References

• R. B. Chase, Ravi Shankar, and F. R. Jacobs, Operations and Supply Chain

MANAGEMENT OF PRODUCTION Management, 14th ed. McGraw Hill Education (India) Private Limited, 2015.
• R. J. Tersine, Principles of Inventory and Materials management, 4th ed.
SYSTEMS Prentice-Hall International, 1994.
• F. R. Jacobs, W. L. Berry, D. Clay Whybark, and T. E. Vollmann, Manufacturing
Planning and Control for Supply Chain Management, 6th ed., McGraw Hill
Education (India), 2015
• J. A. Tomkins, J. [Link], Y. A. Bozer, and J. M. A. Tanchoco, Facility Planning, 4th
ed. Wiley India, 2013.
• E. L. Grant, and R. S. Leavenworth, Statistical Quality Control, 7th ed. McGraw-
Hill Education (Indian Edition), 2017.

Manufacturing Planning and control framework Manufacturing Planning and control

framework
• Key part of any ERP system
• Linking customer and supplier firms in a supply chain requires coordinating the
MPC activities between the firms.
• Three parts or phases
• The top third, or front end, is the set of activities and systems for overall direction
setting.
• Sales and operations planning balances the sales/marketing plans with available
production resources.
• The master production schedule (MPS) is the disaggregated version of the sales
and operations plan.
• The MPS must support the sales and operations plan.

Manufacturing Planning and control Manufacturing Planning and control

framework framework
• Resource planning determines the capacity necessary to produce the required • The supplier systems provide detailed information to the company suppliers.
products now and in the future. • In firms using MRP systems, execution of the detailed material and capacity plans
• The middle third, or engine of MPS framework encompasses the set of MPC involves detailed scheduling of machines and other work centers
systems for detailed material and capacity planning. • The MPC systems provide inputs to the financial, distribution, marketing, and
• The master production schedule feeds directly into the detailed material planning human resources systems that require the information.
module.
• MRP determines (explodes) the period-by-period (time-phased) plans for all
component parts and raw materials required to produce all the products in the
MPS.
• Bottom third is the execution systems

1
Operations and Supply Chain Management Operations and Supply Chain Management
• United Parcel Service (UPS), the U.S. • Operations Vs Supply chain
package delivery company.
• UPS serves an essential segment of the
supply chain: the part that moves goods • Operations refers to manufacturing and service processes that are used to
from one point to another. transform the resources employed by a firm into products desired by customers.
• Moving material in the supply chain is
referred to as logistics
• The company installed digital remote • Supply chain refers to processes that move information and material to and from
control so drivers can quickly lock and
unlock the truck’s overhead door. the manufacturing and service processes of the firm.
• The chief operating officer reported that • These include the logistics processes that physically move product and the
the new $70 million wireless system will warehousing and storage processes that position products for quick delivery to
save 1.75 seconds per stop, or 6.5 minutes
per driver per day, while reducing motion the customer.
and fatigue.

HISTORICAL DEVELOPMENT OF OPERATIONS AND SUPPLY CHAIN

Operations and Supply Chain Management MANAGEMENT
• Supply chain management’ wasn’t coined until the 1980s • The late 1970s and early 1980s saw the development of the manufacturing
• Supply chain management is the process of planning, implementing, and strategy paradigm
controlling the operations of the supply chain with the purpose to satisfy • Emphasized how manufacturing executives could use their factories’ capabilities
customer requirements as efficiently as possible. as strategic competitive weapons.
• It spans all movement and storage of raw materials, work-in-process inventory, • This requires trade-offs among such performance measures as low cost, high
and finished goods from point-of-origin to point-of-consumption quality, and high flexibility in designing and managing factories.
• The 1980s saw a revolution in the management philosophies and technologies -
JIT, TQM and lean manufacturing concepts.
• Service quality and productivity
• However, Henry Ford seems to have realized many of these philosophies about
60 years before – assembly line concept

HISTORICAL DEVELOPMENT OF OPERATIONS AND SUPPLY CHAIN HISTORICAL DEVELOPMENT OF OPERATIONS AND SUPPLY CHAIN
MANAGEMENT MANAGEMENT
• Focus on quality, Six sigma concept • The central idea of SCM is to apply a total system approach to managing the flow
of information, materials, services from raw material suppliers through factories
• Quality Gurus - W. Edwards Deming, Joseph M. Juran, and Philip Crosby. and warehouses to the end customer.
• 1990s – Business Process Reengineering - It is implemented by taking a fresh look • Trends such as outsourcing and mass customization helping companies to meet
at what the organization is trying to do in all its business processes, and then customer demand in a flexible manner.
eliminating non–value-added steps and computerizing the remaining ones to • The focus is on optimizing core activities to maximize the speed of response to
achieve the desired outcome. changes in customer expectations.
• Dramatic expansion of Six Sigma concept as an extensive set of diagnostic tools • E-commerce
was developed. • Growth of service sector - retail (franchises, ecommerce, Amazon, eBay),
• The term supply chain management become very popular in 1990s communication (telephones, Whatsapp), transportation (airlines, FedEx),
Financial (e-brokers), as well as information (Google).
• Business analytics - analysis of data to better solve business problems.

2
 Supply Chain - Introduction Supply Chain - Introduction
• A supply chain consists of all stages involved, directly or indirectly, in fulfilling a • These functions include, but are not limited to new product development,
customer request. marketing, operations, distribution, finance, and customer service.

• Supply chain includes not only the manufacturer and suppliers, but also Example: Customer at Wal-Mart retail store to purchase a detergent or a beauty
transporters, warehouses, retailers and even customers themselves. soap.

• Within in each organization, such as a manufacturer, the supply chain includes all
functions involved in receiving and filling a customer request.

Supply Chain - Introduction Supply Chain - Introduction

• The customer transfers funds to Wal-Mart. • Most supply chains are actually networks.
• Wal-Mart conveys point-of-sales data as well as replenishment orders to the • A typical supply chain may involve a variety of stages. These supply chain stages
warehouse or distributor. include:
Customers
Retailers
Wholesalers/distributors
Manufacturers
Component/raw material suppliers

Supply Chain - Introduction Supply Chain Stages

• Each stage in a supply chain is connected through the flow of information,
product, and funds.

• These flows often occur in both directions and may be managed by one of the
stages.

• Each stage of the supply chain performs different processes and interacts with
other stages of the supply chain.

3
Supply Chain - Stages Supply Chain Management
• Each stage in the figure need not be present in a supply chain. • Supply chain management involves the management of supply chain assets and
• Example: Dell supply chain – two different supply chain structure product, information and fund flows to maximize total supply chain surplus.
Through retailer
Direct selling (1996)

Supply Chain Management Supply Chain Management

• The objective of every supply chain should be to maximize the overall value • Supply chain profitability is the difference between the revenue generated from
generated. the customer and the overall cost across the supply chain.
• Value – Supply chain surplus – supply chain profitability • It is the total profit to be shared across all supply chain stages.
• Supply chain surplus or value is the difference between what the final product is • Supply chain success should be measured in terms of supply chain profitability
worth to the customer and the costs the supply chain incurs in filling the and not in terms of the profits at an individual stage.
customer’s request.

Decision phases in a supply chain Decision phases in a supply chain

• Successful supply chain management requires many decisions relating to the flow • Supply chain strategy or design decisions
of information, product and funds. • Supply chain Planning decisions or tactical decisions
• Each decision should be made to raise the supply chain surplus. • Supply chain Operation level decisions
• These decisions fall into three categories or phases, depending on the frequency
of each decision and the time frame during which a decision phase has an impact.
• The design, planning and operation of a supply chain have a strong impact on
overall profitability and success.

4
Supply chain strategy or design decisions Supply chain strategy or design decisions
• Company decides how to structure the supply chain over the next several years. • These decisions include but not are limited to:
• Long term decisions Whether to outsource or perform a function in-house
• Very expensive to alter on short notice. The location and capacities of production and warehousing facilities
• Consequently, when companies make these decisions, they must take into The products to be manufactured or stored at various locations
account the uncertainty in market conditions over the next few years.

Supply chain planning or tactical level

Supply chain Operation level decisions
decisions
• Time frame considered is a quarter to a year. • Time frame considered is weekly or daily, and during this phase,
• The supply chain’s configuration determined in the strategic phase is fixed. companies make decisions regarding individual customer orders.
• Companies start the planning phase with a forecast for the coming period (may • At the operational level, supply chain configuration is considered fixed,
be a year). and planning policies are already defined.
• Include decisions regarding which markets will be supplied from which locations, • The goal of supply chain operations is to handle incoming customer
the subcontracting of manufacturing, the inventory policies to be followed, the orders, in the best possible manner.
timing and size of marketing and price promotions.

Supply chain Operation level decisions Process views of a supply chain

• During this phase, firms allocate inventory or production to individual orders, set • A supply chain is a sequence of processes and flows that take place within and
a date that an order is to be filled, set delivery schedules of trucks, and place between different stages and combine to fill a customer need for a product.
replenishment orders. • There are two different ways to view the processes:
• Less uncertainty compared to other decision phases. • Cycle view
• Push/Pull view

5
 Push/Pull view
Cycle view
• Processes are divided into a series of cycles, each performed at the
interface between two successive stages of a supply chain. • The processes in a supply chain is divided into two categories depending on
whether they are executed in response to a customer order or in anticipation of
customer orders.
• Pull processes are initiated by a customer order
• Push processes are initiated and performed in anticipation of customer orders.

Push/Pull view of the supply chain Push/Pull view of the supply chain
• Let us compare a make-to-stock environment like that of L. L. Bean and an
assemble-to-order environment like that of Dell to compare the push/pull view
and the cycle view.

Push/Pull view of the Dell supply chain Push/Pull view of the L.L. Bean supply chain

6
Push/Pull view of the supply chain Responsiveness and Efficiency

• Pull processes may also be referred to as reactive processes because they react • Supply chain performance can be measured in terms of responsiveness and
• to customer demand. Push processes may also be referred to as speculative efficiency.
processes because they respond to speculated (or forecasted), rather than actual, • Supply chain’s responsiveness includes a supply chain’s ability to do the following:
demand.  Respond to wide ranges of quantities demanded
• Pull processes operate in an environment in which customer demand is known.  Meet short lead times
• They are, however, often constrained by inventory and capacity decisions that  Handle a large variety of products
were made in the push phase.
 Build highly innovative products
 Handle supply uncertainty

Responsiveness and Efficiency Responsiveness and Efficiency

• Responsiveness, however, comes at a cost. • Cost-responsiveness efficient frontier shows the lowest possible cost for a given
• For example, to respond to a wide range of quantities demanded, capacities must level of responsiveness.
be increased, which increases costs.
• This increase in cost leads to the second definition, supply chain efficiency.
• Supply chain efficiency is the inverse of the cost of making and delivering a
product to the customer.
• Increase in cost lowers efficiency.

Drivers of supply chain performance

7
Drivers of supply chain performance Drivers of supply chain performance

Various drivers in a supply chain interact with each other • These drivers interact with each other to determine the supply chain’s
• Facilities performance in terms of responsiveness and efficiency.
• Inventory • The goal is to structure the drivers to achieve the desired level of responsiveness
at the lowest possible cost.
• Transportation
• These drivers are classified into two categories:
• Information
• Logistical drivers – Facilities, Inventory, Transportation
• Sourcing
• Cross functional drivers- information, sourcing, pricing
• Pricing

Competitive strategy
• Competitive strategy targets one or more customer segments and aims to
provide products and services that satisfy these customer’s needs.

• A competitive strategy will specify, either explicitly or implicitly, one or more

customer segments that a company hopes to satisfy.

• A companies competitive strategy defines, relative to its competitors, the set of

customer needs that it seeks to satisfy through its products and services.
• Eg: Walmart aims to provide high availability of a variety of products of
Supply chain decision making framework
reasonable quality at low prices.

Supply chain strategy Competitive Dimensions

• Supply chain strategy determines
The nature of procurement of raw materials • Cost or Price
Transportation of materials to and from the company • Delivery Speed
Manufacture of the product or operation to provide the service • Delivery Reliability
Distribution of the product to the customer, along with any follow-up service • Coping with Changes in Demand
Specification of whether these processes will be performed in-house or • Flexibility and New - Product Introduction Speed
outsourced.
• After-sale support
• Supply chain strategy also includes design decisions regarding inventory, • Environmental impact
transportation, operating facilities, and information flows.

• Dell’s initial decision to sell direct, its later decision to start selling PCs through
resellers are all part of their supply chain strategies.

8
Competitive Dimensions Types of production
• Job production (intermittent production)
• An operation cannot excel simultaneously on all competitive dimensions. • Batch production (intermittent production)
• Consequently, management has to decide which parameters of performance are • Mass production (flow production)
critical to the firm’s success and then concentrate the resources of the firm on
these particular characteristics.
• Trade-offs
• Project production- generally not in a factory

• Continuous production-special type of mass production. (The difference between

this and mass production is that continuous production is on 24 hours a day, e.g.
blast furnace can run for many years without stopping).
50

Types of production Mass production (flow production)

Types of
Production • In mass production, same type of product is manufactured to meet the
continuous demand of the product.
• Usually demand of the product is very high and market is going to sustain
Flow Intermittent same demand for sufficiently long time.
production Production • E.g. Automobiles, toys, etc.

Mass Job Batch

Production Production production
52

51

Characteristics Of Mass production (Flow Production) Merits and Demerits of Mass production ((Flow Production)
• Merits
• Volume output is large
• Volume of production is high
• Operation sequence are standardized • WIP inventory is minimum
• Automation is possible
• Product layout • Requires less storage space
• Material handling is easy
• Special purpose automatic machines
• Demerits
• Materials are fed at one end and received at the other end
• Less variety or flexibility
• Fixed path material handling equipments • Fault in one operation stops the entire operation
53 54

9
 55 56

Characteristics of intermittent production

Intermittent production
• Flow of production is intermittent
• The basic product design changes from time to time • Volume of production is small compared to flow production
• Wide variety of product is produced
• General purpose machines are used
• Facilities should be flexible enough to handle wide varieties of • They are more complex
products
• Types

• Job Production
• Batch Production
57 58

Job production
Characteristics of Job Production
• In this system products are made to satisfy a specific order.
• However that order may be produced- • General purpose machines required as changes are quite frequent.
• only once • Planning and control system should be flexible enough
• or at irregular time intervals as and when new order arrives • Man power should be skilled enough to deal with changing work
• or at regular time intervals to satisfy a continuous demand conditions.
• The machines are adjusted in such a manner so as to suit the requirements of a • Schedules are actually non existent in this system as no definite data is
particular job. available on the product.
• Ex – General purpose workshop • In process inventory will usually be high as accurate plans and schedules
do not exist.

59 60

10
 Merits & Demerits of Job Production
Characteristics of Job Production Merits
• Small investment in machine and equipment
• Product cost is normally high because of high material and labor costs. • Flexible
• Grouping of machines is done on functional basis (i.e. as lathe section, milling • Fault in one operation does not result into complete stoppage of entire process
section etc.)
Demerits
• This system is very flexible as management has to manufacture varying product
types. • Low volume of production
• Material handling systems are also flexible to meet changing product • WIP inventories are high.
requirements. • Unit cost is high as the speed of work is slow

61 62

Batch production Characteristics of batch type production

system
• Manufacture of products in small or large batches or lots at intervals by • As final product is somewhat standard and manufactured in batches, economy of
a series of operations. scale can be availed to some extent.

• Every batch is different form the other • Machines are grouped on functional basis similar to the job shop manufacturing.

• Mixture of job and mass production • Semi-automatic, special purpose automatic machines are generally used to take
advantage of the similarity among the products.
• Ex –Clothes, medicines, machine tools, shoe making

64
63

Characteristics of batch type production

system
• Labour should be skilled enough to work upon different product
batches.

• In process inventory is usually high.

• Normally production planning and control is difficult due to the odd

size and non repetitive nature of order.

65 66

11
Other production systems Lean Manufacturing
• Lean manufacturing stipulates the attainment of continuous improvement through
• Lean Production System the systematic elimination of wastes.
• Just-In-Time (JIT) production • Time and again, the intensification of competition has been triggering the mankind to
adopt various strategies.
• It is believed that the roots of lean manufacturing began to grow in Ford’s car
manufacturing plant during the beginning of twentieth century.
• It is reported that Henry Ford minimized the material handling distance and cost of
manufacturing by applying a few of the lean manufacturing techniques and tools that
are widely applied today.

Lean Manufacturing Lean Manufacturing

• It has also been successfully applied to administrative and engineering activities • Lean manufacturing is a Japanese method focused on 3M’s.
as well. • These Ms are: muda, the Japanese word for waste, mura, the Japanese word for
inconsistency, and muri, the Japanese word for unreasonableness.
• What Lean has done is to package some well-respected industrial/manufacturing • Muda specifically focuses on activities to be eliminated.
engineering practices into a system that can work in virtually any environment. • Within manufacturing, there are categories of waste.

• The challenge of eliminating inventory and associated wastes in modern

organizations is met by lean manufacturing concepts whose origin can be traced
to Toyota (TPS or JIT)

Lean Manufacturing 5S concept in Lean Manufacturing

• Waste is broadly defined as anything that adds cost to the product without • One of the techniques that is found useful in the application of lean
adding value to it. manufacturing paradigm is 5S.
• Generally, muda (or waste) can be grouped into the following categories: • Seiri: Sort/Structurise
1. Overproduction • Seiton: Streamline/Systematise
2. Delays
• Seiso: Shine-cleanliness
3. Inefficient Transport
• Seiketsu: Standardize
4. Poor process design
• Shitsuke: Self-discipline
5. Excess Inventory
6. Excess Motion
7. Defects

71 72

12
Just-in-time manufacturing Just-in-time manufacturing
• The just-in-time (JIT) inventory system is a management strategy that minimizes • JIT is a manufacturing philosophy, which aims at having the right part at right
inventory and increases efficiency. time, and in the right quantity.
• JIT production systems cut inventory costs because manufacturers receive • In JIT system, a company receives goods as close as possible to when they are
materials and parts as needed for production and do not have to pay storage actually needed.
costs. • For example, if a car assembly plant needs to install airbags, it does not keep a
• Just-in-time manufacturing is also known as the Toyota Production System (TPS) stock of airbags on its shelves but receives them as those cars come onto the
because the car manufacturer Toyota adopted the system in the 1970s. assembly line.
• The success of the JIT production process relies on steady production, high-
quality workmanship, no machine breakdowns, and reliable suppliers.

73 74

Just-in-time manufacturing Kanban system

• With JIT, components are "pulled" through the system to arrive where they are • It is a pull system of inventory and production control.
needed and when they are needed. • It uses Kanban cards as the information transmission device.
• Concentrated a lot on waste elimination and inventory reduction • Kanban is a Japanese word meaning signal or visual card.
• Waste is anything other than the minimum amount of equipment, materials, • A Kanban is a card or a tag usually attached to work-in-process parts and it is
parts, space, and worker’s time which are absolutely essential to add value to the used to facilitate the proper movement of these parts.
product.
• The number of Kanban cards or containers sets the amount of authorized
inventory

75 76

Kanban system PRODUCTIVITY

• The number of Kanbans (containers) =
• Productivity is one of the commonly used term in industrial engineering
Demand during lead time + Safety stock
𝑆𝑖𝑧𝑒 𝑜𝑓 𝑐𝑜𝑛𝑡𝑎𝑖𝑛𝑒𝑟
• In-plant Kanban systems use standardized, reusable containers that protect the
specific quantities to be moved. • It is a measure of how well the resources are utilized to produce an output.

78

13
 PRODUCTIVITY PRODUCTIVITY
• It is the quotient (ratio) obtained by dividing output by one of the factors of
production. • Two types of productivity should not be directly compared.

• Factors of production: Land, Labour, Capital, Raw materials, Machines, etc.

E.g. Productivity of labour and productivity of capital.

79 80

Kinds of productivity measures Kinds of productivity measures

• Labour productivity – in terms of Labour hours (Denominator)
• Direct labour cost productivity – in terms of direct labour costs.

• Capital productivity – e.g. Value of capital investment.

• Direct cost productivity – All items of direct cost associated with the
• Raw material productivity – e.g. weight of raw material consumed in the resources are considered.
denominator.

• Energy productivity – The only resource considered in this case is the

amount of energy consumed.

81 82

• Present production= 24000 components

Example Problem 1 • Present labor productivity

Output 24000
A company is manufacturing 24000 components per month by employing   1 component/manhour
100 workers in 8 hour shift. The company gets additional order from Input 100X8X30
government to supply additional 6000 components/month. The
management decides to employ additional workers. What will be the
productivity level (labour productivity) when the no of additional workers With increased order
employed to produce this additional requirements are (i) 20, (ii) 25 and (iii)
30. (Assume number of working days in a month is 30). a. When 30 workers are hired
b. Production = 24000+6000=30000

Output 30000
  0.96 component/manhour
83 Input (100  30)X8X30 84

14
 With increased order
With increased order
1. When 20 workers are hired

a. When 25 workers are hired 2. Production = 24000+6000=30000

b. Production = 24000+6000=30000
Output 30000
  1.04 component/manhour
Output 30000 Input (100  20)X8X30
  1 component/manhour
Input (100  25)X8X30

85 86

Productivity, Efficiency and Effectiveness

• Production and productivity are two different things

• Efficiency indicates a measure of how well the resources are

• Increased production does not mean increased productivity. utilized to attain a result.

• Higher productivity means that more is produced with the same • Performing or functioning in the best possible manner with the
expenditure of resources; (i.e. at the same cost in terms of land, least waste of time and effort.
material, machine, time or labor.)

• Efficiency is the ratio of actual output attained to the standard

• Alternatively, same amount is produced at less cost in terms of expected output.
land, labor, material etc; thereby releasing some of these
resources for the production of other things.

• Productivity is always associated with the context in which it is87 88

calculated (e.g. material productivity, energy productivity, etc.)

Total productivity and partial productivity

• Total productivity – ratio of aggregate output to aggregate
Productivity, Efficiency and Effectiveness input.

• Effectiveness is the degree of accomplishing the objectives; producing the • Partial productivity is the ratio of the aggregate output to
intended or expected result. any single input.

• Being effective is about doing the right things, while being efficient is about doing • Examples of partial productivity are labor productivity,
things right. capital productivity, etc.

• Productivity is the integration of both effectiveness (i.e. performance) and

efficiency (i.e. resource utilization).
89 90

15
Productivity index
• Since productivity is a relative measure, for it to be meaningful or useful it must
be compared to something.

For example Forecasting

Businesses can compare their productivity values to that of similar firms
Other departments within the same firm or
Against past productivity data for the same firm or department

91

Forecasting Characteristics of forecasts

• Forecast is an estimate of an event which will happen in future • Forecasts are always inaccurate and should does include both the expected
value of the forecast and a measure of forecast error
• Forecasts of future demand are essential for making supply chain decisions • Long-term forecasts are usually less accurate than short-term forecasts.
• Aggregate forecasts are usually more accurate than disaggregate forecasts
as they tend to have a smaller standard deviation of error relative to the
• Demand forecasts form the basis of all supply chain planning mean.
• In general, the farther up the supply chain a company is, the greater is the
distortion of information it receives.
• As a result, the farther up the supply chain an enterprise is, the larger is the
forecast error.

Forecasting Forecasting
• To forecast demand, companies must first identify the factors that influence future
demand and then find out the relationship between these factors and future demand. • Forecasting methods are classified into two categories:
• Companies must balance objective and subjective factors when forecasting demand.
• Some of the factors that are related to the demand are:
Qualitative forecasting
Past demand
Quantitative forecasting
Promotions
State of the economy
Planned price discounts
Actions by the competitor

16
Qualitative forecasting Quantitative forecasting
• Quantitative forecasting methods use historical data to make a forecast
• These methods are primarily subjective and rely on human judgement
• They are based on the assumption that past demand history is a good
• They are most appropriate when little historical data is available indicator of future demand
• These are the simplest methods to implement and can serve as a good
starting point for a demand forecast

• Time series analysis and causal forecasting methods are two types of
quantitative methods of forecasting

Time series analysis Time series analysis

• The systematic component of demand consists of Level, trend and seasonality
• Time series is a sequence of observations ordered by a time parameter components
• Prediction of future demand from past interval data • Level component shows current de-seasonalized demand
• Time series methods are most appropriate when future demand is • Trend component shows rate of growth or decline in demand
related to historical demand, growth pattern and any seasonal pattern. • Seasonality component shows predictable seasonal fluctuations in demand
• Demand patterns or components of time series data – level, trend, • A company cannot forecast direction of random component
seasonal and cyclic • All a company can predict is the size and variability of the random component which
• With any forecasting method, there is always a random element that provides a measure of forecast error
cannot be explained by historical demand pattern • The forecast error measures the difference between the forecast and actual demand
• Therefore, any observed demand can be broken down into a systematic
and random component.
• Observed demand = systematic component + random component

Time series forecasting methods Causal method of forecasting

• Simple moving average method • These methods assume that the demand forecast is highly correlated with certain
• Weighted moving average method factors in the environment such as the state of the economy, interest rates, etc.
• Simple exponential smoothing method • Causal forecasting methods find this correlation between demand and
environmental factors and use this to forecast future demand
• Trend adjusted exponential smoothing method
• For example, companies can use causal methods to determine the impact of price
• Trend and seasonally corrected exponential smoothing method promotions on demand
• Linear regression analysis is a causal method of forecasting

17
Moving average method Simple moving average method
• The moving average method is used when demand has no observable trend or • In this method, demand is forecasted as the average demand over the most
seasonality. recent N periods.
• There is only level component • Thus, this method is also called as N period moving average method.
Two types: • In this method, average moves whenever new period demand is obtained.
• Simple moving average method • Forecast for (t+1)th period, Ft+1 = Lt
• Weighted moving average method
• Lt is the average demand over the last N periods

Simple moving average method Simple moving average method

Solved example 2: Ans:
• A supermarket has experienced daily demand of milk of 120, 127, 114 and 122 • F5 (using 3-period MA)=121 litre
litres over the last four days. (a) Forecast demand for day 5 using 4-period and
3-period moving average methods. (b) Suppose that the actual demand for day
5 is 125 litre, forecast demand for day 6 using the above two methods. • F5 (using 4-period MA)= 120.75 litre

• F6 (using 3-period MA)= 120.33 litre

• F6 (using 4-period MA)= 122 litre

Weighted moving average method Weighted moving average method

• In this method, weights are assigned to the data in different periods rather than Solved example 3:
providing equal weights as used in simple moving average method.
• Forecast for (t+1)th period, Ft+1 = Lt • Weightage for last 3 periods are 20 %, 30 % and 50 %, where 50 % is the
weightage for the last period. (a) Forecast demand for day 5 using 3-period
• Lt is the weighted average of demand over the last N periods. weighted moving average method. (b) Suppose that the actual demand for
day 5 is 125 litre, forecast demand for day 6 using the above method.
Ans:
• F5 (using weighted MA)=120.6 litre

• F6 (using weighted MA)= 121.9 litre

18
Simple exponential smoothing method
Simple exponential smoothing method
• The simple exponential smoothing method is appropriate when
demand has no observable trend or seasonality. • Ft+1 = Lt = Lt-1 + α(Dt - Lt-1)
• In this case the systematic component of demand = level = αDt + (1- α) Lt-1
component • Where t is the current period (period at which the most recent demand is
• This method does not eliminate any past information, but so known)
adjust the weights given to past data that older data gets
increasingly less weightage (hence the name exponential • α - smoothing constant (the proportion of error that will be incorporated
smoothing) into the forecast)

• The current estimate of the level in exponential smoothing method is a

weighted average of all the past observations of demand with the recent
observations weighted higher than older observations

Simple exponential smoothing method Trend adjusted exponential smoothing method

• Larger values of α gives more weightage to recent demands and this • Holt’s Model
provide responsive forecast. • This method is used to incorporate correction for trend in the demand data.
• Smaller values of α produce more stable forecast. • This method is appropriate when demand is assumed to have a level and a trend
• An initial estimate of level is essential for determining the forecast. pattern in systematic component.
• One way to determine the initial estimate of level is to take the average of • Systematic component of demand = level + trend
all historical data. • An initial estimate of level and trend is essential to use this method.

• Solved example 4:
• Ans: 521 units

Trend adjusted exponential smoothing method Trend adjusted exponential smoothing method

• Trend adjusted exponential smoothing forecast = Ft+1 = Lt + Tt • An initial smoothed value and values for smoothing constants α and β are
needed for using exponential smoothing method.
• Lt = αDt + (1- α) [Lt-1 + Tt-1] • There is no good rule for finding the values of α and β.
• Tt =β (Lt- Lt-1) + (1- β) Tt-1 • These are the limitations of exponential smoothing methods.

• α, β – smoothing constants
• β – smoothing constant for trend adjustment Solved example 5:
Ans: 12072 units

19
 Trend and seasonally corrected exponential Trend and seasonally corrected exponential smoothing
smoothing method method
• Ft+1 = (Lt + Tt)*St+1
• Winters model of forecasting • Lt = α (Dt/St) + (1- α) (Lt-1 + Tt-1)
• This method is appropriate when the systematic component of demand has a • Tt =β (Lt- Lt-1) + (1- β) Tt-1
level, a trend and a seasonal factor.
• St+p = γ(Dt/Lt) + (1- γ)*St
• In seasonal forecasting models, usually the year is divided into a number of Notations:
seasons and data fluctuates in each season.
• St – seasonal component (seasonal factor) in period t
• Systematic component of demand =
• St+p – seasonality at (t+p)th period
(level + trend)* seasonal factor • p – periodicity of demand
To begin with initial estimates of level, trend and seasonal factors are required. • γ – smoothing constant for seasonal factor
• Seasonal factor is the amount of correction needed in a time series to adjust for the
season of the year.
Solved example 6: [Ans: 9645.4, 14291.2]

Determination of seasonality index: Simple

Linear regression analysis – Causal method
proportion method
Solved example 7: • Regression is defined as a functional relationship between two or more
• During the past years, a retailer firm sold an average of 1000 units of a particular correlated variables.
product each year. On the average, 200 units were sold in the spring, 350 units in • It is used to predict one variable given the other.
the summer, 300 units in the fall, and 150 in the winter. Determine the seasonal • It is classified into simple linear regression and multiple linear regression.
factor in each season. If the firm expect an annual sales of 1100 for the next year,
forecast demand for each of the four seasons in that year. • In simple linear regression, there is only one independent variable and in multiple
regression there are two or more independent variables.

• Ans: [0.8, 1.4, 1.2, 0.6],

• Ans: [220, 385, 330, 165]

Simple Linear regression Simple Linear regression

• Least square method to fit a line
• In simple linear regression, the relationship between dependent and
independent variables forms a straight line.
• y = a+bx
• Least square method is used to fit a line to the given data
• i.e., in least square method, a line is fit to the data in such a way that the sum of
the squares of the error is minimum.

20
Simple Linear regression
• Y = a+bx
• a = 𝑦 − 𝑏𝑥
𝑥𝑦−𝑛𝑥 𝑦
• b= 2 2
𝑥 −𝑛 𝑥
Simple Linear regression - Standard error of
• a = Y-intercept
• b = slope of the line or trend the estimate
• Y = dependent variable
• x = independent variable
• n = number of periods under consideration

Solved examples 8 & 9:

Checking the Estimating Equation Interpreting the standard error of estimate

• To measure the reliability of the estimating equation, statisticians have developed the • The larger the standard error of estimate, the greater the scattering (or
standard error of estimate. dispersion) of points around the regression line.
• This standard error is symbolized se and is similar to the standard deviation. • If SE is zero, what is the meaning?
• The standard error of the estimate, denoted se, is a measure of the standard deviation of
the errors in a regression model. • Assuming that the observed points are normally distributed around the
regression line, we can expect to find 68 percent of the points within ± 1Se (or
• Instead of measuring the average distance from the mean we are measuring the average
distance from the regression line. plus and minus 1 standard error of estimate), 95.5 percent of the points within
±2Se, and 99.7 percent of the points within ±3Se.
• It represents the average distance that the observed values fall from the regression line.
• The value of se tells us, on average, how much the dependent variable differs
from the regression line based on the independent variable.

• We lost two degrees of freedom while estimating the regression line (a,b)

Determination of seasonality index using

Interpreting the standard error of estimate
regression line
• Consider the solved example 9: Solved example 10:
• If the amount spend is 8 Lakhs, estimated value = 36 Lakhs • Forecast the demand for each quarter of the next year using trend and seasonal
• Determine the Standard error factors. Demand for the past two years is in the following table:
3.24 Lakhs
Year Quarter Demand
• On average, profit differs by 3.24 Lakhs from the regression line based on the I 1 300
amount spent on R&D 2 200
• Now we can interpret that we are roughly 95.5 percent confident that the actual 3 220
profit will be between (29.52 and 42.48)…How? 4 530
II 1 520
2 420
3 400
4 700

21
Determination of seasonality index using
Qualitative methods of forecasting
regression line
• Y= 52.262t + 176.07 • Qualitative methods or subjective methods are used when no past data is
• Seasonality index: 1.25, 0.79, 0.70, 1.28 available or if some data is available but cannot be used for long run
forecast
• Forecast: [808, 552, 526, 1029]
• Mostly used for new technology or new products or new region
• Require knowledge of experts and require much judgement.
• For example, forecasting of new fashion merchandise in a retail store –
input from customers, store managers, etc.
• No wild guesses……but well-thought-out and structured decision making
approach….

Qualitative methods of forecasting Qualitative methods of forecasting

Some of the methods are: Consumer opinion survey
• Consumer opinion survey • Buyer is asked about their future buying intentions of products and their
• Personal interviews of sales representatives brand preferences.
• Market trial • From the representative samples, demand is forecasted.
• Survey of supply chain partners
• Delphi method

Qualitative methods of forecasting Qualitative methods of forecasting

Personal interviews of sales representatives Market trial
• In this method, opinion of sales representatives is collected • This method is used for new products.
• Aggregation of opinion of sales representatives is used to obtain sales • New products are exposed to limited market demand and feedback from
forecast of a region customers are carefully analysed to make a forecast

Survey of supply chain partners

• Estimate of expected sales can be obtained from distributors or from
retailers

22
 Qualitative methods of forecasting Forecasting Error
Delphi Method • Every instance of demand has a random component
• In Delphi Method, several knowledgeable persons (experts) are asked to provide estimates of
demand forecasts through a questionnaire. • This random component appears in the form of forecast error
• The experts may provide several opinions • If a firm's forecast tend to consistently over or underestimate demand, this
• Experts answer questionnaires in two or more rounds, after each round a facilitator provides an may be a signal that the firm should change its forecasting method.
anonymous summary of the experts responses as well as the reasons provided for their
judgements
• Forecast error in period t = Ft – Dt, where Ft is forecasted demand in
• Experts are encouraged to revise their earlier answers in light of the replies of other members period t and Dt is the actual demand in period t.
of their panel
• New questions may be added in each round
• Thus, Delphi Method is iterative in nature
• Consensus usually emerged at the end of the exercise
• Usually in Delphi Method, all participants remain anonymous to avoid personal bias.
• Usually achieve satisfactory results in three rounds.

Measures of Forecast Error Measures of Forecast Error

• Mean forecast error Tracking signal
• Mean absolute deviation • It is a measurement that indicates whether the forecast average is keeping
• Mean squared error pace with any genuine upward or downward change in demand.
• Mean absolute percentage error • It is the ratio of bias or running sum of forecast error (RSFE) and mean
absolute deviation.
• Tracking signal
• It can be positive or negative
• The bias will fluctuate around 0 if the error is truly random.
• If the tracking signal at any period is high, this is a signal that the forecast
is Biased and is either under forecasting for over forecasting

Measures of Forecast Error Measures of Forecast Error

Tracking signal Solved example 11:
• For example, suppose that the demand has a growth trend and the
manager is using a moving average method, the forecast will be always • The forecasted value of 160, 164, 167, 170 and 172 units had been
less than the demand. made for the demand in five consecutive periods. Actual demand data is
• This will result in a large negative tracking signal 150, 160, 165, 175 and 180 units respectively. Calculate MFE, Mean
absolute deviation, Mean squared error, Mean absolute percentage
error and Tracking signal in all the periods

23
 Measures of Forecast Error Measures of Forecast Error
Period Forecast Demand Error Absolute Squared Absolute %
error error error
Period Forecast Demand Error Absolute Bias MAD TS
1 160 150 10 10 100 6.67 error (RSFE)
2 164 160 4 4 16 2.5 1 160 150 10 10 10 10 1
3 167 165 2 2 4 1.21 2 164 160 4 4 14 7 2
4 170 175 -5 5 25 2.86 3 167 165 2 2 16 5.33 3
5 172 180 -8 8 64 4.44 4 170 175 -5 5 11 5.25 2.09
Sum 3 29 209 17.68 5 172 180 -8 8 3 5.8 .52

• MFE = 3/5 = 0.6

• MAD = 29/5 = 5.8
• MSE = 209/5 =41.8
• MAPE = 17.68/5 = 3.54

Measures of Forecast Error Measures of Forecast Error

Solved example 12: Solution:
• A specific forecasting model was used to forecast the demand for a product. The • MAD = 52.5
forecasts and the corresponding demand are shown as follows. Use MAD, • TS at the end of sixth period = 1.05
tracking signal technique and MAPE to determine the forecast error.
• MAPE = 6.53 %

period Forecast Actual demand

1 660 700
2 840 760
3 750 780
4 835 790
5 910 850
6 890 950

Aggregate planning Aggregate planning

• Aggregate planning is a process by which a company determines ideal levels of • Aggregate planning answers the question
capacity, production, subcontracting, inventory, stockouts, and even pricing over a “how should a firm best utilize the facilities that it currently has?’’
specified time horizon – aggregate levels
• Sales and Operations planning
• Goal of Aggregate planning is to satisfy demand while maximizing profit
• To be effective, it requires input from throughout the supply chain, and its
• Specify planning horizon – 3 to 18 months results have a tremendous impact of supply chain performance.
• Duration of each period in planning horizon
• It is an intermediate range planning.

24
 Aggregate planner requires the following
Aggregate planning
information
• Demand forecast • Constraints
• Production costs • limits on overtime
• Labour cost, regular time and over time costs • Limits on layoffs
• Cost of subcontracting • Limits on capital available
• Cost of changing capacity • Limits on stock out and backlogs
• Labour/machine required per hour • Constraints from supplier to the enterprise
• Inventory holding cost
• Stockout / backlog cost

Aggregate planning Aggregate planning

• Using this information a company makes the following determinations through • Ideally, all stages in the supply chain should work together on an aggregate plan
aggregate planning that optimizes the supply chain performance.
• Production quantity from regular time, overtime, and subcontracted time • Lack of coordination can result in:
• Inventory held lost sales and thus lost profits or
• Backlog or stockout quantity
large amount of excess inventory and capacity, thereby raising costs.
• Workforce Hired or Laid-off
• Machine capacity • Aggregate planning is an important tool to optimally match supply and demand.

Aggregate planning
• An aggregate planner must make trade-offs among capacity, inventory and
backlog costs.
• Most profitable combination
• The aggregate planning problem can be stated as follows:
“Given the demand forecast for each period in the planning horizon, determine the
Aggregate planning strategies
production level, inventory level, and capacity level (internal & outsourced) for
each period that maximizes the firm’s profit or minimizes the total cost while
satisfying various constraints over the planning horizon”

25
 Chase strategy
Aggregate planning strategies • Using capacity as the lever
• With this strategy, the production rate is synchronized with the demand rate by
• There are three distinct strategies. varying machine capacity or hiring and laying off employees as the demand rate
varies

Chase strategy Level strategy

• It can be expensive to implement if the cost of varying machine or labour capacity • With this strategy, a stable machine capacity and workforce are maintained with a
is very high constant output rate.
• It can also have a significant negative impact on the morale of the work force
• It results in low level of inventory - advantage
• It should be used when the cost of carrying inventory is very high and cost of
change levels of machine and labour capacity are low

Level strategy Time flexibility (from work force/capacity) strategy

• Shortages and surpluses result in inventory levels fluctuating over time • In this case the work force is kept stable but the number of hours worked is
• Production is not synchronized with demand varied (eg: overtime is used) in an effort to synchronize production with demand
• Employees benefit from stable working conditions • It avoid some of the problems associated with the chase strategy related to
changing the work force.
• Either inventories accumulate in anticipation of future demand or customer
orders may delayed during periods of high demand. • It can be used if there is excess machine capacity (for example, machines are not
used 24 hours a day, seven days a week)
• It should be used when inventory carrying cost and backlog costs are relatively
low. • This strategy results in low levels of inventory compared to level strategy.
• Large inventories and customer orders may get delayed are disadvantages • Used when machine capacity is relatively inexpensive.

26
 Aggregate planning strategies
• In addition to the above strategies, Subcontracting can also be used as a strategy
to meet demand in peak periods.
• If a single strategy is used it is called a pure strategy.
• Most strategies that a planner actually uses are a combination of these strategies
and are referred as mixed or tailored strategy.
Aggregate planning methods (example)

Linear Programming Model Linear Programming Model

• It is an optimal method to solve aggregate planning problem. • The linear programming model includes three basic elements
• Linear programming finds the solution that creates highest profit or least cost • Decision variables that we seek to determine
while satisfying the constraints that the company faces. • Objective function that we aim to optimize
• Constraints that we need to satisfy

Cost parameters
• r : Regular time production cost/unit
• Decision Variables Formulation of linear programming problem
• o : Overtime production cost/unit • 𝑅𝑡 : Regular time production in period t
• i : Inventory cost/unit • 𝑂𝑡 : Overtime production in period t
• 𝐼𝑡 : Inventory at the end of period t • Identifying the objective underlining the problem
• h : Hiring cost/additional unit produced
• 𝐻𝑡 : Additional units obtained through hiring in period
• l : Lay-off cost/unit reduction of production t • Identifying the decision variables
• out : Outsourcing or subcontracting cost/unit
• 𝐿𝑡 : units reduced through layoff of workers in period t • Construction of the objective function
• 𝑂𝑢𝑡𝑡 : Units outsourced in period t
• S: stockout cost/unit • St: stockout units at the end of period t • Construction of the constraints
Other notations • Non-negativity restrictions
Dt: Projected demand during the period t
k: maximum allowable overtime as a proportion
of regular production.
α: minimum amount to be ordered to use
subcontracting/outsourcing.
I* - Maximum inventory that can be carried in a
period to the next period
n: number of periods considered

27
 • Objective function
Minimize
𝑛 Aggregate Planning - Linear Programming
𝑟𝑅𝑡 + 𝑜𝑂𝑡 + 𝑖𝐼𝑡 + ℎ𝐻𝑡 + 𝑙𝐿𝑡 + 𝑜𝑢𝑡. 𝑂𝑢𝑡𝑡 + 𝑠𝑆𝑡
𝑡=1
• The objective function minimises the total cost including regular time production,
overtime production, subcontracting, Hiring & Laying-off workers, stockout cost
Constraints and inventory cost.
(𝐼𝑡−1 +𝑅𝑡 + 𝑂𝑡 + 𝑂𝑢𝑡𝑡 ) − (𝐷𝑡 + 𝑆𝑡−1 ) = 𝐼𝑡 − 𝑆𝑡 (1) • First constraint represents the inventory balance equation.
𝑂𝑡 ≤ k 𝑅𝑡 (2) • i.e. the difference between the total quantity available in a period (obtained
using regular time, overtime and subcontracting production, along with previous
𝑅𝑡−1 + 𝐻𝑡 − 𝐿𝑡 =𝑅𝑡 (3) period remaining inventory) and the demand and previous backlog in that period
is equal to the current period remaining inventory or backlog quantity.
𝐼𝑡 ≤ I* (4)
• It is assumed that no shortage is allowed.
𝑂𝑢𝑡𝑡 ≥ α or 0 (5)
All variables are positive (6)

Aggregate Planning - Linear Programming Aggregate Planning Example Problem

• Second constraint ensures that the overtime production in a period should
be less than or equal to a known percentage (k) of regular time capacity. Solved example 13:
• Jason Enterprises (JE) produces Telephones for the home market. Quality is not
• Third constraint shows that the total available regular time capacity in a quite as good as it could be at this point, but the selling price is low and Jason can
period is obtained by adding (or subtracting) the capacity hired (or laid-off) study market response while spending more time on R&D. At this stage, however,
to the previous period regular time capacity. JE needs to develop an aggregate production plan for the six months from
• Fourth constraint shows the upper limit of inventory that can be carried in January through June. You have been commissioned to create the plan. What is
a period to the next period.
the cost of each of the following production strategies?
a) Produce exactly to meet demand; vary workforce (assuming opening workforce
• Fifth constraint shows the minimum amount (α ) to be subcontracted. equal to first month’s requirements).
b) Constant workforce; vary inventory and allow shortages only (assuming a
workforce of 10).
c) Constant workforce of 10; use subcontracting.

Aggregate Planning Example Problem

Solution:

Plan description Total cost

Plan a $203300
Plan b $202000
Plan c $281400

28