Introduction to Industry 4.

0 and Industrial Internet of Things

Industry 4.0: The Fourth Revolution

In this lecture, and in the subsequent modules, we are going to go through an understanding
of what is this industry 4.0 and what are the different changes in the revolution that is
happening in order to transform the industries towards Industry 4.0. So, Industry 4.0
basically corresponds to the fourth revolution in the industries.

Revolution, as you know, that English meaning of this term is basically some kind of
shift, some abrupt change that is required in order to transform the way things have been
happening. An abrupt change in the way things are being done, so that is the revolution.
If you look at revolutions in the past earlier, if we go back more than 10,000 years ago, our
predecessors used to collect food through foraging techniques. So, basically wandering
around collecting food, bringing them, eating the food through the collected food materials
like fruits, vegetables, etc whatever they used to find.

Further, foraging behavior was transformed to farming. Different crops were grown,
different vegetable plants, different fruits, food plantations were started. The result of
this transformation from foraging to farming was increased production, and increased

communication between different humans. There was growth of food production as the
population growth increased. We are talking about this more than 10,000 years back.

Then came the industrial revolution where new technologies, new machines were
produced, new approaches to the production processes were introduced. This shifted the
economy from the primitive economy with simple agrarian-centric economies to more
aggressive machine-oriented production systems. So, that was the industrial revolution.

Consequently, the economic models changed, the social architecture, everything changed
with the revolution in the industries, with the introduction of new technologies, and new
approaches.

The industrial revolution again went through different stages. Back in 1760s to 1840s, it
was the first industrial revolution. And this first industrial revolution was started with the
invention of steam engine, introduction of trains, mobility increased, construction of
railways stimulated the overall revolution. So, this resulted in the utilization of machines
in production, during the first industrial revolution in the 1760s to 1840s.

Then came the second industrial revolution, which was during the transition from the
19th century to the 20th century, with the advent of electricity and increase of electricity
in the society. This resulted in mass production, machineries could use electricity for
large scale and faster production.

Then came the third industrial revolution that was in the 1960s and so on. It was around
that time that computers were starting to get popular. So, gradually with the increase of
computers, different computers, and computing devices, peripherals, etc, the
transformation in the industries also happened. There was increased use of digital
technologies in the industry. The use of computers in digitization and so on was another
revolution--the third industrial revolution, which again increased the production of goods
and commodities in the industry.

The production was due to the increase in semiconductors and semiconducting devices,
which was almost in parallel with the growth of computers. The result was increase in
computing technologies such as mainframe computers, and personal computers. And
eventually the connectivity between these different computers, internet, world wide web
gradually started, and these basically were introduced in the industries to improve the
efficiency of the machinery, processes in the industries, manufacturing processes and other
processes in the industries. So, that was the third industrial revolution, which was basically
the introduction of computers and information technology in the different machinery and
manufacturing processes in the industry.

The fourth industrial revolution or this industry 4.0 is something, which has the origin in
the German economy. So, it was required in the early 21st century to uplift the German
economy. The individual industries having individual IT infrastructure everything was
through the third industrial revolution. But how could we improve the production even
faster and make the processes even more efficient.

So, people thought about how things could be done. There were different sensors and
sensing technologies, which were becoming very popular. With the introduction of sensors,
actuators, along with the regular infrastructure, the IT infrastructure, the internet together
basically was able to transform the existing IT-based infrastructure in the companies to
more efficient ones to connected, sensed machinery and so on, so that was the fourth
industrial revolution. And this is fourth industrial revolution or the Industry
4.0 that we are going through at this moment.

So, this industry 4.0 or fourth industrial revolution started. The sensors and actuators are
very small in size, cheap, and powerful. So, small sized, sensors, actuators, much more
powerful internet and IT infrastructure together, and autonomous monitoring, use of
technologies such as artificial intelligence, machine learning, cyber physical systems, use
of all of these together is basically how this transformation is happening in Industry 4.0.

Computers have become more sophisticated. They are smaller in size, but much more
powerful, less energy consuming. These computers can be integrated and connected
together. This has resulted in the radical transformation of the global economies, the
societies, and the different industries. So, the fourth industrial revolution, this term was
coined as the second machine age by Prof. Erik from MIT and Andrew McAfee from the
same institute.

So, Industry 4.0, which is the synonym of industrial fourth industrial revolution; the term
Industry 4.0 was coined in the Hannover fair in Germany in 2011. So, only a few years
back, in 2011, the whole thing started. Now in the industries, through this kind of
transformation, with the help of sensors, actuators, computers, IT infrastructure, much
more efficient one, smaller in size, cheaper, and everything connected together, this
fourth industrial revolution started.

So, what is the scope of the fourth industrial revolution? Smart connected machines.
Internet work, communication, connecting different standalone machinery, typically which
used to be all IT driven, computer driven; and smart means we are using different for
smartness, we need autonomous behavior things, which can be detected, corrected and
taken forward in an autonomous manner that is basically the smartness. The introduction
of small, cheap, energy-efficient sensors, and actuators have made it possible to make
machines and connected machines smarter.

The smart factories are basically similar kind of concept extended beyond simple
machinery, but having the entire factory operations, machinery in the factory, all of
which made smarter with the introduction of connected sensors, connected machines and
so on in the industries and so on.

Then came technologies such as nanotechnology, renewable energy, quantum

computing, biotechnological interventions innovations like gene sequencing and so on.
Everything together has helped in the overall growth. IT, sensing technology, sensor
networks; then we have the biotechnology - gene sequencing, DNA sequencing, and so on;
nanotechnology and quantum computing, everything packaged together is helping in the
transformation to the fourth industrial age which is Industry 4.0.
There has been some profound and systematic change in the fourth industrial revolution.
This change has been in innovation disruption; disruptive technologies have been
introduced. And this disruption and innovation has happened in both the scale and scope.
The scope has increased manifolds, and has increased.

The scale and scope of innovation of fourth industrial revolution has basically defined
today’s acute disruption and innovation in technologies and the transformation of industries
accordingly. Companies like Alibaba, companies like Uber, Airbnb, Amazon and Flipkart
in India, so all of these companies are essentially transforming the way the operations,
delivery, have been carried on in the past. Everything has changed the way things are
happening at present.

These are using different state of the art technologies such as cloud, sensor networks,
technologies such as drones, networked drones and many different other technologies are
being used in order to transform the industries and their operations. So, these are newer
ways in which these companies are working.

For example, iPhone was launched in 2007, but since then only within few years, billions
of smartphones are being mass produced at present. So, you see that how this disruption
and penetration of these technologies are happening and how fast they are happening and
in what scale they are happening right. So, these are all like different phenomena that are

happening in terms of change, systematic and profound change, that are happening in
this fourth industrial age.

Take the company Google we all know how Google is transforming the world at present.
In 2010, as many of us know just still few years back, Google basically announced the
fully autonomous car. And already we have seen that self navigating cars, fully autonomous
cars are already in the roads. But what is very important also is the increase in the use of
technologies such as artificial intelligence and machine learning, that is making these kind
of newer technologies, a reality.
So, not only the speed of change, but also the scale of change, the profound change, both
are very equally important and are increasing in rapid pace. So, like going back to the
1990s, there were different industry giants in the Detroit area, in US, which had a combined
market value of 36 billion dollars, in terms of capitals. Their overall revenues were in the
order of about 250 billion US dollars and the number of employees working in that area
was about 1.2 million.
In 2014, if you look at in the Silicon Valley, the different industry giants together had a
combined market of about 1.09 trillion dollars. So, the capital has increased manifold.
On the other hand, if you look and compare with these industry giants in the Detroit area
in 1990s, in the Silicon Valley area, the number of employees was only 1,37,000. There

is a drastic reduction in the number of manpower that is being used in the industries, 1.2
million employees to 1.37 lakhs employees being used.

On the other hand, capitalization has increased manifold. With the reduced number of
workforce, we are able to increase the number of the value of capitalization and growth
of the industries. Consequently, the revenue that is obtained from the different sales has
also increased.

With marginal costs, the digital business creates unit of today’s wealth with fewer
workers. We are able to increase the wealth, but with the use of fewer workers and that has
been possible with the introduction of digitization.

Business provides information goods, which has virtually zero transportation and
replication cost. Consider companies such as Instagram, WhatsApp, Facebook, Twitter.
They are able to supply information-centric products, and goods. There is almost like
zero transportation cost involved and that is unlike the manufacturing industries which
have lot of capital goods, lot of transportation costs are involved, shipping from one
location to another, and logistics. The transformation to the introduction of digitalization
and advanced IT and different other technologies is making the scaling up a possibility in
this Industry 4.0 revolution.
In this context of Industry 4.0, digital fabrication technologies are able to communicate
with biological world. Now, it is possible that through the advancement in technology, it
is possible that we can have different sensors inside the physiological systems.

These sensors and different other advanced technologies are able to communicate with
the outside world that means outside the human beings. This is now possible that
remotely signals are send to the human body to perform certain physiological operations
within a human, without basically having the human go through or rather the human
being or the patient being able to know what is actually happening. These things are
possible with the introduction of all these new technologies. Consequently,
transformation has been possible in this fourth industrial revolution, transformation of
the biological systems, and physiological systems.
Now, it is also possible to manipulate the way biomolecules within a human body
operate. It is possible to sequence the different genes in the body. It is also possible to
manipulate the DNA within a body. These things are possible due to profound change with
the introduction of all these different technologies, IT, biotechnology, nanotechnology, and
quantum technology.

So, designers and architects are, now, combining, computational design, additive
manufacturing, material engineering, synthetic biology and so on. Consequently, they are
able to increase the number of production of the number of objects. These different

technologies will be used in order to increase the production of number of objects and these
objects are also being produced, because of the use of these different types of technologies.
It is now easy to mutate these products, and have these products adaptable to different
changes in the environments of their operation.

In the context of the fourth industrial revolution, use of AI or artificial intelligence, machine
learning, also has made it possible to have self driving cars, virtual assessment, transitional
software, discovery of new drugs, and prediction of cultural interest, have been made
possible with the use of artificial intelligence.

Everybody knows about the application Siri. Siri basically is nothing, but an application of
voice search, it is an Apple product, which basically uses artificial intelligence techniques
in order to have and in order to perform voice search. Somebody speaks and this application
will automatically recognize the voice and make searching in the internet possible. And it
is very similar to the Cortana by Windows. This is basically the voice search with the
introduction of use of strength of AI machine learning.

So, there are different drivers. These drivers are basically driving this change. Change in
terms of scientific breakthroughs. Introduction of newer technologies, current
transformation in terms of changes in the megatrends are happening, future transformation,
the tippling points, everything are happening at present with the introduction of all of these
technologies.

These megatrends are due to the introduction of recent technologies and using or leveraging
the pervasive potential of digitization and information technologies. In termsof physical
transformation of machinery, manufacturing machinery, digital transformation, in terms of
the introduction of IT, biological transformation through the introduction of biotechnology,
biotechnological systems, all of these are newer megatrends are happening at present.
In terms of the physical megatrends, we have now autonomous vehicles, 3D printing,
advanced robotics, connected robotics, new materials, lightweight materials, cheaper
materials, and stronger materials.

Autonomous vehicles are already in place. We now have autonomous trucks,

autonomous drones. Drones are basically autonomous self-driven, airborne vehicles, where
there is typically no pilot or any kind any kind of human pilot or machine pilot. Aircrafts,
driver-less aircrafts are a reality now. And also driver-less boats and many other like
particularly in agriculture people are now talking about use of driver-less, driver-less
tractors right. So, all these are possible with the introduction and use of artificial
intelligence and advancement in robotics.

3D printers, basically, give any shape these machines 3D printers. They will manufacture
a particular product according to the specified shape. 3D printers have applications in wind
turbines and medical implants.

Advanced robotics, different robotics, robotic equipments, connected robotic equipments

are being used in the automotive industries. Robots are also used in medical domain for
robotic surgery. Robotic surgery is basically something that is happening worldwide.
Robots and their use in agriculture robots and their use in nursing. Connected robotics is
something that is quite common now in automotive industries in agricultural fields. We
ourselves have developed some robots, which will go and plant some seeds in the
agricultural field. So, these are all these advancements that are happening in different
industrial fronts with the advancement of robotics and connected robots.

Newer materials, lighter materials, stronger materials, materials that are recyclable,
recyclable plastics are basically a reality now. Use of nanotechnology, Graphene, carbon,
nanotubes these are also making these materials lighter, stronger, having different
advanced properties for being used in different application domains such as aviation
industries, for different other manufacturing industries.

This digital transformation has been happening. The introduction of sensors, connected
sensors, connected actuators. These are all about the use of internet of things. Sensors being
used RFIDs, NFCs, tracking of package delivery, typically by courier companies this is
quite common. Complex supply chain, monitoring systems are all a reality in this fourth
industrial revolution.
We have all heard about use of digital currencies. Bitcoin for example is one such, and
use of Block chain, for securing bank transactions and government transactions and so
on.

Companies such as Uber are transforming the models of transportation. Car pooling has
increased the revenues that are earned by these companies. So, the reduction in fuel
consumption overall and also consequently reducing the pollution in the environment
through the use of car pooling, reduction in fuel consumption, increase in revenue,
making everything very smart in the sense that have all of these different cars and
vehicles connected to each other. Instantly, which car is where, and what is the
availability status, and protecting the environment from unnecessary pollution all of
these things are now possible.

Biological transformations adoption of different technologies such as gene sequencing,

DNA writing, recommender systems, so recommender system in the physiological
domain is about like if you tell some of the molecular level, composition of a human
being then what is the best precise drug that should be administered to that particular human
that recommendation can come in through the use of different advanced systems that have
been produced now. Cell modification is possible now, and also advanced genetic
engineering is a reality.

Tipping points or the changes that are coming in the future. So, basically by 2025, it is
expected that we will have the different clothing, the different fabrics, connected to the
internet. There would be unlimited and free storage available to everyone. There would
be trillions of sensors connected to the internet. And we are going to have a world, where
there would be some robotic pharmacists, which is going to help in the pharmaceutical
industry.

With this we come to an end of the fourth industrial revolution lecture. We have
understood, what is this fourth industrial revolution, what were what were the previous
three industrial revolutions that we have seen. And in this fourth industrial revolution
age, what are the new things that have come in technologies such as self driving cars,
technologies such as AI enabled Siri.

Technologies such as connected robots, standalone robots, smart robots being used in
different application domains such as agriculture, medical industries, manufacturing
industries. Technologies that are going to come in the future and people worldwide are
lot of research words that are going on in different labs in the academia and in the industries
R & D industries. All of these things we have gone through. And we have also seen how
these transformations are happening gradually and what is the benefit of all these
transformations that are happening.

So, with this we come to an end, these are some of these references that you might want
to go through if you are interested to know further about any of these. There are many more
references that would be there if you search in the internet or if you look at relevant books
like the ones that are listed over here. With this we come to an end.
 Introduction to Industry 4.0 and Industrial Internet of Things
Industry 4.0: Sustainability Assessment of Manufacturing Industry

In this lecture, we are going to go through some of the very important concepts for
sustaining the Industry 4.0 revolution. So, basically what is required is not just the
development, development in terms of introduction of new technologies, new types of
services, and so on, but what is also required is to be able to sustain the newer systems,
newer methodologies, newer techniques that are introduced. So, what is required
essentially is to ensure that the sustainability of the newly introduced concepts,
technologies, methods.

The term sustainability means to continue at a fixed rate. A sustainable industry provides
energy efficiency, conservation of resources, and low waste production. So, the last one
particularly is very important. Energy efficiency is very important because there is a global
concern about saving energy and thereby saving the environment. So, energy efficiency is
definitely very important to reduce carbon footprint on the environment. But what is also
very important is to ensure that the amount of resources that are used are all conserved and
the amount of waste that is produced should be reduced significantly.

So, that will be a sustainable method of manufacturing, sustainable method of production

or development. So, let us consider the context of sustainable manufacturing industries.

In the context of Industry 4.0, the proposition is to include the characteristics of previous
industry revolution in a much more sustainable way. So, not only that we want to introduce
newer things, we have to ensure that whatever has been existing from the previous
industrial revolutions continue in the same form that it was before. To introduce newer
techniques methods and continue the same things in a much more consistent, sustainable
manner, so that it does not become a one-time kind affair.

Industry 4.0 or the fourth industrial revolution is a comprehensive industrial revolution,

which takes into account this sustainability issue because it introduces newer concepts,
newer technologies, but also to ensure that there is sustainability in the long run.
Industry 4.0 incorporates globalization and emerging issues as well.
So, when we talk about sustainability first, what is required is to consider some industry
in our case, the manufacturing industry and assess the sustainability issues. First of all
you have to assess, how much sustainable that is in particular industry, in terms of the
processes, that are existing or the processes or the product, that is being manufactured.

A manufacturing industry is considered as a base of modern industrial society and is the

cornerstone of the world economy. In order to estimate this sustainability and assess it, it
is required to evaluate this term S over SD.

This S over SD considers different issues, different parameters, which are linked to the
issue of globalization, are considered as one of the important drivers of sustainable
industries, globalization issues affect the sustainability of any development or any
manufacturing or any production system.

There are five different elements of globalization that are depicted. So, the first one is the
business models, the next one is the energy price, information and communication
technology adoption, supply chain management, which is a very important thing,
particularly, in the manufacturing context, and the inclusion of emerging markets. So, these
are some of the important issues or the cornerstones behind globalization.

We will start with the supply chain management. So, supply chain management talks about
consideration of the different stages through which the production system starting from the
production till the supply goes through. In these manufacturing industries, there are
different suppliers, the production system as a whole and different customer. Each of these
together go through different stages. The sequencing of the stages for the whole system is
what SCM basically talks about, supply chain management.

The most important stage in SCM is the selection for outsourcing components or parts of
raw material. So, supply chain management has many additional environmental concerns
as well. So, these are the issues that basically contribute to the overall sustainability. Issues
of climate change, contamination – contamination of through the introduction of
different wastes to the environment, contamination resource consumption, how much
resources are consumed, different human resources, non-human resources. It is required
basically to optimize the resource consumption. These are the different supply chain
management issues.

The second thing is the introduction of ICT -- Information and Communication

Technology. As we know that ICT overall is the backbone of most of the modern
manufacturing industries today. So, if there is no information technology or
communication technology there is no communication within the enterprise and across
enterprises. This communication technology is required even to distribute we have
distributed communication between different people, distributed communication between
different people, distributed communication between the different labs of the same
organization, across the different locations, or different campuses of the same
organization or even it is also required for having proper communication between the
different partners of a particular organization, who contribute to the manufacturing of the
products or the services.

So, overall ICT is very important and we are talking about ICT, as a whole, in the bigger
context, not just the introduction of computers and communication technologies, but also
different other things that we talk about in the industry 4.0 context like sensors, actuators

and the connectivity between them and the different other things. So, it is very required
to have communication between the customers, the producers, and the suppliers.

They need to be able to share the information between themselves and with the help of
these computers, computing technology, and communication technology. It is required to
enable Enterprise Resource Planning, ERP-based systems, then wireless communication
technology, which is sort of like why wireless is; wireless is something that makes
portability a reality, portability of different equipment, then mobility across different
parts of the company, that are being used in the manufacturing process.

Wireless communication technology is very important thing, a third one is the GPS and
the fourth one is the radio frequency identification system. With the help of RFIDs we have
gone through in the introduction. You need to tag the different parts in a production system
or different equipments that need to be tracked or different items or elements or agents that
need to be tracked and that basically helps to have a complete and efficient monitoring, of
the mobility and portability of these different parts and constituents of the whole production
system.
The second, the third one is the energy prices. If you are talking about large-scale
enterprises there is a consideration of the energy consumption as a whole. So, larger
enterprise means larger energy consumption, but that is sustainable and beneficial for the
environment more energy consumption basically leads to bigger impact on the

environment and which is not very desirable. So, for enterprises it is very important to
ensure that there is reduced energy consumption, through the introduction of these newer
technologies.

Energy has to be first created, then transferred, transformed into a different form, and
then gets consumed. For instance, if you are talking about electricity; electricity gets
generated in the generating power plant, then it is transferred through different grids and
transmission lines it is transferred from one location from the generation station to
elsewhere to this station to the electricity substations and different other points. So, one
form of energy is transformed to another form of energy.

In our example, electricity is transformed, let us say that if we are talking about lighting
lamps and bulbs in the warehouses, then electricity is getting transformed into light energy.

So, what is required is whatever be the cycle, however, the energy consumption and this
transformation takes place, whatever be it what is important is to ensure that there is
reduced energy consumption overall and that basically we will also have an impact on
the economy and the environment as I said earlier.

So, it is required to have energy supply also at reasonable price. The increase in the price
of energy is not good, because if you are increasing the price of energy then that will affect
the overall price of the product or the service that is being created and that is not going to
be sustainable over all from another perspective.

Increasing energy price effects sustainability and we have seen that different ways it is
affecting, it is not sustainability has different facets. So, what is also not desirable is to
increase the cost of the product or the service that is being generated or being developed
and also it is also the cost of energy, if it increases, it is not very useful from the
environment point of view as well.
The reduction in energy consumption is also required and energy consumption, energy
production can be done not only from the non-renewable sources of energy, but also
from the renewable ones like solar, wind.

The next one the next globalization issue is the emerging markets and its consideration. So,
markets are able to meet the standards of newly developed innovative products. If you
think about the emerging markets, whenever a particular product is being introduced.
So, it goes through typically a phase of dictatorship; dictatorship means like it is a
monopoly kind of thing. The company which introduced the product basically has
monopoly. And, then gradually it has to transform towards the free market and free
economy where it will be made accessible to the greater part of the world. So, the
consideration of emerging markets is very important, particularly for developing countries.

Business models: we need to think about business models which will be helpful for the
greater society. So, it is required to have mass customization, which will incorporate the
knowledge including the consideration of international culture across different countries,
different societies, and so on and also the local culture where things have been
introduced and produced first. So, business models basically have direct linkage with the
mass customization. So, what is required is from the globalization point of view the product
that is manufactured should not only cater to the needs of the local community, but also to
the international community.

So, business models are to be developed which should take the strategic approach by
considering the bigger issues the strategic issues; that means, high level issues for a
particular organization business and the greater issues overall. It is also required to have
the business model, which will maximize the economic profits for an enterprise by
taking into account the competitive benefits and promoting the product value.

Emerging issues are also there like the globalization issues which also contribute to the
sustainability factor. There are many of these emerging issues which contribute to such
sustainability. One is technology, growth of population, government regulation,
consumption of natural resources, and consideration of crisis, recession, and depression.
These are the five different contributors to the to the sustainability factor from an emerging
issue viewpoint.
So, the first one is the technology. Technology considerations are very important. So, if we
are talking about technology, let us say, technology broadly can be classified as hardware
and software. Within hardware we have different types of technologies that are used
commonly technologies such as computer. When we are talking about technology, we have
the hardware technology and the software technologies.

Within hardware we have different technologies in the manufacturing industries such as

the computer numeric control machines, we have the barcodes, barcode base technology
barcode, we have different technologies such as RFID, NFC, all of these are used quite
widely and these are the technologies that have also contributed to the overall advancement
of the manufacturing industries.

In terms of software, we have technologies such as ERP, the GPS-based software

technologies. GPS itself is hardware, but the GPS-based software technologies and like
this there are many other software technologies that have emerged and contributed to the
advancement of manufacturing industries.

These are very important and what we need to ensure is that all these different newer
technologies should be included in order to have better sustainability. Advancement in
technology facilitates manufacturing with higher quality products, lower cost products
and products which are manufactured in reduced time. So, these are very important. So, the
quality of the product that is developed should be improved with the introduction of

these newer technologies, then the cost of the production should be less, so that the
overall product comes to the market at a reduced price and comes faster to the market,
the production lifecycle should be reduced. So, there is reduced manufacturing time.

The role of technology advancement in the global market is about converting from the
traditional system of manufacturing to the automated system. Technology can help in
having greater agility, faster development, and flexible, changeable, maintainable products.

In terms of government regulation, it is also required to be able to protect the public and
the private sectors. It is required to protect the enterprises. So, for protecting the enterprises,
different enterprises have their own different requirements, and those will have to be taken
into consideration while arriving at different regulations and rules, which can help these
organizations these enterprises to offer better services and low cost goods.
So, government regulation will help in basically avoiding unfair competition and also to
promote sustainable environment considered development for everyone including the
employees of the organization or the industry. Therefore, government regulations are
very important.
We are talking about government regulations, there are different considerations or the
different types of regulations. Some of these regulations are basically having direct
impact on the economic issues, some of them have considerations of the social issues,
and some the environmental issues.

Economic issues talk about some of these regulations, when a business will enter, when
an institution basically enters a business. Price estimations, there are government
regulations typically, which will help in this price estimation. Then social issues, which
will essentially help in opening channels between channels of communication, maybe
between employees and the enterprise or the management and environmental issues will
concern the protection of the environment.

Environmental issues are very important in the manufacturing process. Government

regulations should be there to protect the environment, because if there is some kind of
production process, which produces lot of waste and in turn harms the environment the
water, the land, are full of different industrial wastes. There are government regulations,
which talk about how to reduce these wastes and also the wastes that are produced will
have to be handled properly.

There are government regulations concerning employment, advertisement, labor – labor

laws are there, environmental regulations are there, regulations concerning the safety of
the workers, safety of everyone, health regulations are there, and privacy protection of
individuals.

Basically, the employment and labor rules represent laws concerning wages and salaries,
things such as benefits to the workers in terms of retirement plans, compliance with
health and safety issues, proper working conditions, issues of expatriate employees such

as visas and so on. Equal opportunity in employment in terms of promotion and

consideration of all workers from different ethnicity in the equal platform, provisioning
of authority or higher ranking position; these are different types of classes of regulations
that are typically there in any industry.

In terms of advertisement – advertisement regulations protect customers, firm honesty

about a product, information regulation publicly, then transparency on distribution and
manufacturing process.
Environmental regulations or rules meant, are maintained by different acts, different
agencies such as the environmental protection agency. Maintaining clean air, reduction
of chemical effects in soil, river water of different water bodies, then privacy regulations
concerned with the information of safety and security, particularly the sensitive
information, that is collected about the different employees and the other stakeholders by
a particular enterprise.

fety and health regulations will concern the health issues providing healthy and
working environment, overall safety, and workplace safety. These are the different types
of government regulations.

Population growth: monitoring population growth is important for manufacturing industry.

It affects the industry growth, food supplies, fertility, sociology, economics politics,
industry locations, and use of available lands.

There are three different types of countries based on the population growth, developed
countries then emerging countries, and developing economies.

So, the population growth of countries developing and disadvantage is typically greater
than the population growth of the developed and advantaged countries.

Based on the United Nations report the population growth is from 1950 to 2050, reduced
between 32 percent to 13 percent in developed countries, increased between 8 to 20 percent
in emerging and developing countries.

So, economic view on the population growth can be of two types: pessimistic view and
optimistic view.

The pessimistic view of population growth basically effects the economic growth and the
optimistic view basically on the contrary talks about increase of the globalization issues

such as trade and commerce due to the growth of population. So, pessimistic view basically
effects, hinders the overall economic growth, on the other hand, and the optimistic view
on the other hand, basically, increases the globalization issues such as trade and commerce.
Population growth is a very important consideration in terms of sustainability. So,
basically, if the growth is not conformant with the overall growth of the economy then that
will become a disaster for that particular economy.

Then the last one is basically the consideration of economic crisis recession and depression.
So, economic crisis basically takes place over a duration not more than a few months, so
that is the crisis. Recession on the other hand, talks about the decline in the economic
activity recession; that means slowdown of the economy, exponential decline happens.

When the economic activities again increase, then there will be commencement or the
rise of the economy. So, basically from the recession, then there will be some kind of
acceleration in terms of the growth.

Depression is the extremity of recession, which is observed by exponential

unemployment increase, reduction in available credit, significant reduction in trade and
commerce and huge number of bankruptcies might also consequently happen and there is
volatility in currency value and the duration is more than two years and this is basically the
extreme case of recession.

So, we start with the economic crisis then that is that will take place for few months, then
recession, a rapid slowdown of the economy and thereafter we have the depression. This
depression basically is the extreme case of recession, where all of these things would
[Link], let us now look at the view of each of these. So, basically if we talk about
crisis so, let us say that this is the economic crisis. Then comes the recession phase,
recession, and the next one is basically the depression. So, conceptually it would look like
this.

If there is economic crisis on recession, then there will be reduction in prices of different
major commodities and what is important is to increase the productivity and reduce the
overall cost that becomes the solution in such a case. So, it is very important to ensure
and avoid this kind of crisis and recession as much as possible.

In terms of consumption of natural resources this is one of the very biggest concerns,
because that has impact on the sustainable development which is environment friendly. So,
this is one of the biggest issues, in contrast, to economically sustainable development as
natural resources are the main source of revenue in developing countries. It is one of the
major sources of social conflicts. So, basically concerns about mining, extraction of oil
and gas, demographic shifts, societal behavior, politics, technology, economic situations,
difficult economic situations all of these are major contributors in terms of the
consideration of natural resources.

So, natural resources can be of two types, the renewable ones and the non-renewable
ones. Coal, oil, gas, are the non-renewable examples of non-renewable sources of natural
in energy and then solar, water, wind, are the examples of renewable sources of energy.

So, all of these things we have discussed because if you recall that we started with that
formula S over SD.

So, for computing this S over SD, which is basically the sustainability factor we have gone
through all of these different issues. Issues of technology, government regulation, growth
of population, crisis, recession, depression, and consumption of natural resources, and they
are corresponding things also we have understood that how these, what are these
different issues and what are what is important for consideration of these different issues.

So, while we have understood all of these things we now need to basically look at this
particular S over SD formula.

S over SD is basically sustainability over sustainable development and for this S over SD
has all of it is a function of all these parameters, right. So, all these parameters that we have
seen in the previous slide and so, this basically can be again rewritten as a function of all
these is and where I equal to S over E and S is basically the change towards the
sustainability and Y is basically the exponent of the change towards sustainability S of E.

So, this is basically the overall formula that we wanted to arrive at to start with we have
looked at this S over SD formula and these are these different contributors to the
different parameters that contribute to this computation of S over SD and this factor
basically talks about whether a particular effort is sustainable or not

So, finally, these are some of these references that you could go through in order to
understand these concepts, in greater detail.
 On Introduction to Industry 4.0 And Industrial Internet Of Things
Industry 4.0: Lean Production System

We need to understand the sustainability issue that we have discussed in the previous
lecture. There are different ways in the development of the product, the manufacturing of
a product, it can be done in different ways. One of the considerations for sustainable
development of a product is to have the adoption of lean production system. So, lean and
thin production process should be adopted.

Therefore, we need to have a cheaper product of higher quality developed in reduced

time. And in the consideration of lean, and ensuring that there is reduced waste disposal
to the environment. Reduction in wastes is the prime consideration in the lean
manufacturing process or lean production process.

The concept of lean basically has its origin in Japan. The major global automotive
manufacturer, Toyota, adopted their own production of their own vehicles, a
manufacturing process, which was known as the Toyota Production System, TPS. So, it
was done only for within their company for basically having the advantages at the outset.

This became popular globally, for adoption across different industries concerned about
manufacturing of different products. It become very popular to ensure that at the end you
have a product of high quality, produced in reduced time, and is produced with minimal
wastes. Wastes of all kinds not just the tangible wastes, but wastage of time, wastage of
human resources. So, improving profits having higher quality products, produced in
reduced time with reduced waste of resources is what is the overall goal of sustainability.
And lean production contributes to this overall issue of sustainability.

So, what is this lean production system? It is all about eliminating wastes from the
processes, manufacturing processes in manufacturing industries for instance having
higher quality products. So, it has its origin in the Toyota motor corporation company. This
TPS or lean production system mainly focuses on addressing the customer’s needs directly,
which is key to this lean production process. Whatever the customer needs, eliminating the
over usage of resources, use whatever resources are precisely required to eliminate all kinds
of wastes, and finally produce a good which the customer basically would be satisfied in
reduce time and having higher quality.
If you look at the comparison of lean approach versus other approaches, lean approach is
basically look from the customers’ perspective, whereas other approaches basically look
from the task and production perspective. So, this fundamental difference between lean
and the non-lean approaches.

This lean production system has basically considerations of two different types of systems.
One is the JIDOKA, which talks about that whenever there is a problem, problem
of any kind with the machines, or in the process, or whatever be it. If there is a problem,
stop the production, then and there, stop producing defective products in turn. So, this is
this JIDOKA process.

And just-in-time basically talks about ensuring that each process produces whatever is
exactly needed by the next process, in a continuous flow. JIT or just-in-time, is the short
form, it is also known as popularly known as JIT. So, where each process produces what
is needed by the next process in a continuous flow. There are different other components
of this lean production process, such as the PPS, Toyota production system.

There are different types of wastes. And these wastes have to be minimized. So, these
wastes are transportation wastes, which talks about excessive movement of people from
materials or information. Transportation of different materials, transportation of
information, in excess, that is not required. Second is the waiting, which is basically talking
about the period of inactivity or people for material or information to arrive or to be able
to use. It should not happen that the workers are waiting for something to arrive and the
time gets wasted. So, this waiting has to be minimized.

Third form of waste is the motion, which is non value-added movement of people.
Unnecessary movement of people from one point to another within the factory, outside
the factory, across different locations of the same factory or organization, or between
different organizations as the case may be. But whatever is required you should have
only that kind of movement, non value-added movement should be reduced.

Inventory waste, which is the cost of inventory such as raw materials, work in progress,
and finished goods. So, all of these things whatever inventory would be required should be
used, and not unnecessary inventories should be used built up and procured.
Over-processing - doing more work in the product, then whatever the customer finds value
in. So, customer values, talks about what the customer needs precisely. And the production
system should talk about only addressing the customers’ needs. Producing exact product
or the service that the customer needs, this is what has to be done. And over-processing
doing more work for the product you know having different features, which customer will
not need.

Defects can be in the product, in the service, in the process, or in the paper works.
Defects of all kinds should be reduced. Overproduction - producing more product sooner
than the customers’ requirement, and that should be also reduced. So, these are the seven
different forms of wastes.

The overall lean system talks about this value streams. So, value streams are all the actions
that are required for a product from order by the customer to the delivery of the product to
the customer. And it is not just the product, it could be the service as well. So, starting from
the ordering of the product or the service to the delivery of the product, the service is what
the value stream basically talks about.

There are three types of works. One is value added; second is incidental, and pure waste.
These are the different, which is very much desirable value added to the value stream, then
pure waste is not desirable.

There are five steps of walk in the value streams. Focusing on a single value stream is
very much desirable in building a leadership team. Scheduling the date and time properly
for each, and everything whatever the customer needs. Walking it, that means, discuss
the value, value in terms of the customer requirements, precise customer requirements,
discussing the value of those walking together with all stakeholders, with the customer,
listing, prioritizing ideas, and doing everything together what is desirable in terms of
offering the value to the customer. Schedule follow up is the last one. These are the five
steps of work in the value streams.

In the context of Industry 4.0, lean production is paramount; it is something very important.
This lean production talks about four different considerations. Concerns about the
integration of humans in the plant, continuous input improvement, value added
activities, identifying waste in the processes, and eliminating them, as soon as possible,
in an environment friendly manner.

The impact of lean production system in terms of offering what is going to happen if you
adopt this lean production system. You are going to have products of best quality, produced
in lower cost, and with shortest time of production. These are the three different
impacts of production using lean.

Implementation of lean leads to the implementation of the full manufacturing system. It

does not focus on just the adoption of the lean tools, but also it focuses on different other
areas such as business requirements, operation improvement, operation of everything,
operation of the machinery, operation of the product being developed, operation of the
processes. So, overall production process, production operations o the overall
production process, people management, and performance [Link], implementation
of lean has different requirements. First is the business requirements, which talks about
setting the right objectives. Clearing the strategy completely from a business point of view;
business strategy should be clarified.

Second, the performance management dimension in terms of lean implementation. Here,

we are talking about people management, use of different KPIs to assess. KPIs means
key performance indicators, so use of these different KPIs to assess the performance in
the production process. Consideration of top-down management and we are talking about
the use of KPIs to measure the performance in terms of productivity, quality of the goods
and products, cost of delivery, cost of production, delivery time, and safety. These are
the different performance attributes or the KPIs that have to be considered from a
performance management viewpoint, in the lean production system.

Operation improvement should have clear knowledge about all tools of the toolbox, and
should not have massive toolbox with unnecessary tools. Then minimize the wastage in
terms of all different resources, and overall improve the operations.
The other dimension of a lean production is basically people engagement. And as this name
says it basically to ensure that people, the constant stakeholders are all engaged, and
follow these different objectives learn, do, teach. Learn means having clear knowledge
about the tools; do means perform, all the tools you know act with all thetools that are
available, and teach into the role of a teacher about these different tools. So, engaging all
these peoples in these three different directions is also very important.

This is basically we have been talking from the starting. So, basically to reduce the cost,
the time of production, and to improve the quality, would be the contributors to the
overall continuous improvement, continuous improvement, in the manufacturing process.
So, let us now look at it from another direction, holistically, what lean is all about. If we
are talking about the lean production system, it is required to have these
differentcomponents of the lean production system, lean manufacturing, adopting lean
manufacturing methods, ensuring total quality management, and adoption of it solutions
to improve the overall production, reduce time, improve upon everything, virtual
integration of different machinery, and have a lean supply chain.

If required as much as possible to outsource some parts of the product or the

components, as the case, may be if it is cheaper and can be done faster by some experts,
who have more expertise in developing those components. And just-in-time is the
another one that I have already mentioned before adopting just-in-time concepts, and
having flexible manufacturing. These are the different components of the lean production
process, lean production system.

So, here are the references that you could go through in order to have further understanding
about whatever we have covered with respect to lean production system, lean production
process, and lean manufacturing.
 ntroduction to Industry 4.0 and Industrial Internet of things
Industry 4.0: Smart and Connected Business Perspective

Another very important aspect of Industry 4.0 is the business perspective; the Smart and
Connected Business Perspective. From a business point of view, looking at the smartness
and connectivity that we have been talking about in the context of Industry 4.0; trying to
look deeper into these different issues concerning it.

First of all, the question is that we have been talking about that in industry 4.0; we need
this connectivity, connected machinery, connectivity everywhere, sensing connectivity,
and so on.

So, the question is that why do we need this kind of connected products, what is so good
about it? Why we want to transform ourselves from whatever we were doing in the past
to connected products? Why do we need connectivity between these different products this
machinery? Why at all we need to make these machineries smarter?

First of all, this connectivity is required in order to connect these physical objects, different
machines in the industries. The second thing is that it is very important tocollect
these different data from these physical objects and share between themselves between
themselves and also share the data through some network to elsewhere where it is going
to be analyzed.

So, this connectivity is essentially going to be required, because through this

connectivity and smart perspective; we are going to increase the overall resource
efficiency. Overall efficiency is going to be increased in terms of resource consumption,
resource utilization, and improves the overall productivity in the manufacturing process.

So, improved resource efficiency, improved productivity, are the reasons why we need to
take smart and connected business perspective.

So, what are the benefits once again? Faster we are going to have these products
manufactured at a faster rate than before it was being done in the past without the use of
IoT-centric solutions.

Cheaper products are going to be developed; these products will be used in a better way
through the incorporation of these smart solutions. Improved recall process of the
product, it happens that certain product will have to be recalled right by the industry. So,
that recall process is a very complex process and hard to track. This recall process can be
improved with the incorporation of smart and smartness and connectivity.
Decreasedenvironmental impact is the other one and finally, improved supply chain; smart
supply chain, these are the different benefits of having smart and connected products.

The medium of getting smart and connected are the use of sensors, the core of making
smarter systems. Sensors are associated with the actuators, all these sensors and actuators
will make these systems smarter. In IoT the network is important; cloud is also
important, because lot of data are coming in, who is going to process the data; huge amount
of data to handle.

So, cloud is going to handle so much of data in terms of storage, in terms of processing and
so on. And this is the most important of all embedded systems. By the help of these
embedded systems, fitting all of these different objects, the physical objects with
different embedded devices, which again will have sensors communication device; so all
of these are going to be embedded into each of these physical objects.

These are the fundamental building blocks for smartness and connectivity. Number one
is the customer values, blueprint of profits; key resources and key processes. Customer
values are about the value proposition. So, value proposition means like through the
incorporation of all of these things that we are talking about; the smartness, IoT sensors,
actuators whatever how the product is going to be improved; in terms of usage. From a
customer viewpoint, how the product has improved, what is the value proposition and the
services that the product offers; how these services are going to be improved, what is the
overall value proposition.

Next thing is the blueprint of the profits. The smartness and connectivity is going to help
in coming up with a blueprint for improving the profit. The efficiency of the key
resources in the companies, industrial processes, manufacturing process will have to be
improved, they will have to be tracked and optimized over time. And same thing goes for
these different key processes as well.

The next thing is that; what are the different categories of these smart and connected
products; monitor, control, optimization, and automation; these are these different
categories of smart and connected products. So, monitoring, control, optimization, and
automation these are the ones, that are self-understood.

Monitoring the resources; resources such as sensors, sensor-equipped resources such as

this machinery and mnitoring the external data sources, these are important;
somonitoring of all kinds of different resources and also the monitoring of the effects
through the use of these different resources.

The effects such as the health monitoring of this machinery, health monitoring of the
products; if something has gone down, if something is likely to increase the downtime in
the future generating alerts, beforehand, is also something that will help in improving the
overall efficiency. So, this is another effect and taking action against the odds; this is also
the third effect in terms of monitoring.

Control of these resources through the use of different custom designed, software or
different other software used for the resource control optimization. And optimization of
these resources through the use of these different software, this software can help in the
optimization process and the effects; controlling the products and the personalization of
the products based on the different requirements of the end-users.

We are discussing about the use of different optimization solutions and different
algorithms. The effect is to increase or enhance the overall performance of the system,
machinery, enabling remote services, and assisting in repairing the product.

Automation of these resources, monitoring, control, and optimization capabilities, different

software algorithms will have to be used for the automation, and the effect is having
autonomous performance of these different products.

Smart systems from a business perspective, is very important. So, what is this business
model? What is the smart business model? Why at all we need this smart business
model? So, we need the smart business model in order to make the current processes less
costly, more efficient, and to improve upon the receiving of increased revenue. So, you
need to achieve the expected revenue plus, to have some surplus that is also required often.
So, you need to have a smart business model, that is, that will help in reducing the overall
process cost, making the processes more efficient and meeting the expected revenue and
in fact, exceeding the expected revenue.

So, these are some of the key attributes of the smart business model value proposition,
which I have talked a lot earlier. So, what value it is going to have to the customers in terms
of the product and the services it is offering to the customer; value proposition. This is
the first key attribute when you are trying to come up with a smart business model.
Revenue stream means like what are the different sources of the revenues that are coming
in, what are the different sources that from different sources the revenue can come for
helping in the manufacturing process or undertaking a particular project.
So, what are the different sources of the revenues that is the revenue stream and finally,
that revenue is going to be used in order to help in the manufacturing process; incurring
the different costs. So, that is the revenue stream and the different technologies that are
going to be used are the three different key attributes of a smart business model.

If we are talking about the value creation in a smart business model, there are different
perspectives, different aspects, that will need to be understood.

One is the novelty centricity, second is the efficiency centricity, lock-in-centricity, and
the complementarity; these are the four different perspectives that will need to be

So, what is important is to monitor in a smart business model; if you have to come up
with a smart business model. These are the key things that; we have seen we need to
monitor, we need to have a control mechanism, and have some optimization.

We have things like monitoring the health of the machines. So, health of the machines,
taking some actions, offering different alerts; then, for control we have what? Control of
the product; personalization, so based on the custom requirements, this kind of control
and personalization can be done. We have the optimization; here through the use of
different optimization algorithms, we want to improve the performance of the system the
process and so on performance, and then product repair, and also remote service.

This is one thing; so all of these things can be done with the help of what? What is the
backbone? The backbone is use of different sensors, data services, actuators and in fact,
the different applications and the network. One very important thing I have not
mentioned is the algorithms, the different algorithms that can help in achieving all of these.
Algorithms for monitoring, control algorithms, and optimization.

So, we need to come up with something like this; the business model, a smart business
model that can help achieve this overall thing that we have talked about. So, a smart
business model will need to take into consideration all of these different things. Why do
we need a smart business model? Because in order to improve upon the efficiency,
productivity, and essentially will help in achieving autonomous systems for improving
business efficiency.

We were looking into the value aspect the value proposition where we have seen there is
this novelty aspect, efficiency, lock-in, and complementarity, these are the different
value creation aspects in a smart business model.
We need to have a value-centric business model, which the customers find to be useful to
be of value to the customers. This kind of value-centric business model will open up new
markets, new services will give solutions, which are innovative, new, which we are not
there before customers find it useful exciting, so that is the value centricity.

Next thing is the efficiency; efficiency in terms of improving the products, faster
productivity, simpler production processes, transparent production processes, which
willeliminate errors or reduce the number of errors from occurring; overall efficiency
centricity.

Lock-in centric business model prevents the customer from migration. There will be a lock
in time period during which the customers will not migrate from one vendor to another.
This will help in reducing the switching cost and improves building the trust between these
different customers, suppliers or the vendors.

Complementary business model talks about things such as the product and services,
online and offline assets, technologies, activities all these different complementary
things, which are required in order to come up with the smart business model.

So, what are the layers and technologies that will help in creating values? Number one is
the physical layer, second is the connectivity layer, and the third is the digital layer.

This is the overall architecture that can help in this thing and these are the different
components in it; we have the physical layer, digital layer, and connectivity layer.

So, in each of these layers-we are talking about sensors, microcontrollers; digital layer
we are talking about processors, which can help in execution of these different
algorithms the in the processes overall the functionalities, that will have to be
implemented. So, all of these things are taken care of in the digital layer by the
processor.

The storage means the storage of these data. And in terms of connectivity all these different
technologies ZigBee, Bluetooth, 3G, 4G, NFC, RFID, Z-Wave, Wi-Fi; is the basic
introduction component of this course, all of these are the constituents of the connectivity
layer. These are the three layers in the overall architecture.
The physical layer is responsible for collecting and acquiring data with the help of
sensors and are equipped with different microcontrollers and microprocessors.

Connectivity layer talks about the overall connectivity, smart devices, smart services, using
different connectivity technologies such as IP networks, ZigBee, NFC, and Bluetooth.

Digital layers talks about storing the data analyzing the data processing the data and so
on.
So, here are two examples of smart and connected business model; one is the Amazon’s
dash button, where the value proposition is the lock-in value proposition, the revenue
streams are low cost revenue streams. Physical layer is equipped with Wi-Fi enabled
embedded devices, connectivity layer is Wi-Fi enabled and the digital layer is connected
through mobile applications.

Similarly, we have the other product the Semios, which talks about improving the
efficiency overall, the value proposition is efficiency in improving the efficiency. The
revenue streams are through yearly subscriptions 24X7 monitoring and assistance; the
physical layer constitutes of sensors for soil moisture, insect, disease, climate monitoring
and so on. The connectivity layer is cellular and the digital layer consists of different mobile
applications. These are the two examples of smart and connected business models being
implemented, in practice.
So, these are again some of the references.

This lecture covers the concepts that are very important to understand the overall
business perspective of why at all we need to do, whatever we intend to do; that means,
transform towards industry 4.0.

We intend to transform towards industry 4.0, through the incorporation of all these smart
devices and so on. We want to transition towards making everything smart by, but the
question is that why at all we need to make smart? What is the value proposition? From
ausiness perspective; what are the the things that are going to be improved through this
kind of adoption, transformation from the traditional to the smarter ones through a
connected one, and this is what we have discussed in this particular lecture.

So, understanding these things are very essential for the building of the industry 4.0. And
this needs to be also educated to the industry workforce, who needs to get into this
transformation to industry 4.0. Because until this value proposition and the use of all of
these things the transformation is understood by the workforce; they will not be able to
do this transformation, they will not be able to appreciate this transformation, and
because of which the entire exercise might fail.

With this we stop over here and these lectures will these the concepts, that we have covered
in this lecture will help in further understanding of the later lectures, that we are going to
cover in this course.
 Introduction to Industry 4.0 and Industrial Internet of things
Industry 4.0: Smart Factories

Smart Factory is a term that is used commonly in the context of Industry 4.0 revolution.
People are talking about making smart factories in the industries. So, existing factory,
factory operations, processes, products everything being made smarter, but then we need
to understand that what smart factory is all about.

So, there are different definitions of smart factory that is available. Let us look at one
such definition of smart factory.

So, smart factory as per this definition, “The smart factory is a flexible system. It is a
flexible system, that can self optimize the performance across a broader network and self
adapt and learn from the new conditions in real or near real time and autonomously run the
production processes”. There are different aspects that are captured over here through this
very good definition of smart factory, what are these different aspects. It is a flexible
system; we need to have a system, which can be changed easily. If you need to introduce
certain things in an existing process, the system should be flexible enough in order to
incorporate in order to help in incorporating that particular [Link]-optimizing the
performance over time when the system is performing, this smart factory or some
component of it in a smart factory is able to optimize its performance over time. As you
can understand, realize based on whatever we have gone through in the previous lectures,
this concept can be achieved, implemented with the help of incorporation of different
algorithms; algorithms that can estimate how much the performance is going to be of a
particular machine or a process in the future and then optimizing its performance. This is
going to self-optimize the performance across the broader network. So, a network it has to
be there, which will help in the process and self- adapting to the different conditions, which
are coming in real time or non-real time, and automate autonomously running the different
industrial processes, the production processes.

So, let us before getting into all of these things let us first try to understand this smart
factory concept in little bit more detail. Let us say that we want to build a smart factory
[Link], what are the different components of it? We need to have smart machinery, smart
manufacturing processes, smart engineering solutions, and these can help in arriving at
the smart product. So, for all of these in order to make them smarter, we need to use
smart devices. So, this is a smart factory and there are there could be few other things
that you might be able to identify, which can help in making the product smarter.
Then we are talking about this interaction. This is the smart factory and the different
interactions in terms of the customer behavior, interactions with the customer, the behavior
of the customers on the factory, in the system. Smart grid is very important in terms of the
energy management and another thing such as when we are talking about any kind of
smartness, we have to deal with the analytics; cloud and big data.

And try to understand the different other aspects. So, why at all we need to have smart
factories? Why at all we need to have smart factories? In order to evolve our
technologies, the market is highly competitive nowadays. If you have smarter machineries,
smarter processes, smarter systems overall, it is quite likely through the use of these
systems you are going to have improved productivity, efficiency, improved monitoring of
this machinery, and reduce the down time. This will help in improving the competitiveness
in the market. In order to deal with this competitiveness, we need to have the smart
factories and high amount of production, within minimum timeline and this is quite self-
understood.

The last one is very important is reducing the risk of failure, which will affect the
productivity. So, reducing the risk of failure can be achieved through the use of smart
factories.

So, what are the advantages of running smart factories? Reducing the overall cost, cost
of production; increasing the efficiency of productivity, efficiency of the product, that is
being done that is being manufactured, improving the efficiency of the surfaces,
efficiency of the production processes can also be improved. So, overall efficiency in all
different terms can be achieved with the help of incorporation or transitioning to smart
factories.

Improving the quality, improving the predictability of different things like the health of the
machine; in the future predicting the health of the machine, predicting whether a fault can
happen in the future, improving the safety of the machinery, and the overall safety of the
workers in the factories.
These are the different components of the smart factory, smart machines, smart
engineering, smart devices, smart manufacturing process, then three things improving upon
the innovation whatever was existing innovating the product, processes and all these
things can be done with the help of information technology. Information technology
can help in making the overall factory smart. So, this is a very important component, IT,
use of IT in the factories can help in making the companies, the factory smart and the data
analytics. The data of all these smart factories are the factories when you are incorporating
IoT solution, sensors and actuators; they will throw in lot of data which will have to be
analyzed in order to make them smarter.
Smart machines help in communicating with other machines, communicating with other
smart devices, and with humans.

The smart devices will be connected. They will be connected between themselves and other
field devices, other mobile devices, that are operating. So, this connectivity issue is very
vital in the context of smart devices. Smart manufacturing process talks about the
dynamism in the manufacturing. So, taking care of dynamic changes in requirements. So,

autonomic autonomously the manufacturing is going to be done, without any human

[Link] timeliness is very important; efficiency is very important; thus, all of
these help in having smart manufacturing processes.

Smart engineering, smart design of the product, smart development of the product, smart
planning all of these are different constituents of smart engineering.

IT is something that I was also telling you, that it is very important in this overall
business of or building smart factories, we need IT solutions in terms of smart software,
smart hardware, smart management, smart monitoring, and smart control.

These are some of the characteristics of smart factories. Overall, when we are talking about
smart factories there is connectivity between different machines, connectivity between
machines and the different operators, connectivity between these different operators. Thus,
we have machine to machine, machine to human, and then human tohuman. So, all
this different connectivity is very important characteristic of smart factory.

Optimization, these smart factories are such that they will optimize their solutions, their
productivity, their processes, and overtime. They will have some kind of self-learning
behavior incorporated, implemented in them, through different algorithms. Thus,
transparency is very important proactivity and agility in terms of moving very fast. So,
moving very fast in terms of changes the dynamism, changes in the requirements,
changes in the environment, and changes in the environmental parameters.

Connection-connected smart devices, connected smart machines, and all of these

connected devices will bring in lot of data; so connectivity with the data center and
connectivity between the different processes.
Optimization-optimizing the tasks, scheduling of the tasks, optimizing the usage of energy,
optimizing the cost of production, optimizing tracking, optimizing throughput, optimizing
reliability, and many different other things will have to be optimized in a smart factory.

Transparency, because now you are able to track a particular component, a particular
machinery, track some logistic item, which are done in real-time. This improves upon the
transparency as well. In real-time, whenever it is required, in a smart factory that were ina
particular location. Therefore, tracking becomes very efficient, it helps in improving the
transparency.

Real-time monitoring and taking required action on time. If something has gone wrong
trying to know about it as quickly as possible in real-time and taking the requisite action
is very important and then generating alert messages to the people, who actually need to
have this information.

Proactivity-predicting the quality issues, improving safety, forecasting the future outcomes,
and predicting the future challenges.

Agility talks about overall flexibility which I was telling you earlier, flexibility in terms
of incorporating different changes, changes in requirements, maybe, changes in the
different production processes, changes in the legal compliance, and there are many
different things that might change over time, which are able to incorporate flexibly, what
agility talks about.
Adapting with the changes in all these production processes, changes in the legal
requirements, changes in the overall requirements. So, all these things are adaptation,
which is a very important component of the agility attribute. Self-configuration the
machines should be able to self-configure, whenever required they would be able to start
up on their own, work on their own, and configure on their own.
Supporting technologies for smart factories, big data technology is very important, cloud
computing is very important. Big data is about getting lot of data, in huge volumes, in high
velocity, having high variety, and many different attributes represented in terms of different
Vs.
These data will have to be handled and smart factories have produced lot. They are the
producers of lot of data, and these data will be useless until they are stored, processed,
and analyzed in certain way. So, for this big data technologies like hadoop or different
other big data technologies are used, different cloud computing platforms, public cloud,
private cloud. Different companies nowadays also have their own private cloud, and public
cloud services like Microsoft azure and Google cloud, all of these things are available,
smart grid is a very important supporting technology, because we are talking about
factories.
Factories are essentially the largest consumers, one of the largest consumers of electricity
and different other energy. So, being able to efficiently use the energy is very important
and having smart grids connected to the smart factories is very important.

The use of cloud computing in smart factory helps in the processing capabilities, providing
the capability of high performance computing, giving tools for running different
applications, different development platforms, storing the data. In a more scalable fashion
is what cloud computing provides in the smart factories domain.

Big data is about big data analytics, generates the knowledge from the data that is received.
So, we have the data, extracting information out of the data, generating knowledge out of
the information, that is processed from the data. These are very important for
improving the value streams, predicting the future, and taking care of the performance
indicators, improving upon the performance indicators, the KPIs is also a very important
aspect.

Use of smart grid in smart factories, persistence in energy consumption; if you are using
smart grid with smart factories, energy consumption will be taken care of in a smarter way.
There will be optimized energy consumption based on the exact requirements even the
energy consumption can be balanced across different processes throughout the day or
throughout the different times of the year.

Load balancing is very important, certain parts of the company might require certain
types of load electric load. So, certain components, certain companies’, certain parts of
the factory will require more load compared to the other parts of the company. This
differential loading and balancing the load throughout the day, across the different parts
of the company. These are the important aspects of smart grid.

This is the automation pyramid of a smart factory at the very bottom, we have the
devices in the device level, then we have the control level, then we have the manufacturing
system level, and the ERP level, in between we have all these different communication
layers.

Use of augmented reality is quite rampant in smart factories. This will help in operating
instruments remotely. So, remote monitoring of different machinery, different
equipments in factory is possible, with the help of augmented reality.n fact, there is another
one which is the virtual reality, that we have discussed in another lecture, augmented reality
and virtual reality together will help in remote monitoring, remote operations of
instruments in a smart factory. These technologies can also help in improving the
precision, providing precision in the processes, in the production processes, improving
the safety especially for radioactive zones, and different other safety critical zones.

So, with this we come to an end of the lecture on smart factory, I hope that you have
some basic understanding about what smart factory is, and why smart factories are
required, and what are the essential constituents of a smart factory.
These are some of the references that have been given for you.

And if you are further interested you can go through, in the internet you will be able to find
lot of different other lectures and different other materials on smart factories.

So, please feel free to go through them, if you are interested to know more about the
smart factories. But smart factory is more of a concept there are different building
technologies and different components. So, all these different technologies the enabling
technologies for building smart factories, this is what we are trying to get a glimpse of
through the different lectures of this particular course. And if you really need to build a
smart factory you have to start implementing these concepts that we are going through in
this course.

So, with this we come to an end of it, smart factory is a very important concept, in the
context of Industry 4.0. Industry 4.0 one of the most fundamental, and central theme in
Industry 4.0 is building smart factories. With this we come to an end.
 Introduction to Industry 4.0 and Industrial Internet of Things
Industry 4.0 Cyber-Physical Systems and Next Generation Sensors

In this module, we are going to talk about Industry 4.0 and the different aspects of it. In
Industry 4.0, first of all is a very popular wave, that is going on worldwide, among all
different industries. Everybody wants to transform themselves into Industry 4.0
compliance. For Industry 4.0, there are different aspects that are important and should be
understood in order for the industries to be able to comply themselves with the
requirements. For instance, one of the very important aspects of Industry 4.0 is automation,
full automation. Automation where there is no human intervention ideally or to a large
extent, there is reduced human intervention.

So, the entire product line, the entire production process in a manufacturing plant, for
example, would be made highly autonomous without any human intervention. And so, how
can it be possible? It can be possible with the help of all these technologies that we have
gone through so far like sensors, actuators different aspects of communication between
them. So, these are going to be the enabling technologies for complying with the
requirements of Industry 4.0.

So, we need sensors, actuators, communication, automation all of these things. But how
these are going to be done? These are going to be done with the help of these IoT devices
and something called CPS--Cyber Physical Systems and these Cyber Physical Systems is
what we are going to go through in this particular lecture.

So, the idea is that there are different enabling components of cyber physical of Industry
4.0; Cyber Physical Systems is one of them. Likewise, there are different other ones, which
we are going to go through in the next few lectures of this module on Industry 4.0 basics.

So, what is cyber physical system? There are different definitions that you will be able to
find in the literature and people still confuse between what is IoT, what is CPS, what is
the difference between them, what is M2M, like that there are different allied technologies,
which have overlapping scope. But cyber physical systems is something that the scope
is very clear. So, as this name suggests cyber physical. So, these are systems where there
is a strong component of the cyber world and the physical world. There is a strong
component of the physical world. So, there is interaction between the cyber world and the
physical world together in these systems. So, how it is possible?

So, let us look at the NIST definition of what a cyber physical system is. So, cyber physical
system also often known as smart systems are co-engineered interacting networks of
physical and computational components. Interacting networks of physical components and
computational. So, computational is the cyber one and physical is basically the physical
world in which these systems are operating, physical world.

So, these systems will provide the foundation of our critical infrastructure; so, different
infrastructure. All these public infrastructures that are there not only public, private
industry infrastructure all this critical infrastructure that are there. These will be
supported using these cyber physical systems in the next generation internet, the Industry
4.0. So, these are going to be all supported with the help of these critical structure will be
supported with the help of these cyber physical systems. So, we need to
understandreally what these cyber physical systems are through the next few slides. But
what is important is with the help of these critical, these cyber physical systems the
quality of life will be improved because gradually we will be getting into smarter
world, where there will be smarter systems offering smarter services.

So, for smartness autonomy in every respect is very important; autonomy and
suggestions, based on the data that are collected are very important. So, that basically
completes this definition of the cyber physical systems as per NIST.

Now, the question is that what is a cyber physical system, but before that we need to
really understand what is an embedded system. Cyber physical systems are embedded
systems. So, these are basically cyber physical systems are you can think of conceptually
as cyber physical system, conceptually as embedded system plus the physical system
together you what you get is the cyber physical system.

So, what is an embedded system? So, you have some kind of software that is working on
some kind of hardware. The software is embedded in the hardware. Embedded systems
have found a lot in the cars; that means, vehicles, these are found in aircrafts embedded
systems are found everywhere.

Nowadays, embedded systems are enabling technologies for making systems smarter and
that is why these embedded systems are very popular. So, these embedded
systemsirrespective of where they are used, they possess certain capabilities of
computation, communication and control, compute, communicate and control capabilities.
So, where there is interaction with the physical world, through different sensors and
actuators examples of it would be medical instruments, many medical instruments are
embedded they have embedded systems in them, transportation vehicles, defense systems
have embedded systems in them.
For example, the radar and different other systems, that are used even the aircrafts etcetera;
they are all having different embedded systems in them. Robotic equipments likewise
process monitoring and factory automation systems, all of these have embedded systems
in them. So, this is the cyber physical system.

So, now let us move forward and see how an embedded system works. So, as I told you
that an embedded system has the computational capabilities. They also have the processing
capabilities, so, computation communication capabilities and so on. So these systems
would help in building the cyber physical system. So, in the cyber physical system, you
have the physical systems--the physical space. So, this is basically the cyber space. This is
the physical space, and the interaction between them will make it the cyber physical system.

So, what does this physical space have? So, a physical space we are typically talking about
different real objects like human beings, like houses, factories automobiles,buildings
and so on. And different objects, these communication objects, for example different other
household objects or even the objects that are in the industries like cell phone, cameras,
washing machines, refrigerators. All of these are different objects these are different
objects that work in the physical space.

So, these objects basically will sense certain data. They will sense different information
in the place, where they are operating and then using certain communication devices like
the ones shown over here. These data are going to be sent further. So, this sensing once it
is done, this sensed data is sent for further processing. So, this processing will be done
using different servers, server firms, cloud or, whatever. So, that becomes the
cyberspace.

So, it will be done this processing will be done in the cyberspace and based on the results
of this processing, those results will be sent back to the physical space for actuating
different devices. So, these devices could be maybe, starting a washing machine, starting
a refrigerator, if there is certain thing, that has happened; starting the air conditioning in air
conditioner in the room.

So, let us say, that it has been sensed that the temperature has gone down in the room.
So, based on their analysis that is performed in the servers over here maybe based on
certain rules that are already pre-programmed in those cyber spaces. So, finally, based on
that the actuation is going to happen and the physical space will be actuated with the
intention of starting some device or operating some motor or anything like that.
So, these are some examples that I have given you. This could be generalized further and
we can try to understand how, in general, a cyber physical system is going to work.

So, before I go any further, let us again try to understand in little bit of depth about how
this cyber physical system, CPS is going to work; how the CPS is going to work.

So, if you are talking about cyber physical systems, we have as I told you we have the
cyber world, the cyber world, and the physical space. So, what is going to happen is this
physical space is going to send some information. It is going to send certain information to
the cyberspace and then some decision is going to be made over here some decision is
going to be made. Based on this decision certain action is going to be performed
andbased on this action, the physical state of certain device, physical state of a device
is going to be changed. Interaction primarily between devices device to device, but it could
also be device to humans or it could be even human to human. So, all kinds of
interactions are going to happen, but this is basically the kind of loop that we are talking
about this loop is going to take into effect.

So, what we have essentially is some kind of a loop, some kind of a loop between the
sensor. You have sensing, rather let me call it not sensor, but let me call it sensing. So, you
have a loop between sensing, then computation, and then control. So, this kind of loop is
going to exist in cyber physical systems. So, going back so, we need to understand
that what is so, special about these cyber physical systems and how they differ from
the embedded systems in general. So, we have in embedded systems we are talking about
devices alone the devices that will process some information and this information are
basically embedded in these devices, they are stored in these devices and so on.

On the other hand, the cyber physical systems, these are complete systems, where there
is an interaction between the physical components and the software or, the cyber world.
The next difference between cyber physical systems and embedded systems is in an
embedded system, typically, you are talking about a single device and this tip the single
device the embedded system is basically confined to this particular single device. On the
other hand, in a cyber physical system, we are talking about a network set of such kind of
embedded systems.

So, the networking of different embedded systems will typically exist in a cyber physical
system. Embedded systems have limited resources for performing limited number of tasks.
On the other hand, a cyber physical system may not be resource-constrained and there you
can perform large number of different tasks at the same time. In an embedded system, the
main issues are real-time response and reliability, on the other hand, in a cyber physical
system in an addition to these the main issues are timing and concurrency. So, these are the
differences, primary differences, between cyber physical systems and embedded systems.

So, here are some features of cyber physical systems. Reactive computation is very
important and that we have seen when I explained to you about this feedback loop, the
control loop and so on. So, reactive computation means interaction of the system with
the environment in a continuous fashion. So, basically there is a sequence of observed
inputs and outputs in the process and that has to be dealt with by these systems suitably.

Concurrency is very important, concurrency as this name says we are talking about
concurrency of execution of multiple processes at the same time. So, these processes,
concurrent processes would exchange information to achieve certain expected result and
these operations could be synchronous or, asynchronous in terms of their operation or, their
modalities of their operation.

Some of the other features of cyber physical systems; there is a feedback control loop
that I already talked about and these basically these systems will have some kind of a
control system in them there is some kind of a control element in them with certain
feedback loop. So, basically the sensors would sense the environment and these actuators
will basically influence it and there is hybrid control system these are basically hybrid
control systems for complex tasks. If there is a complex task, simple control systems will
not work, it will be a network of different control systems it will be a hybrid control system,
which will basically help in achieving the task. Real-time computation is quite well
understood and supporting safety critical applications is what is very attractive of cyber
physical systems.

There are different applications of cyber physical systems. Healthcare is a very important
domain where cyber physical systems are widely used. So, these basically in healthcare
as that what is very important is to have medical devices and systems, which will
measure things at a very high level of accuracy. Some physiological parameter of a patient,
for example, has to be measured very accurately by different medical devices and
systems. So, obviously, as you can understand if you think a little bit in deep you will
be able to realize that cyber physical systems will be very attractive of use for healthcare.

For example, you must have heard about the use of robotic surgery or image-guided
surgery. These are examples of cyber physical systems and they are used in surgery or
healthcare, in general. Then we if we are talking about fluid flow. So, control of the fluid
flow, control of blood flow into a patient, when required for medical purposes may be
somebody having a bone marrow transplant or some kind of blood transfusion is taking
place. So, cyber physical systems can find use to perform the tasks accurately. The
system itself will be taking care of the accuracy to a great extent which if a human body.
If a human was performing, then they would not be able to do it that much accurately; so
cyber physical systems.

So, robotic surgery, already probably you must have heard that robotic surgery is very
popular because of the precision, precision in surgery that can be obtained using
roboticsurgery or image-guided surgery, these are very precise and known. So, the doctors
can perform this surgery very precisely accurately. So, that is where in healthcare this kind
of cyber physical systems can be useful.

So, let us now talk about other kinds of engineered systems such as exoskeletons,
orthotics, like entertainment robotics, therapeutic robotics, brain machine interfaces,
prosthetics, and so on. So, these are like different other types of examples of cyber physical
systems and their use in healthcare and these are based on cognition and neuroscience.

So, the next application domain of cyber physical systems and their use is transportation
cars vehicles in general aircrafts they all use cyber physical systems. So, there could be
infrastructure-based transportation cyber physical systems for real-time monitoring of
traffic infrastructure such as traffic signals, cameras etcetera and traffic control vehicle
infrastructure, coordinated transportation, CPS for transit signal priority, queue warning,
example for ambulances etcetera. So, these are very important priority-based, priority-
based queuing, priority-based system, integration and so, on. So, all these things in
transportation could be addressed, with the help of cyber-physical systems.

Vehicle-based transportation cyber-physical systems, where proximity detection for safety.

So, you must have boarded vehicles, where the vehicle will warn you if the vehicle
is too close to the vehicle, another vehicle in front or if there is another vehiclecoming
from the back, which is close to that particular vehicle. So, all these things are examples of
cyber physical systems operating on the road on the vehicles and so on.

Likewise, other application domain would be in the energy sector, in the smart grid.
Smart meters are examples of cyber physical systems smart meters can do demand side
management with distributed generation, automated distribution with intelligent
substations, wide area control of smart grid. So, all of these things can be performed with
the help of smart meters and smart meters are good examples of cyber physical systems
and their use in the energy sector. Cyber physical systems in smart grid smart meters is
one, but then there are other examples also like PMUs Phasor Measurement Units, Data
Acquisition Unit, and so on. So, all of these are basically examples of cyber physical
systems in the energy sector for smart grid.

In the industry, in general, manufacturing industries will use cyber physical systems. All
these different machines that we talked about some of these may be robot assisted
machines; some of these could be simple machines. So, machines, such as let us say the
lathe machine in a workshop, milling machine in a workshop, the welding machine in a
workshop. So, all of these could be equipped with different components to make them
smarter for manufacturing, improved manufacturing, improving the efficiency of
production using different sensors and actuators and different other processing. These
could be made smarter and these would be also examples of cyber physical systems, in
the industry, in the manufacturing industry.

So, these smarter cyber physical manufacturing systems can perform different things
such as smart control, optimal resource utilization, smart diagnostics and maintenance,
safety, safety of the industrial environment, safety of these machines, smart monitoring
of the health of these machines. So, all of these things could be performed using cyber
physical systems attached to or rather machines, which have been made cyber physical.
So, these would be examples of cyber physical systems for use in the manufacturing
industry.

Basically we were talking about Industry 4.0 and we have seen that at the outset, I
mentioned that cyber physical systems are very attractive for use in the manufacturing
industry and other industries also for moving towards, for leaping forward, towards
achieving Industry 4.0 goals, in general. So, in terms of Industry 4.0, there are 5C
architecture goals, which can be achieved with the help of cyber physical system.
Connectivity, conversion, cyber cognition, and configuration, these are the 5C architectural
aspects. So, connection: so, we are talking about smart connections. To do what to ensure
that the accurate data is obtained from these different IoT devices, IIoT devices, in the
industry; these are obtained from these devices. So, here two factors should be
considered obtaining seamless and tether-free data and selection of sensors with
proper specifications.

Conversion we are talking about by this term conversion. As the conversion of machine
data to some information, that can be made meaningful and useful. So, data analysis
tools and methodologies can be developed for the prognostics and health monitoring of
different components, different parts of the machines, and for multi-dimensional data
collection, and data correlation analysis. So, basically what will happen is these
machines will be made smarter, they would be made self-aware.

Then comes the cyber component and this is quite obvious. We are talking about analysis
of the data that is informed that is received at some kind of a hub which will be
comprised of different high end workstations, servers, cloud or whatever. So, that is the
server component that we are talking about.

So, this data that is obtained can be used for analyzing the data, mining the data, storing
the data and then based on certain algorithms that are run on it on the data or using the data.
Then making others aware of what is going to happen next or deriving some useful
information at the same time. So, this is basically done by the cyber component of it. So,
this is the C of the 5Cs in the architecture of CPS for a IIoT for or Industry 4.0.

Cognition is another one. So, here by cognition what is meant is basically the proper
presentation of information to users for generating thorough knowledge of the system.
So, we are talking about collaborative diagnostics, decision making for prioritization,
optimization of processes and so on. So, all of these are different aspects of cognition
that come into picture in the CPS systems.

Configuration is the other one where we are talking about supervisory control to determine
actions to be taken by the machines themselves. Examples would include self-
configuration for resilience, self-adjustment for variations, and self-optimization for
different disturbances. All of these self properties would help in the supervisory control for
performing certain actions by the machines themselves. These machines are going to be
self-adaptive consequently.

So, there are different challenges of cyber physical system development; challenges with
respect to safety, security, robustness, computational challenges real-time embedded
system abstractions sensing and mobility challenges are there architecture and modeling
verification validation and certification and including education and training of these
systems. These are all challenges for of the deployment of CPS for in the industries towards
the compliance of complaints for Industry 4.0, the next little while we are going to discuss
about the next generation sensors, that are going to be used.
So, all these sensors that we talked about in the introductory module before these are simple
sensors. These sensors can only sense and then send the sensed data forward. So, these can
be thought of like smart sensors, but then for next generation sensors throughout the
world industries academic labs and so, on.

They are working on making sensors intelligent. So, not only that these sensors the smart
sensors would sense and send, but these would be made intelligent because certain small
algorithms lightweight algorithms will be executed in these sensors to do certain tasks at
the sensors themselves. So, these are, this is very important in the context of Industry
4.0.
So, here basically you are talking about integration of sensors and actuators, with a
processor and a communication module. This is basically these traditional sensors, how
they look like. So, there will be a sensor component an actuated component optionally
and a processor for processing certain, whether the data that is sensed and certain
communication module for connecting these different sensors with a central hub for
processing or the sensors themselves. So, these are the simple smart sensors these
sensors will also have to do certain functionalities like self calibration, communication,
computation, multi-sensing cost improvement of their [Link], these are some of these
different features of the smart sensors, but these smart sensors have severe limitations.
They have predefined embedded functions that cannot be used for certain customized
scenarios. So, you have to customize them based on certain requirements and that
makes it difficult for use in real life industry scenarios. Narrow application spectrum is the
second limitation sensor data aggregation using the sensors is not possible or limitedly
possible and external processor for sensor calibration is required for these smart sensors
and also they would support very basic lightweight communication protocols.

So, for overcoming all of these different limitations of smart sensors, next generation
intelligent sensors have been proposed. They are in the works and these are the different
features of these intelligent sensors.

So, these intelligent sensors are capable of processing sensed data and performing
predefined functions by processing the data. So, they are capable of customizing embedded
algorithms on the fly unlike the smart sensors which were not able to do so. They are
capable of managing and controlling external sensors and devices and they would comprise
of a sensor, a microcontroller unit, a memory unit comprising of flash RAM, ROM, and a
platform for running different sensor applications.
So, there are; obviously, different advantages of these intelligent sensors, in terms of
reduced data communication, reduced power consumption, shorter software development
time, improved compatibility of sensors, possibility of continuous calibration and
monitoring of the sensors. These are the different advantages of these intelligent sensors
and that is why these are very attractive.

So, this is how is this is how a smart sensor would look like a smart sensor this is an
example of it. So, as you can see over here, there is a digital interface and
embeddedthere are different embedded features signal conditioning circuit, then you have
ADC Analog to Digital Converter and the sensors or MEMs which actually sense the
material, environment or, sense whatever it is supposed to sense.

So, in addition to all of these things in the intelligent sensor on the other hand, you have
mostly whatever is present in the smart sensors, but also a component for processing and
computation. So, you have this RAM, ROM microcontroller flash and so on. All of these
basically make these smart sensors intelligent, by virtue of having these components in
them.
So, there are different applications of next generation sensors these intelligent sensors.
Automatic assembly in factories, smart fabric and intelligent textiles, advanced driving
assistance systems, fault detection and forecast using machine intelligence, non-invasive
biomechanical analysis, chemical component analysis resource lifecycle management.
These are some of these different applications of these next generation, intelligent
sensors.

So, there are different design challenges in the manufacturing of these intelligent next
generation sensors. There are different hardware issues. These sensor nodes, they
themselves are already limited in terms of power, they have very high response time and
there are synchronization issues, issues of limited bandwidth of operation, security issues
are also there and there are different other software issues such as the issue of software
partitioning for complying with different applications and their different processors in
them. So, all of these are different design challenges with respect to hardware and
software. So, with this we come to an end of cyber physical systems and next generation
intelligent sensors, discussions on them. So, this brief of brief idea about cyber physical
systems and the use of intelligent sensors basically equips you with an understanding of
how you can go forward in the years to come, if you have to make your industry compliant
with Industry 4.0 expectations, the objectives of Industry 4.0 and so on.
So, cyber physical systems and IIoT are strongly interlinked. So, essentially what we are
trying to do is we want to use these cyber physical systems. The abstractions of these cyber
physical systems in the form of different objects; objects means we are talking about virtual
objects not the physical objects. So, these different physical or virtual instances of the
corresponding physical objects in the physical world, these could be networked together
and the data would be received at the other end for further processing analysis and so on
for making the industry smarter. So, we are gradually leaping forward towards building
industrial internet of things and in the industrial internet of things IIoT using CPS, we are
going to talk about in further detail later on.

So, these are some of these references and if you are interested you can go through these
references in further detail, to know what are the different other issues of the cyber physical
systems and how next generation sensors could be developed; if you have to develop these
next generation sensors.
So, these are different concepts knowing these concepts is required, but then if you have
to develop a cyber physical system yourself or if you have to develop a next generation
sensor, an intelligent sensor, that is a complete completely different ballgame. So, you need
to be specialized, there are researchers who spend years together to build a sensor and
intelligent sensors it will take even more effort to build these intelligent sensors.

So, obviously, that is a completely different ballgame, but here we are just getting an
exposure to these different concepts of cps, next generation sensors, and so on which will
be required for building Industry 4.0.

.
 Introduction to Industry 4.0 and Industrial Internet of Things
Industry 4.0: Collaboration Platform And Product Lifecycle Management

For Industry 4.0, there are different technological considerations that are required. One
thing that is often not looked upon seriously, but it is very important from our viewpoint
for Industry 4.0 compliance is to take help of the knowledge base. Knowledge base from
the same industry from a previous production processes, existing production processes in
other units, or from the knowledge base that can be obtained from other industries
following similar kind of production processes.

So, this knowledge base will be very helpful, that is why we need some kind of a
platform that can help build a knowledge-sharing mechanism. Collaboration platforms
are very important. And also in this lecture, we will talk about product lifecycle
management. How you can achieve automation in PLM - products lifecycle management
for trying to step forward to meet the Industry 4.0 objectives. So, let us look at some of
these features, the high level understanding about each of these, in the next little while.

So, what is a collaboration platform? So, basically we are talking about a platform which
can be a software platform, which can be a software-hardware platform, which helps the
employees to share information and solve certain business problems. So, employees can
share information between themselves using a collaboration platform. The employees of
one industry can share the information with employees of another industry by virtue of
use of these collaboration platforms. And this thing you can generalize and extend to
different scenarios, likewise.

So, collaboration platforms essentially are talking about building platforms that will
include in the sharing of knowledge, a platform for managing the knowledge, knowledge
management and including it in the business operation for renovation of the business
processes, and improving upon the efficiency of these business processes in the future.
So, this is what a collaboration platform will help in doing.

So, what is a collaboration platform? There are some different perspectives to build
collaboration platforms. There is a social layer component that is introduced in the business
social layer; because we are talking about employees we are talking about users who will
share their different knowledges with the help of these platforms. Social layer component
is combined and integrated, with the provisioning, with the provision of business
utilizations, but then you need some kind of a platform. So, you need some kind of a tool,
which can be integrated with the existing products or the new products, that are being
developed or, are being acquired and implanted in the industry. So, there are some common
attributes in business collaboration platforms. These collaboration platforms should be
easily accessible, and this should be easy to use. And they require some familiar function,
which help in the collaboration between the team members or across different teams. One
such platform is named known as pro-work flow that is the name of a collaboration
platform that is often used in the industries.

So, this is a web-based project management platform that is designed for managers. And
then this platform basically helps employees to collaborate and improve upon the project
delivery in the production process, in the project management process. So, this is an
example of a collaboration platform. Let me now show you how this collaboration
platform is going to work.

So, we are talking about a collaboration platform. This collaboration platform will help
in collecting knowledge, collecting knowledge. So, this knowledge will be used in
knowledge management. So, basically what is happening essentially is this collaboration
platform is helping or, it is enabling knowledge management.

So, this knowledge management is important for different industry agents, there could be
other industry agents, likewise, and industry agent. So, these industry agents would be
sharing information between themselves, and they would be working together for achieving
the objectives of the business. So, this is how a collaboration platform works, and how it
is linked to the knowledge management aspect of it. So, essentially, what is happening the
idea is towards the objective of improving the productivity, improving the efficiency of
these business manufacturing processes, and so on for fulfilling the overall objectives of
Industry 4.0, we are saying that these collaboration platforms would be found to be useful.
So, collaboration productivity is going to be improved, is going to be increased. So, there
are four different aspects of increase in the collaboration productivity. And this will be
enabled with the help of different key parts such as IT, single source of truth,
industrialization, and coordination. Say let us look at each of these individually in little bit
more detail in the next few minutes.
IT for IT proliferation. So, basically what has happened is as we know that computers,
computational devices has a huge impact in our economic growth in the last few decades.
And there has been increase in the capital stocks and shares across different companies
worldwide. So, these industries worldwide are required to consider and promote global
information technology, global information systems, with the help of this computing
power. So, storage capacities have increased computational high speed computational
capacities have increased and they are increasing even more day by day. So, this IT
proliferation has basically helped in building the computational platform that will be
required for coming up with the collaboration platform.

Single source of truth means we are talking about a single repository, where all the data
are going to be stored. So, essentially we are talking about a data model and information
model that will help in storing every data element exactly once. So, this single source of
truth must employ the right software, at the right time, at the right place for right
decision making. So, single source of truth is needed to be realized across the whole
product life cycle, and this is very important.

Industrialization. So, it is basically the bridging between the virtual world and the
physical environment. So, the physical environment is linked with the virtual world
using cyber physical systems which we talked about earlier. And these cyber physical
systems are equipped with computers, sensors, actuators. So, these systems will require
intuitive and self-effective elements to be adopted in them.

So, this is the industrialization aspect of it. And coordination is basically to ensure that
there is stronger coordination between multiple industry agents in the same industry, across
different industries, and so on, so that the collaboration productivity can be increased even
further. Coordination can be initiated in two steps. In the first step, we can establish a
network which communicates with the overall target, and in the second, providing
authority to decision makers in a decentralized system.

So, this network we are talking about in the context of coordination is maintained by
encouraging the exchange of employees or by using smart devices between different
employees, because exchange, physical exchange, may not be required, if we equip your
employees with different smart devices. This exchange of information can be maintained
for stronger coordination between the employees of the same organization or across
different organizations and so on.

Now, let us try to understand the product lifecycle management aspect of it. So, product
lifecycle management as this name suggests, this term suggests it is actually a widely
used terminologies in the industry. So, it talks about the lifecycle of a product from ideation
till deployment the entire lifecycle of a product is captured using these product life cycles.

So, this product lifecycle management works as a management system for a company’s
products. PLM handles a product completely from single part of the product to the entire
portfolio of that product. Examples of product lifecycle management systems are basically,
the CIS system, computational intelligent system, and there are many different other
industry specific PLM systems that are widely used in the industries. And there are so
basically these PLM systems are quite popular.

The main goal of product lifecycle management is to maximize the product revenues, to
decrease the product associated costs, and to increase the products value. So, obviously,
as you can understand, these are very attractive features, which will be required to be
implemented in order to move towards Industry 4.0, improving efficiency in the Industry
4.0, improving automation objectives in Industry 4.0.

So, P, L, M, this P, P means product. So, the product is basically the central theme in the
industry, everything is governed around products and processes. So, P corresponds
toproduct, which is basically the central agent of consideration in PLM in the industries.
So, the product is the origin of company earnings. So, this P is obviously, as I was telling
you so far the most important component of product lifecycle management. And we can
understand that if you do not have product, you do not have its services. So, every
service is basically linked to that product.

The L of product lifecycle management stands for lifecycle; L stands for lifecycle. And
there are five different parts, five different phases in this product--lifecycle so in the
lifecycle of PLM. So, one is basically visualization visualize explain perceive use and
then dispose. So, visualize means ideation, ideation explaining the idea, perceiving the
idea, realizing the idea, supporting the system, the product that is built, and then
disposing or retiring the product, that is built. So, these are the five different parts, five
different phases in the product lifecycle of any product lifecycle.

Visualization as I was telling you corresponds to the ideation. People have different ideas
regarding a product. So, they visualize this product to be built that is the ideation part. Next
is the explanation this idea has to be transformed into some kind of a representation that is
basically captured through the phase explanation phase. Perceiveness or perception is
basically at the end of the previous phase, the product has to be built in its final form. And
then use or support is basically the customer basically starts to use the product in the user
support phase. And then once it is used, once it is no longer useful to the company, the
product has to be retired.
So, why we are talking about all of these things, we need to understand the different phase,
the different phases of the lifecycle of any product. M in PLM means management.
So, product lifecycle management, M stands for management.

Management, we are talking about product management over here, where the
considerations of coordination and institution of product-related devices are important it
is required to fix different objectives. And enforcing the capabilities of decision making,
and taking result of the control, these are the different important considerations, in the
management aspect of PLM. So, the whole idea in the management aspect of PLM is to
ensure that a product works well, it is managed across its lifecycle and the management
guarantees that the product will earn the profit for the company.

In terms of Industry 4.0, it is required to improve efficiency, effectiveness, automation

has to be poured in order to be able to achieve it. So, automation in the PLM is
something that the industries will have to be striving for, with the help of different things
such as computers alone embedded systems, cyber physical systems everything has to be
taken together in order to improve upon this efficiency and effectiveness in the automation
of PLM. So, this efficiency and effectiveness of PLM improves the market share and
market size, with increasing revenue.

These are some of these business objectives of PLM for Industry 4.0, financial
performance, which obviously talks about increasing the market share, increasing the
market revenue, reducing the development cost. Time reduction basically, reducing the
project time overrun, decreasing the profitable time; improving quality talks about
decreasing, the defect rate that means the rate at in which, rate at which the
manufacturing defects in these different products are going to be there.

So decreasing that rate improving the customer satisfaction and so on, these are basically
improving quality aspect of the business objectives of PLM. And overall improvement of
the business decreasing the delay time in product release ensuring 100 % configuration
conformity, improving 100 % customer satisfaction, these are all the different business
objectives of PLM for Industry 4.0.
So, if we are talking about PLM, we need to understand few things. So, let us look at some
of these different aspects.

So, let us talk about the scope of PLM, in the context of Industry 4.0. So, first of all you
need to design the product. Then this product has to be built in its final form conversion.
Conversion of the design into the physical form of the product and then this product will
be delivered. So, all of these, basically will have to be automated to the extent possible in
order to comply with the Industry 4.0 objectives. So, we will be going back what is the
scope of PLM. So, there are nine components in PLM to handle a product across its
lifecycle. And these are the different components of it. These are the different components
objectives and matrix, organization and management, activities, people, product data,
product data management system, different PLM applications, equipments and facilities,
techniques, and methods. So, all of these different nine components will have to be handled
in the lifecycle of a product in PLM.

o, objectives and metrics, the objective of a company for PLM is to improve the quality
and business, reduce the time, improve the financial performance. So, KPI is commonly
known as KPIs key performance indicators are considered in the companies these are
basically some kind of a metrics that are considered in the company. And these are targeted
by the employees to be achieved, towards the development of a particular product.

The next thing is basically the organization and management. Here we are talking about
resource management aspects. And the overall management of the different resources
resource management means all kinds of resources. All kinds of resources including the
product, the infrastructure, the human resources and all kinds of resources the management
aspect of it should also be consideration alongside.

Activities, there are many products product associated activities such as idea
management, program management, new product development. People are involved to
progress and maintain a product, example a business analyst, a cost accountant, so these
are the different people aspects. And the product data it is a major asset, throughout the
product lifecycle.

So, product will face problem if we provide false product data. So, product data means like
you falsify some kind of a data. You say that the product is defect free, but let us say that
there is some small or big defect, that exists in the product. So, this falsification should not
be done. So, then the product itself will fail, and it will face problem in the market.
Product data management system. As this name suggests, it manages all the generated data
and that is used for improving the life cycle product lifecycle overall. So, it provides correct
information at the right time. PLM applications to get desired performance levels, different
applications are required, which are responsible for enabling the people to take different
decisions. These applications support the people to build and maintain these different
products.

Equipment and facilities talking about the use of different equipments, use of different
facilities, the use of different infrastructure, that are already existing in the industry and
procuring the ones that are not already there in every phase of the PLM. Techniques and
methods, to refine the production across the lifecycle, by means of product progress
time, product cost, many methods and techniques can be used such as activity based
costing, concurrent engineering, design for sustainability, and lifecycle assessment.

Business drivers there are many new business challenges if you have to adopt PLM for
complying towards the objective of Industry 4.0 there are different business challenges
such as reduced or shortened product lifecycle. Increased outsourcing in the new
products in the context of Industry 4.0, and the products structure has become complex.
So, think about IIoT systems, think about a CPS system, in Industry 4.0. We are talking
about small sensor enable devices small actuator enabled devices.

So, the way these IoT systems are developed, it is a kind of different product lifecycle
that is used for these IoT systems. So, the essentially what has happened is in IoT the
product life cycle has reduced it has shortened, and a lot of things are basically lot of
components are outsourced. And many of them are in sourced from different companies.
So, basically this outsourcing and in sourcing in the context of IoT, and their use in Industry
4.0 has increased and consequently the structure of the product the building ofthe product,
because there are so many different complex components in these different products IoT
products that the product structure has also become complex.

So, these are some of these new business challenges for PLM in Industry 4.0 that will have
to be kept in mind. Increase in speed, increase in demand, and quality of product are the
other challenges to drive a particular business.

Industrial requirement. Nowadays, we are not talking about physical presence of the
different units, there could be virtual presence also. There could be geographically
dispersed design teams supply chain itself has been made virtual to a large extent. And this
could be further increased with the increase of adoption of these IoT systems. So, what is
required is that the different industry components will have to collaborate with one
another. So, net-centric technologies have increased in their use in PLM nowadays, and
that has also not only improved not only increased the efficiency of the production process,
but has also made it much more complex, in terms of development.

So, for Industry 4.0, these are the 10 step approaches that can be used for PLM. Number
1 data gathering education of PLM. So, education component is very important and is often
neglected, but education and training is very important. Practice of management in PLM
concept, PLM roadmap generation, development strategy, rate of interest calculation,
management report preparation, executive preparation and executive decision
support.

If you need to know in further more detail about each of these many of these are quite well
understood, and I think most of them you can understand from these names alone. But if
you need to understand in further more detail this is the source that can help you to
understand in further more detail. So, you can go through this particular literature in
order to understand these.

So, with this we come to an end of this lecture. So, collaboration platform product lifecycle
management, their automation improvement in the context of IoT, IIoT, Industry 4.0
is very important. So, these are often overlooked, but should not be from our perspective
and that is why we have included these things in this particular course on Industry 4.0 and
IIoT.
 Introduction to Industry 4.0 and Industrial Internet of Things

Industry 4.0: Augmented Reality and Virtual Reality

In this lecture we are going to get a brief overview of Augmented Reality and Virtual
Reality in the context of a IIoT. So, augmented reality and virtual reality, they are popularly
known as AR and VR, are quite important technologies, that have been adopted and
are being adopted at present in different IIoT context, in the different industries.
Because of different advantages that they offer, there they have some relationship between
them they are interlinked, but they have some distinct differences.

So, in the next few minutes, we are going to understand the specific attributes of AR and
VR and the different challenges, that are there in the adoption of AR and VR
technologies in IIoT settings.

So, from a technological perspective, both AR and VR they can be used in different
context, different industry sectors, they can be used and together they can help in improving
Industry 4.0. So, they play a very important role in the primary stages of manufacturing,
where optimization and productivity are important considerations. Because in any
manufacturing industry it is not just the manufacturing of the product, optimization of
the processes, improvement of the productivity, these are important considerations.
The efficiency of warehouses is also very important and these can be improved using
different AR applications. Both AR and VR are also important in safety scenarios.

So, for example, there are situations I will also talk about a scenario, where VR as well
as AR can be used for training of industrial safety. So, we will talk about that in another
lecture not in this lecture. So, I will showcase how AR VR can be used to improve the
training of safety in industrial plants. So, these are very important technologies. So, that
is the reason why I thought that a basic understanding about these technologies can help
in improving the IIoT infrastructure in the industries.

The different use cases that are served by AR and VR for IIoT are things like machining
and production. These technologies AR VR can be used to improve the overall
machining processes, production processes, and in education, training, collaboration,
assembly line, safety security of different infrastructure in the industries. And digital
prototyping, factory planning, maintenance and inspection, these are some of the
different uses of these technologies in the IIoT settings. So, these are very important, but
remember one thing that through this half an hour lecture, you are not going to learn about,
how to use the AR and how to use [Link], here you are just going to learn about the
different features of AR, different features of VR, how they can be used, where they can
be used, where they cannot be used. So, these are the things that you are going to get an
understanding about. You cannot learn through this half an hour lecture how to become
a master of use of these technologies AR and VR. So, this is basically the overall, this is
the overall scope of this particular lecture.

So, let us now try to understand what is AR. So, augmented reality basically, as this
name suggests augmented. So, physical reality is augmented with certain information,
typically, some computer generated digital images augmented with that kind of information
to give a better feeling, better perception to the users, improve perception of the reality to
the users.

So, it AR basically will improve the feeling of the reality, the physical reality of the
world in which the user is operating. And that improved feeling as I just said can be offered
with the help of computer generated technologies, typically different types of images,
videos, etcetera. So, these digital elements are added to the actual environment and together
we have an amplified environment, where the present perception of reality is improved
this is what AR is all about.

So, some of the key features of AR, it lies in the middle of the mixed reality spectrum.
So, we are talking about different types of reality; we are talking about augmented
reality, we are talking about virtual reality, we are talking about no reality at all, completely
physical environment, we are talking about completely immersed kind of environment. So,
AR is some kind of mixed reality kind of environment VR provides mixed reality
environment.

So, it provides multiple sensor modalities, which can be visual, auditory or, haptic.
Haptic means what? By touch, through touch, basically, you have different sensors
which are added to the eye, to the ears, to the fingers, which can give the user an
improved feeling. And these users could be any user and in the industrial settings in the
IIoT settings we are talk talking about industry users typically. Let us say, a mechanic of
a particular machine, an operator of a particular machine can be trained with AR, with
the help of these different types of sensors, that I just mentioned visual, auditory, haptic
sensors can be attached to that particular user or the mechanic or the operator of an
industrial machine. And, different computer generated images can be shown and that
basically will give a better perception of the physical environment in which that
particular mechanic the operator is operating.
So, they will have improved augmented reality of the actual reality. So, that is going to
be generated out of it. So, this is the whole advantage of the use of AR, in industry settings.

So, let us now go through the history a bit. So, AR has become very popular in the recent
times, but it has been there since long starting from early 1900s. So, in 1901 the idea of
electronic displays and spectacles came into being Frank Baum basically proposed this
kind of idea use of this kind of thing. In 1980s, he published heads-up display for
teaching real world five right skills we are done this publication of public publications of
this kind of heads-up displays was made available by Lintern.

In the 1990s, Boeing researcher, Thamos Caudell coined the term augmented reality. In
1992, Louise Rosenberg was the first, he came up with the first functioning AR augmented
reality system. In 1999, the US Naval Research laboratory came up with the BARS system,
which is the battlefield augmented reality system. In 2013 Google came up with the
Google glass, which is very popular and everybody knows about it at present.

So, this is basically the overall history of augmented reality. So, it has been there since
the last century 1900, early 1900, till recently augmented reality in different forms was
there, but now it has become much more popular, because of its use in different industry
settings. So, different applications of augmented reality in tourism, entertainment, retail,
industrial design, medical environments, hospital settings, for instance, flight training,
educational, military training and so, on augmented reality; reality has different
applications.

So, here in this particular picture, there are pictorially, the different augmented reality
systems are shown. So, on the left hand side we have the augmented reality eyeglasses.
This is the augmented reality eyeglass, this is the heads up display as you can see over here
the individual is wearing it, the display unit heads-up display unit and then we also have
like medical applications for instance as you can see over here the doctors are performing
some kind of a heart surgery and can see using the display, the different aspects of the heart
on upon which the surgery is being performed. So, these are some of the key components
of these different devices, that are used for AR, there are sensors and cameras, projection
screen, processing unit, reflection systems like mirrors etcetera.

So, these are some of the different components that are used for AR. So, let us do one
thing, we will now try to some of the key features of how these AR systems work. So, what
is important? We have these sensors.
We have sensors; these sensors basically are the ones that gather real world information,
then we have also the cameras and these cameras they scan the environment, they collect
the data from the environment, from the surroundings. Then we have the projectors,
basically, they project the information about the environment to somewhere and then we
have the mirrors this is basically the reflection system. So, these are some of the different
components of AR and how together the AR systems work. And these mirrors are very
important because they can change the orientation based on how the users’ eye changes
its position.
So, these are the key aspects of AR and how it works. So, sensors collecting data,
cameras also basically know projecting in the information collecting the information and
then together the sensors, cameras, collecting this information and then sending it to
theprojectors, and this projector basically projects all this information and then you
have this reflection system, which is the mirror. So, this is how the AR systems work
holistically. But as I said at the outset that through this half an hour lecture you cannot
become an expert of AR or VR, this is just to get a brief expository idea about how the
systems work and what is in there.

So, now let us look at the different types of augmented reality. We have marker-based
augmented reality, marker-less augmented reality, projection based augmented reality,
superimposition based augmented reality these are some of these different types of
augmented reality.

Marker-based augmented reality gives an outcome when the reader is sensed by the camera
and the visual marker. So, there are two things one is the camera and the existence
of marker. So, these markers basically will help in recognizing simple distinctive patterns,
which can be easily processed and the camera is well understood what is the functioning
of this camera.

The marker-less augmented reality unlike the previous one does not use any kind of marker
and these commonly used things like GPS, digital compass or accelerometer, which
basically help in offering information such as the location of a person or a device.
Then we have the prediction based augmented reality, that gives an outcome by
projecting light onto the real world surfaces and it allows the human interaction by sending
light. And it differentiates between the expected position projection and the altered
projection expected and the altered.
So, this is these are very important things these are very important things in expectation
and altered projection, this differentiation is very important and is key to AR. Then we
have the superimposition based augmented reality, which partially or, fully substitutes the
original view of the object with the augmented view. Here object recognition basically
plays an important role and the application cannot replace the original view with the
augmented one in this kind of superimposition based augmented reality

So, these are some of the highlights of AR and how it works at a very high level, what
are the different components of it.

Now, let us look at what is VR and later on, after we have gone through the different
highlights of VR, we will see that how AR and VR they compare between each other, what
are the similarities between AR and VR and what are the differences. So, VR is a mixture
of interactive hardware- and software-based artificial environment.

So, it is an interactive hardware unlike the previous one, here it is interactive and it offers
a three dimensional image. A realistic three-dimensional image is created and the user in
VR feels that the user is actually present in the physical world, which is being simulated,
but is actually not being present. So, it is a virtual environment that is immolated, but the
user feels that, the user is acting in the actual physical environment. So, it is a virtual
immersion kind of experience that the user [Link], this information to the user is presented
in such a way that the user feels that the user is operating is acting in the real world or
physical world.

So, the key features of VR are, that it creates and enhances an imaginary reality, this is
very important. It is something that is not real in fact, but is a virtual reality, imaginary
reality. Then we have it gives the perception of being physically present in a non
physical world, this is also very important. It is actually not a physical world, it is an
emulated one, a virtual one, but the user feels that user is interacting with the physical
world, which is being immulated.

So, it incorporates auditory and visual sensory feedback all kinds of sensors are used in
this case also auditory, visual, touch, haptic interfaces are also used. So, user can hold
devices and feel that actually the user is holding the device; for example, a user might
feel that the user is holding the steering wheel of a car, but the actually its a virtual
environment and in immulated environment, there are certain sensors which will give the
feeling to the user that the user is holding a particular steering wheel of the car, but actually
the user is not feeling. I mean not actually holding the user is not actually holding
the steering wheel and he is not driving the car.

So, I hope that this basically makes it clearer to you how it is going to work, not I mean
not the technical details of it, but an overall impression about how VR works. So,
itallows the users to get naturally absorbed into the virtual environment. So, this is also
very important in VR.

So, in terms of the chronological order of virtual reality, how it has evolved over the years.
So, in around 1960s early 1960s the filmmaker Morton Heilig he created an experience
theater. So that basically started the works, the initial works on virtual reality, then the
Sword of Damocles was created by an MIT computer scientist in 1968 and he is considered
sort of like a father of virtual reality displays, then in 1987, Lanier and American
philosopher and scientist, coined the term virtual reality.

So, actually the virtual reality the popularity of virtual reality started with this term in
1987, the virtual reality term started in the 1987 and then David EM was the first artist,
who died in 1984, he produced the navigable virtual worlds. And Eric Howlett, in 1979
created the leap system, which is the large expense, extra perspective, optical system,
and in 2015, Google created the first ever resident artist system and that is a quite
popular system, it is a virtual reality system. So, the Google glass we have talked about
earlier in the context of AR and Google’s in this system the resident artist system is another
system which is an example of the virtual reality system. These are quite popular systems
in terms of AR and VR being used in our society.

So, there are different applications of VR in terms of used by real estate agents,
entertainment industries, robotics, astronauts; astronauts use VR a lot because even
before and a space craft is sent to the in the space, the astronauts are trained in the virtual
environments, using VR systems, the astronauts are trained similarly for the flight pilots,
as well. Fight pilots are also often trained in the using virtual reality systems. And in health
and safety healthcare industries VR are used. So, doctors are trained before actually
performing a life critical surgery, the doctor can try to practice that surgery in the VR
environment. In educational environments it is used by 3D artists, VR are used for mass
surveillance also VR has found applications.

So, before going any further, let me just try to show you how overall how these VR
systems are going to operate.

So, here also we have these different sensors here also like before we have different
sensors. These sensors can detect the motion the direction in space, motion, space, the
position in space, all these things can be detected using these sensors and there could be
different types of sensors that would be used. So, then you also have different lenses, which
could be used to capture and reshape the image of each eye. Then we have the processing
units, which basically will process the data that is captured, then we have finally, the
display screens, which basically give the user the feeling of being used in the physical
world.

So, these displays are very important. So, these are the overall the different components
of a VR system some of the different components, and these sensors are very crucial
different types of sensors are used. These sensors, the display system, these are very
important and of course, the lenses the processing units processing units are very important
and you need to have high end processing units high end processing units are typically used
in these VR systems. Because otherwise that feeling will not come if there is a delay
between performing a certain action and actually getting the experience the perception if
there is a delay, then this VR system is not going to be much useful.

So, I was telling you about the different applications of virtual reality earlier. So, virtual
reality, here we can see few pictures, this is a VR headset that you can see. Here as you
can see over here virtual reality sets are being used by a military person to get trained
with some combat operation or something very similar. So, there are different
components of these headsets are very important. So, these are very important and there
are different components of these headsets like different types of sensors, like
magnetometers, accelerometer, gyroscope, the different lenses, display screens
processing unit, all these as I was telling you earlier these are the important components
of a VR system.

Different types of virtual reality; one is non-immersive simulation, semi-immersive

simulation, fully-immersive simulation these are different types of virtual reality
environments.

Non-immersive, as these name suggests, the user enters into the virtual environment
through a portal or, window and this basically allows the users to have a peripheral
awareness of the reality outside the virtual reality simulations. So, this name, as it
suggests, non-immersive simulation, basically, utilize only a subset of the user senses notall
of it as a non-immersive. But only a subset of the user senses are used in this kind of
simulated platforms, non-immersive simulated platforms.

Semi-immersive again partly it is not like the non-immersive, but here partly partially or
fully-immersive experience, is gathered by the users, through these different sensors and
the users use of these different sensors and the senses sensing up through these different
sensors, this is what is done in this non semi-immersive simulations. And here basically
these are powered by high performance graphical computing systems, coupled with large
screen projectors thse are some of the things that are used for semi-immersive
platforms.

Fully-immersive is high end basically. Here you have complete immersive experience;
complete realistic experience is offered through some simulations to the users, and these
simulations offer a delivery f a wide field of view, using different head-mounted displays,
motion detecting devices.

So, let us now having understood AR and VR so far, let us now try to understand the
similarities between AR and VR. So, there are many different similarities they are very
similar. So, both of these they create new artificial world for the users, both of
thesetechnologies can help create this artificial world. These serve the user with enhanced
experiences that is also very similar, there is great prospect in the field of each of these AR
and VR in fields like medical science, in astronomy, in flight simulations. There is lot
of similarity I mean there is lot of use of AR and VR technologies in these different types
of application domains.
So, this is how these technologies compare augmented reality and virtual reality.
Augmented reality adds digital elements to the actual environment. It delivers virtual
elements as an encrust of the real world and virtual reality it offers immersive
application. It is unlike in the case of augmented reality, what is happening is the
physical world is augmented with certain high end ICT features, so that the user gets
some kind of improved experience, in improved perception of the reality.

And in virtual reality on the other hand complete or partly partial or complete immersive
experience is obtained by the user. So, user feels that the user is there and he is operating
completely immersed or partially immersed in the physical environment, but actually it is
not a physical environment, it is a virtual environment, in which the user is operating, but
the user feels that the user is operating in the real physical world, which is not correct which
is not the real one, but a simulated one.
So, it offers a digital recreation of the real-life setting, which one the virtual reality
platforms.
So, here are some of the references that one can go through. So, here I have given you
some of the web references, but there are many books on AR and VR, there are many
research papers on AR and VR, but I think for this course that, going through in detail of
each of these technologies is necessary, it is not easy to understand each of them in detail
unless you want to really get an in-depth understanding though about AR and VR.

But then again this will not be right course for you to get the complete understand about
AR and VR, you need to register for some other online course on AR and VR. And so, this
understanding about the different features of AR and VR systems, the different advantages
the different aspects of it, this is sufficient for the industry persons, the learners, to know
about how these things can be used in an industry setting to create an IIoT platform in their
respective industries. So, with this we will come to an end.

.
 Introduction to Industry 4.0 and Industrial Internet of Things
Industry 4.0: Artificial Intelligence

In this lecture, we are going to go through some of the basic concepts of Artificial
Intelligence in Industry 4.0. So, at present all of you must have heard about the hype of AI:
Artificial Intelligence and its use in the industries and everywhere else, in fact, AI has
become very attractive in the present times and it finds different applications in different
domains industries inclusive. So, towards fulfilling the objectives of an Industry
4.0, it is envisaged that artificial intelligence will play a big role.

So, let us try to understand the basic concepts behind AI and how it can be used to improve
the overall efficiency and address different challenging issues in the industrial sector. So,
when we talk about artificial intelligence, what comes to our mind, it is some form of
intelligence, we are talking about an artificial form of it. You must have heard about
different things like a robot playing soccer, robo-soccer, robot playing soccer. Then
you must have heard about the driverless cars, which has also become very popular in the
last few years, the driverless cars.

These are all examples of use of AI techniques to solve different challenging problems,
which otherwise are difficult to solve. But, it is not just robot playing soccer, it is not just
the driverless cars where, AI has found application. AI finds the application in different
domains in for credit card fraud detection AI could be used for designing a computer,
which can play the game of chess. And, this is something that has happened since last
few decades; a lot of people have taken a lot of interest in the applications of AI in
games.

In the industries also for making the industrial processes much more efficient, to solve
different problems, which manually was difficult to be solved AI and different other
applications have found interest in the industries.

So, what is AI? Artificial Intelligence so, there are different viewpoints of what AI is. So,
it is quite broadly scoped and there are multiple definitions to describe what artificial
intelligence is, as per one of the definitions by Patterson; AI is a branch of computer science
that deals with the study and the creation of computer systems that exhibit some form of
intelligence. So, we are talking about computer systems exhibiting some form of
intelligence which is very similar to the natural intelligence of human beings. So, striving
to build systems which can try to intelligently think and behave like human beings is
what one of the definitions of artificial intelligence says.
Another viewpoint as I told you that there are many different viewpoints of what AI is;
another viewpoint is how we can use how we can use the different computational models
to improve the mental faculties of humans is what again AI can do. So, that is another
scope of AI. So, like this there are different viewpoints of what AI is. So, the lateral
definition is basically proposed by Eugene Charniak and Drew McDermott

So, we are talking about artificial, which is manmade and integrating with the
intelligence, which is basically the thinking power and together achieving a system, the
development of the system an AI based system Artificial Intelligence system, which is a
creation of man-made thinking power. So, in simplistic form AI is a creation of software,
having intuitive decision making capability. So, remember one thing that AI systems are
typically software-based, but in industrial sector or, many other domains they these AI-
based software will have to work on some kind of hardware. So, you cannot, basically,
leave hardware completely separated when you are talking about AI systems [Link]
you look at the origin of AI, AI in the recent times have found lot of popularity in our
country and globally, but AI has been there since long starting from the 1950s AI has
been in existence. Then in, earlier there used to be different theoretical research works on
AI. There have been different languages that have been proposed for use with AI like Lisp,
PROLOG. Then, in the 1970s the concept of expert systems came into being. Expert
systems are basically the ones where based on certain pre-existing knowledge the systems
are going to perform better in the future.

So, expert systems came into being, then in the 1990s, somewhere in the middle of
1990s, if you recall the Deep Blue became very popular. IBM came up with their computer,
the Deep Blue, which is a AI based system, which can play chess. So, if you recall its
history now, that Deep Blue, basically defeated one of the greatest chess players long time
back. So, the computer, that was when an artificial system was able to supersede
the brain of a human being and an expert brain, an expert chess player was defeated by
Deep Blue in 1990s and that is when the applications of AI became popular in the domain
of games and chess, particularly.

So, what is the difference between a program, which uses AI. So, a computer program
without AI uses large databases and algorithmic search method, whereas, a computer
program with AI uses large knowledge base and heuristic search method. So, it is
algorithmic search versus heuristic-search method.
Algorithmic search basically follows certain well-defined procedures in order to come up
with some optimized solution whereas, heuristic search as this name suggests; heuristic
search is popular in AI. Because, many of these AI-based problems are not easy to solve.
The search space is huge, many of the AI problems is huge. So, where you cannot come
up with some efficient solution using the traditional methods of search like the
algorithmic search methods.

So, heuristic search basically talks about heuristics or rules of thumb being used to come
up with solutions which will be good enough to solve certain problems at a certain point
of time. So, these are the heuristic search methods; heuristics search methods are quite
popular search methods in the domain of AI. And, heuristic search methods have found
lot of use in the applications of AI in games in different games chess inclusive.

So, actually what happens is why the AI is why the search method heuristic search methods
are very popular in AI is, because think about computer games, things about chess the game
of chess. So, sometimes the input space is so, broad and if you do not use you cannot use
the traditional search methods. So, you have to use some heuristic methods to come
up with certain solutions, which will perform superior at certain point of time under
certain conditions. So, heuristic search methods are quite popular in AI and AI for
[Link], AI is heavily based on the concept of knowledge. So, we are talking about
several pieces of data can lead to information and then information can build help in
building knowledge. So, knowledge is that information that can be used to perform a
particular task. So, there are different forms of knowledge particularly in the context of
AI. There is procedural or, operational knowledge, which basically talks about the
procedures that will have to be adopted in order to; the knowledge about certain
procedures, that will have to be adopted in order to come up with a particular problem.

So for example, a quadratic expression or quadratic equation. So, the procedures that will
have to be followed in order to solve it that is procedural knowledge, declarative or,
relational knowledge, this is also known as descriptive knowledge. So, here basically we
are talking about the knowledge, the description of a certain thing, an object an event or
something alike the knowledge about it; storing such kind of knowledge is important. And,
that kind of knowledge will be used whether it is procedural, declarative or, heuristic, that
kind of knowledge is going to be used to make the next move or, to improve the
processes in the industries. The way things are happening based on certain existing
knowledge you can try to improve upon the processes in the future and so on.
So, that is where this concept of knowledge and AI comes in applicable in the context of
industries and industrial processes. Heuristic knowledge is basically these different
heuristics that will or the rules of thumb that will have to be used in order to address certain
problems, certain challenges at different points of time that is heuristic knowledge.

Now, let us talk about the scope of AI. Artificial Intelligence has found use in diverse
fields. In games we have already talked about, in chess game, in soccer games, in
different other games, AI has found applications. In theory improving, NLP is basically
Natural Language Processing. So, AI in NLP means what that you are trying to come up
with some rules, which can make the computer understand not the computers language, but
the way the natural languages like English, French, with which humans are conversant
to communicate with one another.

So, that is NLP, then AI has also found use in vision computer vision, speech processing.
Computer vision is basically trying to enable a computer to see around, to see around, to
feel what is around it and so on. Speech processing, on the other hand, is enabling a
computer to understand, the way humans speak, what the humans are speaking. Not just
the text, but the speech of the human beings can be understood typically in real-time or,
may not be real-time as well. So, if you are speaking in English, the speech processing
would help the computers to understand what the humans are talking about in their own
language.

Then comes robotics AI in robotics, robot performing different actions, robot making
different moves, in a particular terrain, performing different moves, taking different
trajectories. So, the in AI in robotics is very important and in expert systems. Building
expert systems, which are basically in knowledge intensive systems, based on
theknowledge base, that is resident in these systems, these systems can provide expert
advice; expert advice to users about what should be done next or what should be done in
the future.
So, artificial intelligence is very important, it has become very popular. Artificial
Intelligence, ML: Machine Learning, Deep Learning these things have become very
popular in the recent years. So, Artificial Intelligence, ML, DL, these are linked to each
other. All the people are talking about machine learning, deep learning, artificial
intelligence, but there is a linkage and this is this figure in front of you shows what is the
scope of each of these.

Machine learning or deep learning are basically branches of artificial intelligence, but
then they are in artificial intelligence beyond learning there are different issues. Issues of
search, classification, these that it sends means there are a lot of, lot of different issues
are there beyond learning in AI. So, this is how these three recently popular technologies
have become or what is what is what is there scope. So, that is what is defined.

So, what is machine learning? So, machine learning is a part of AI which empowers the
machines to make decisions based on their experience rather than being explicitly
programmed. So, in the computer basic computer programming or fundamental
computer courses in any [Link] program or whatever; what the students are taught
typically is to make to explicitly program certain steps, which the computer are going to
take in order to solve a problem. In machine learning, we are talking about some kind of
a software or a program, which based on the experience, previous experience, past
experience will take actions, better actions in the future. So, with time better and better
actions based on the past experience will be taken.

Deep learning, basically, is a subset of machine learning, which can learn automatically by
finding the features of the object on its own. So, what is so, deep learning is basically where
some deep computational structures are used to come up with some efficient algorithms,
which can do certain things much more efficiently with grater greater accuracy; efficiency
in terms of accuracy, typically. So, better accuracy in a better manner will be done
things will be done by deep learning.
So, it is not like deep learning is the only solution, deep learning has its own application
domains. There are other non-deep learning, the traditional machine learning schemes,
which are also advantageous in certain contexts. It is not the deep learning is the solution
in machine learning, it is not like that. So, you can use deep learning or non-deep
learning methods to solve certain learning problems depending on the certain requirements
that are there.

Now, let me talk about an interesting thing. So, we need to understand the, the difference
between the different AI techniques. So, AI we have talked about machine learning,
there is also something called Symbolic Learning. So whereas, machine learning is based
on data; symbolic learning is symbol based.

So, applications of this thing good applications would be in pattern recognition, machine
learning has lot of applications in pattern recognition, data mining, and here symbolic
learning has lot of applications in for example, image processing, image processing. So,
symbolic learning, basically, we are talking about applications in computer vision, then,
robotics and this can be machine learning, can be of different types; we have already talked
about deep learning.
But, it could be also the traditional statistical learning. So, statistical learning has
applications in natural language processing, has applications in speech recognition. Deep
learning can be of different types: convolutional neural network is one, a recurrent neural
networks is another. Each of these can be used to address again different problems of
computer, vision, and also object recognition.

So, this is basically the scope of artificial intelligence and the different forms of not the
scope of artificial intelligence, entirely, but in the context of learning.

Let us now go back and talk about, we have understood more or less the scope of AI, the
different applications of it, its importance in the context of industries and different other
application domains like games, computer vision, robotics, but let us now try to understand
in little bit further depth. So, that we will be able to realize that why AI is popular in
building systems, which can help achieve the objectives of Industry 4.0.

So, in Industry 4.0 we have talked a lot in the previous lectures about Industry 4.0 and its
scope, but what is Industry 4.0? In Industry 4.0 we are talking about few things, first of
all human machine interaction. This is very critical, humans and machines interacting.
We also in Industry 4.0, we are also talking about in certain cases machine to machine
communication. One machine directly talking to another machine without any human
intervention, machine to machine communication. Cyber physical systems we have
talked a lot in a previous lecture also associated technologies such as cloud computing,
cognitive computing, IoT or IIoT; all of these can help in making manufacturing
processes better.

So, that is Industry 4.0. Now, smart factory smart factory is another one which is talked a
lot in the context of Industry 4.0, smart factory. So, basically this smart factory what it does
is these factory, machines in the factories, the physical objects that are there in the factory,
which interact with one another. The virtual instances of them would be created and those
virtual instances would be made to talk to one another. This physical worldand the
virtual world, instantiation of the physical into the virtual space and making the physical
objects and consequently the virtual ones talk to one another is what is done in the context
of smart factories in Industry 4.0.

So; obviously, things like machine safety, efficient products lifecycle, efficient
manufacturing processes, these could be achieved with the help of AI. And, these are
required for making smart factories which in turn will help achieve in building Industry
So, IIoT, IoT in general and Industrial IoT, these are the ones, which essentially will help
achieve the objectives of Industry 4.0. AI can be used in along with IoT, along with IIoT
and also to transform these systems into efficient systems; IoT and IIoT systems efficient
using AI. So, basically, use of AI helps the machines and equipments to communicate
and relay information with one another. And, this thing can be used in different industry
sectors, finance, retail, healthcare, agriculture. And, you name it and it is possible to use
AI in order to help these machines and these equipments communicate and relay
information with one another much more efficiently.

For this specifically you could use computer vision, robotics, NLP, ML, DL, RL all of
these things that we have talked about so far you could use them. So, with the help of AI
in industries, in the industries are capable of taking the advantage of large amount of data
that is generated by the machines to do something better. So, one aspect is to
improveupon the communication, automation, relay, etcetera, etcetera using AI. The other
one is that once you have used IoT and IIoT, in the industries; what is happening is
these sensors actuators, etcetera from these IoT devices are going to throw in lot of data.

So, that data we will have to be analyzed and one can take advantage of the past data.
The experience that is generated from this previous historical data to make things much
more efficient, to make processes efficient in the future. Examples would be for
prediction of yield, quality of yield, prediction of the quality of yield, in manufacturing.
So, these are different examples of use of AI. AI for improving the automation,
communication, delay of information etcetera and for improving the prediction processes
in terms of quantity of yield, quality of yield etcetera in the manufacturing industries.

So, there are different challenges of AI in IIoT. Connecting different devices,

understanding the data, training the data, making it accessible, making the machines or
whatever actionable these are all different challenges of use of AI in IIoT. So, I do not need
to explain each of these, but only thing that I would like to highlight is the training. Training
is very important because, you are using past data, which you will be using to train the
machines to do something better in the future that is one type of learning in fact, in AI that
is one type of learning. So, there are different types of learning. This is one type of
learning that using the training data set.
So, there are different advantages of AI in IIoT. The usefulness of AI in the industry
scale are to increase the efficiency, save costs, improve security, augment performance and
boost up resources in the industries; resources of all kind as I said in a previous context in
a previous lecture before. So, boosting up all kinds of resources, all kinds of tangible, non-
tangible human resources. So, all kinds of resources could be boosted up, with the help of
use of AI in the industrial [Link], AI has found use in different industries. In the
agriculture industry, AI could be used for crop and soil monitoring, for precision
agriculture. Precision agriculture means like coming up with precise predictions about
certain things, precise predictions about certain things. For example, when do you need in
the agricultural field, when do you need precisely to irrigate the field, to put fertilizers in
the field, and exactly the area, where the fertilizers will have to be applied.

What fertilizers exactly would be required, not that you put in any kind of fertilizer it
will be. So, all these predictions, precisions, in agriculture could be achieved with the
help of use of AI. Supply chain efficiency also can be improved in supply chain in the
context of agriculture and food could also be made much more efficient with the help of
AI.
So, that is the AI in agriculture, AI could be used in education industry, to improve the
student retention rate for making interesting study programs and for providing immersive
technology into the classroom. So, already there are different existing systems which use
AI in the education sector. So for example, the smart learning systems by Carnegie
Learning, then Querium Corporation AI based education system. Like this, there are
many different types of AI based systems in education sector that are available for use.

So, in the manufacturing factors sector, manufacturing industries AI could be used to

improve machines power consumption, detection of machinery fault, maintaining
product supply by predicting consumer demand. So, these are the different applications
of AI in manufacturing.

AI in the aerospace industry extracting useful data from every day of flight, improving
productivity of manufacturing processes. All these top aircraft manufacturers like
Boeing, Airbus, they use AI heavily in their processes.
So, in the case of rail transportation or automotive transportation, etcetera AI is also
used, self-driving car is something that I mentioned at the outset. Like this assisting drivers
to prevent accidents these are all different applications; these are some of the different
applications of use of AI in transportation industry.

So, one last thing before we get into the references, one last thing that I would like to
highlight about machine learning is that machine learning can help do something else as
well. So, if you are talking about something X let us say some parameter advertisement
let us say and let us say that this is the sale, the Y-axis. So, you can have different data
which could be plotted and then you can have some kind of a regression fitting or some
correlation study etcetera. So, if you are talking about this kind of thing, the humans are
very good in doing things because it is a 2-dimensional thing and we know different
ways of curve fitting. So, handling correlation regression in 2-dimension.

Think about the same problem, if you are increasing the number of dimensions. So, if
you are increasing the number of dimensions and doing something very similar using the
human expertise and human eye, human intelligence, human brain, that is difficult. So,
the computers can do it and in fact, more specifically machine learning can help you
achieve these different types of correlation, curve fitting, in multiple spaces having
multiple dimensions, at the same time. So, this is also another application of machine
learning and how humans and machines can perform better than humans, in these contexts.

So, let us now, go back and look at these things. So, these are some of these references that
are there. So, the references for AI, ML, DL, its huge these are some of the ones that,
if you are interested to know little bit more, in depth, you could, but mind you that through
this course you cannot become an expert of artificial intelligence. Artificial intelligence are
semester long courses themselves and all you need to do in this course is just to get yourself
exposed to the different concepts and which is what I have done. But,
hen if you need to actually implement AI-based techniques to solve certain problems,
you have to take separate semester long courses in AI, ML, DL. So, with this we come to
an end.
.
 Introduction to Industry 4.0 and Industrial Internet of Things
Industry 4.0: Big Data and Advanced Analysis

In Industry 4.0, we are talking about use of IoT, IIoT, Industrial IoT, which essentially
are heavily based on the use of different sensors and actuators. So, these sensors
typically sense lot of different data and this data are sent continuously after they are
being sensed. So, this data will have to be processed. So, what kind of data we are
talking about? So, this kind of data in the context of IoT, IIoT, Industry 4.0, exhibit the
behavior of something known as big data. So, we have to analyze the big and the traditional
forms of data, to gather meaning out of it.

So, when we talk about this analysis whatever we talked in the context of machine learning
in AI those kind of methodologies are also applicable over here. So, analytics, statistical
analytics, machine learning-based analytics, use of different methods of neural networks,
SVM, all of these could be used to analyze the data, in addition to the statistical
methods, the multivariate statistical methods that we are all familiar with. So, those could
also be used to analyze the data. So, what is important is that you try to gain meaning out
of the data that you receive from these IoT devices in IIoT.

So, what is big data? So, data which is too big to be handled by processing tools and
conventional databases, conventional processing tools. So, big data are typically the ones
which exhibit the properties of non-structured data, but they could also have structure data.
Structure data means what? Structure data are the ones, which could be stored in the
form of tables in different databases; for example, relational tables. So, for example,
databases like MySQL, Oracle, they use relational tables.

So, you can store those data in these relational tables in the form of these tables you can
store this data so that is structured way of storing the data. And not that all kinds of data
could be stored in the form of tables. So, certain data for example, Web blogs, Facebook
chats, images, the newspaper blogs. So, all of these will exhibit non-structured behavior
of data and they cannot be used, they cannot be stored in relational tables in the
traditional form. So, these are like big data, non-structured as well as structured data, which
will have to be handled in certain ways.

So, let us look at some of the key attributes of big data. We are talking about representation
of data, which acquisition speed the data volume, data characterization, restricts the
capacity of using conventional associative methods to manage successfully; the analysis or
the data, which can be successfully operated with important horizontal zoom technologies.
This is as per one of these definitions of big data.
So, I was telling you that in the context of IoT, IIoT, we will get both structured data as
well as unstructured data. So, structured data as I was mentioning before are the ones,
which can be stored easily in an organized fashion in typically relational tables. So, if
you are able to store this data in the form of tables, so you can use a language like SQL,
Structured Query Language to query these tables to access the data, that are stored in
[Link], if you are talking about unstructured data you cannot store this data in
the form of RDBMS tables and there is no predefined data model that could be used to
analyze this data. So, basically most of the data that we encountered in the present-day
world, would be unstructured. So, you cannot use the traditional database techniques that
you are already familiar with in order to analyze this kind of data. And at the same time
not only that these data are unstructured, but also this data come in huge volumes, so you
are talking about huge datasets that will have to be stored in certain way.

So, these are some of these characteristics of big data we are talking about volume,
variety and velocity. So, these are the main 3 V’s of defining big data. Data which is
large in volume which exhibits variety and which arrives at high velocities at the processing
server or processing agent. These kind of data traditionally were termed as big data, but
as researchers what with big data more and more they also included few more different
other aspects; other aspects other V’s in fact. So, other V’s like variability,
visualization, value, veracity, so these are 4 additional V’s that will also characterize big
data. So, traditional 3 V’s plus the 4 V’s, 7 V’s would be used to characterize big data.

So, let us try to understand these in little bit more detail. So, let us talk about the 3 V’s first
volume, so volume, velocity, variety.

So, these 3 V’s let us consider. So, what is this volume? This volume is large volumes of
data we are topically talking about. For example, some 72 hours of video on some server
such as let us say YouTube; let us just take for example, it could be anything. This is
huge kind of data and huge in volume. Then if you are talking about velocity, if you talk
about twitter for example, or similar kind of other platforms we are talking about some
100s of millions of tweets, on average per day. So, as you can understand that handling
this kind of velocity of data is very important.

And then variety and this variety could be you can have both structured as well as non-
structured data. So, variety of data for example, image data, text data, then video data, all
coming together at the same time, that is variety. So, in big data context, we are typically
talking about data, which is coming you can think of it as a pipe, some kind of a pipe
through which data are coming continuously, huge volumes of data are coming at high
velocities and this data the composition of this data are all very varied, text, image video
and differ different types. So, all these data coming at the same time, handling this kind
of data, using conventional data warehousing, database techniques, etcetera, it is not
possible. So, people are talking about how you can have newer methodologies, newer tools
in order to address this kind of data. So, let us just go back to what we were discussing in
the slides.
So, we are talking about huge volumes of data, we are talking about data not just in
gigabytes and terabytes, we are talking about petabytes, hexabytes and so on of data,
how do you store, that kind of data that is very important.

Velocity, speed of generation, 100s of millions of tweets per day on average that is huge
velocity of data coming in, data getting generated. So, if the data is getting generated you
have to handle the data and this is a high velocity data that is coming in and that has to
be [Link] the other hand, stock exchanges like New York stock exchange. So, they
handle large volumes of data coming in high speeds. So, many exchanges, financial
exchanges are carried on in the stock exchanges, so the speed is also very important of this
kind of data.

Variety, we are talking about different varieties of data--text, image, audio, video
etcetera, etcetera all different types of data without any restriction about the data
[Link], so here we are talking about it is different from variety. Here we are
talking about the data whose meaning is constantly changing. So, example would be
language processing, hash tags, geo-spatial data, multimedia data, sensor data, etcetera.

Veracity indicates the biasness in the data and it talks about the unusualness and noise in
the data. So, it is important in programs, which involve automated decision-making.
Veracity deals with the understandability of the data and not just the quality of the data,
understanding the data is important in this thing; understanding the data is important in
velocity.
Visualization: visualization we are talking about in what form the data will have to be
visualized, in graphical form, in textual form, what are the different patterns that can be
visually identified. So, all these different character characteristics of this data the
visualization of the data is important. And value, which is basically extracting some
business information from the scattered data. So, this is basically the value attribute of
big data. So, what are the sources of data? Enterprise data can be generated from online
trading, data analysis, production inventory data, sales and different other financial data.
These are some of the examples of enterprise data. IoT data like industrial data, healthcare
data, agricultural data, and so on.

Biomedical data, data that are generated from gene sequencing, from medical clinics.
And other data such as computational biology data, astronomy data, nuclear research
data, these are the different sources of data. Astronomy, so some of these telescopes have
been planted in certain parts of the world to look at the sky continuously, round the clock
365 days, a year, these are scanning the sky. And so as you can understand continuously
in the, so much of data are coming from these telescopes. So, these kind of data will have
to be handled; these are data which come in huge volumes. So, these are some of the
sources of data.

But in the context of industries similarly these machines, cyber physical systems
machines continuously do by virtue of their continuous operation interaction and so on they
are generating data, which typically is a combination of structured and unstructured or non-
structured data. And these will have to be handled using these different analytical
techniques for big data, data collection. So, you have to collect the data. So, the sources
could be from log files, record files, from sensors, from different other sources like RFID
devices and these could be coming from different sources.

Data transmission is then important. We are talking about data inter DCN transmission and
intra DCN transmission. What is this DCN? DCN is basically data center network. Within
a data center network of data center; that means, different servers interconnected with each
other in a data center within that the transmission of the data and then from one data
center to another data center, inter-data center network transmission, that is another type
of transmission.
Then the data will have to be processed. But before that it has to be pre-processed. The
data will have to be pre-processed to remove noise redundancy, inconsistency. And this
pre-processing of relational data mainly follows integration, clearing, and redundancy
mitigation.

Integration is basically combining the data from various sources and delivering the users
a constant data value. Clearing is spot, if there is something that is spotted to be
incorrect, insufficient or uncooperative, then that kind of data will have to be spotted,
corrected or they have to be removed. Redundancy mitigation--if there is any redundancy
in the data there has to be filters in place which will in remove the repetition of such kind
of unnecessary data from being transmitted, because the bandwidth is also very limited.
So, you do not want to transmit, redundant data unnecessarily.

Data storage is important here we are talking about how the file systems would be for
storing the big data. So, GFS is a distributed file system that helps in supporting in the
storing, storage of large scale files. Another one is HDFS, HDFS is basically something
that is followed in the Hadoop technology, Hadoop system, which is quite popular for
use with storage of data, in the big data context. So, HDFS file system is a notable file
system derived from the open source course of GFS. So, GFS and HDFS our file systems
that are quite popular for use in different contexts of big data.

Databases, databases of different types non-traditional database wearing non-traditional

relational database, like NoSQL language is there; like SQL, SQL is for the traditional
relational databases use with querying this traditional relational databases and for non-
relational databases. NoSQL query language could be used.

So, why do you need data analytics? So, in the context of IoT, IIoT, Industry 4.0 in general
sensors that are used are very small in size, and they can be placed anywhere and these
data from the sensors can be transferred over some networks, typically, wireless in nature.
Because of this expression of data moving to systems from systems would be using these
different sensors and the data will be moving from one sensor to another.

So, some data are irrelevant for systems. So, how can one know which data are relevant
and which are not, so that is why this data will have to be analyzed. And also, there is
one more point to it that once you have analyzed, the data you can perform certain actuation
on the system, on the physical system, on the physical environment. So, those equations
can be performed, based on the results of analysis of the data that is obtained.
So, analytics is very important. For a customer, analytics will help the customer to learn
about competitors, learn about competitors, help the customer to join and activity, which
expands business. For the customer to protect from wrongful decisions, improve market
effectiveness, and picking appropriate product recommendations, making appropriate
product recommendations.

For example, if you are getting into Amazon or eBay or something you must have observed
or even like Flipkart, etcetera they use recommender systems a lot. So, recommender
systems basically what they do these are also AI based systems. They analyze the data
using certain techniques maybe, machine learning or whatever and they will be making
recommendations or suggestions to the user about what they could do next. So, these are
the recommender systems.
Big data analytics talks about a different way of handling data from the conventional
way, data are handled in databases and data warehouses. Big data apps cannot be fit in
traditional applications, cannot be fit big data applications cannot be fit in traditional data
warehouse architectures because here we are talking about large volumes of data,
Exadata, Teradata.

For Industry 4.0, we are talking about use of internet, industrial internet, which require
an approach to manage and process data coming from thousands of sensors
forperceptions. To manage and handle this huge data in health services manufacturing
other industries agriculture inclusive, is not new. For example, an event is detected by a
sensor and sent to the operational recorder and an operational recorder in it is a database,
which stores the data. And after that this data is optimized by querying the system about
something like, what about this hour’s production from the norm from the normal. So,
this kind of queries could be made from the data. So, that is where this big data analytics
could be used in the context of Industry 4.0.
So, IIoT can be recognized as a big benefactor or big data. It needs new technologies to
manage first data. Cloud can come to the help, because cloud can help in storing the data
and also processing the data, because there is as such virtually there is no limit on the
storage because if everything the storage is also obtained on demand. So, in IIoT for
helping in the management of the data technologies such as Hadoop, which is an open
source cloud based distributed data storage platform, cloud can be used for the analysis
of this kind of data, which is big data. So, Hadoop is quite popular in the context of big
data analytics.
So, cloud-based methods for analytics would be; so first of all, cloud could be used for
handling data big data, more specifically. So, there are different features of these on
demand self-service, cloud basically helps in getting through a self-service kind of
mechanism computational resources on-demand based on what the user requires. So, the
scalability is also much better.

So, this is basically in a cloud offers wide network access, it offers the capability of
grouping different methods, faster flexibility, and offering measured service. Measured
service means based on the units of usage of cloud, the user is going to be charged
accordingly.

So, there are different types of analytics--prescriptive analytics, predictive analytics, and
descriptive analytics. And prescriptive analytics talks about what is the best action,
should we try this and so on. Predictive analytics talks about what is next, what is the
pattern that is there. And descriptive analytics is when, where, etcetera what happened,
these are all descriptive analytics. Predictive, prescriptive, and descriptive analytics are
different forms of analytics.

So, these are some of the references on big data and analytics. So, if you are interested you
could go through any of these references, but there are huge number of references on big
data and analytics at present and you could go even beyond these lists of references that
are there for you.

So, with this we come to an end of analytics. Remember one thing that artificial
intelligence, analytics etcetera these are not fields which are separate and are different from
Industry 4.0 and IIoT. Without these enabling technologies, like the other technologies,
you cannot basically survive to build IoT systems, IIoT systems or to achieve the vision of
Industry 4.0. So, these are very important technologies for an analysis of the data, which
could be used for analysis of the data, in the big data context in Industry 4.0.