Manufacturing Automation

Implementation Guide
Deciding What to Automate - and When
Custom automation is a valuable tool for cutting costs, improving quality, and boosting productivity. It’s one of the
best measures domestic manufacturers can use to remain competitive, provided they have product demand that
justifies the cost.

Why Automate?
01 High demand, increased capacity, and cost savings: An increase in
demand requires capacity. Rather than replicating the existing process, 

3 Commons Scenarios a decision is made to automate and obtain labor savings.
Most manufacturers automate a process for two reasons: cost and capacity.
But while the cost and capacity generally drive the process, other considerations
02 Cost savings through labor education: A labor-intensive process is

replaced with automation to reduce costs, streaming workflows, and

also impact decision-making. Three common scenarios include: making operations more efficient.
New product introduction with high demand: For new products,
03 capacity projections require multiple production lines. A decision is made
to pursue an automated solution that can meet demand with a single
system.

Reasons to Automate
Beyond Capacity and Cost
Improved quality. Automation provides consistency which can be difficult to Elimination of repetitive motion injuries. Repetitive jobs often lead 


maintain with manual processes. It also frequently includes automated to injuries. Some of these injuries can be designed out of a process, but

inspection, which can continuously verify product quality throughout the automation often provides the best means to completely eliminate an issue.

production day.
Elimination of labor-intensive boring work. Truly boring jobs can be a

problem to support in the factory, as workers tend to migrate away from them,
Elimination of hazardous jobs. Anything that involves toxic materials 

creating an on-going “hole” that must be continually filled with new workers.
or physically dangerous processes is a candidate for automation, as the

elimination of risk is always a powerful driver for change.

Towards Balance Utilization

Good automation projects are based on a growing demand for 
 As a rule of thumb, most systems should be designed to deliver 100% of demand
the product, and the need to expand capacity. However, the principal feature of a at 75% to 85% utilization. If product demand varies considerably over a year, the
cost-effective automation project is utilization.
safety factor between rated capacity and demand should be set higher.  

The system must be utilized at, or near, its designed capacity, or it probably won’t One should be careful about setting the rated capacity too high relative 

make economic sense. Custom automation is not cheap. However, it can be to the expected near-term demand. System capacity drives system 

quite cost-effective when the system capacity is in balance with system cost – as one increases, so does the other. If the capacity target is set too high
utilization. System utilization is in balance with capacity when the rated output of above the actual demand, utilization will be low, and the system is likely to prove
the system is slightly higher than annualized product demand. too expensive to implement.

14725 SE 36th Street Suite 300 Bellevue, WA 98006 | (425) 201-3713

Justification & Management Approval
Before considering an automation project, the buying engineer should be familiar with the approval process, particularly with respect to
x
large capital e penditures. At some point during the process, the buying engineer or manager will need to actively sell this concept to
management.

“Show Me The Money”

Ultimately, the approval of all automation projects hangs on their economic

justification. Automation projects typically require approval from plant or

corporate management, and that approval will be based on the projected return

on investment.

A typical justification for an automated system should include the following

An estimate of savings per unit, broken down by category – labor, quality 


or other

An estimate of the savings per year based on the current forecast

An analysis of the return on investment

A discussion of other benefits that may not be quantified in the economic

analysis

Some form of approval/authorization to be signed by management.

Estimating Standard Labor Costs

To accurately assess the cost reduction through labor savings that an automation project will bring, one must first have a clear estimate of the current labor costs
associated with the current or proposed less-automated approach. For most economic justifications, we are interested in the portion of the semi-burdened labor rate

that would be reduced if the resource was no longer needed, which includes vacation, tax, and healthcare expenses, but not overhead.

Burdened Labor Rate = Example

(Direct Labor Cost * 1.075) + (Healthcare Direct Labor Cost

$18 / hour

Cost + Vacation Cost)/2000 Direct Labor Cost with Taxes

$19.35 / hour

Where Healthcare Cost

$13,800

Direct Labor Cost

Resource’s hourly rate
Vacation Cost (@ 80 hours)
$1,548

Healthcare Cost
Annual cost of healthcare plan
Annual Benefits Cost
$15,348

Vacation Cost Annual cost of vacation plan

Hourly Benefits Cost $7.67 / hour
(hourly rate * days off)
Burdened Labor Rate =
$27.02 / hour or $54,040/year

In its simplest form, per-unit labor costs can be derived from the hourly cost for resources, divided by the

Estimating Per
number of units made.

Unit Labor Costs

Per Unit Labor Cost ($/Unit) = Hourly Labor Cost / Units per Hour

14725 SE 36th Street Suite 300 Bellevue, WA 98006 | (425) 201-3713

14725 SE St #300, Bellevue, WA 98006 | (425) 201-3713
Making Intangible Benefits Tangible
Savings in scrap costs should be reduced to a per unit savings, using the scrap Where:
cost over a known period, and the production volume in that same period.
#Task
Number of unique tasks

Per Unit Scrap Savings ($/Unit) = Scrap Cost / Units Produced #Feeders
Number of parts to be automatically fed

#Transfer Number of expected transfer systems

Guesstimating a System Cost Small Parts Assembly & Test:

$25,000 ≤ Cost per Task ≤ $100,000

Automation integrators typically develop a system cost estimate from a detailed Medium Parts Assembly & Test:
$50,000 ≤ Cost per Task ≤ $125,000

concept of the system, based on the requirements and specifications provided Large Parts Assembly & Test: $75,000 ≤ Cost per Task ≤ $150,000
by the customer. A detailed cost estimate is a difficult and laborious undertaking,
and not something that easily falls within a simple set of rules.

Each system is unique, and the ultimate system cost will depend on the Payback Analysis
complexity of the requirements, and the efficiency and ingenuity of the
In essence, payback analysis weighs the attractiveness of a cost-saving
integration.
investment by calculating the time required for the savings to pay back the
However, system cost is driven by the following generalities: investment. Payback is usually expressed in terms of months or years.
The number of tasks that will be automated. This includes loading and
installing components, and any special processing steps, like soldering, Payback (years) = Capital Payback (months) = Capital
Or
crimping, welding, etc. Investment / Annual Savings Investment / Monthly Savings

The number of part feeders that will be used. Automated feeders can be a
significant cost in any overall system concept.
Total Cost of Ownership
The number of transfer systems that will be used. Fully automated systems
have a minimum of one transfer system. Some systems can have multiple In a simple TCO analysis, one determines the cost per unit of each solution. This
transfer systems that hold and transport the product in different cost includes operating cost per unit of the solution, plus the cost of any capital
configurations, and possibly at different rates. equipment amorti ed over the total product volume expected over the life of the
z

product. If the total volume required exceeds the capacity of the solution, it is
Based on these three inputs, it is possible to generate a rough order of assumed that the solution will be replicated, and the capital cost ad usted to j

magnitude (ROM) system cost based on the following formula: reflect multiple production lines.

System Cost = (#Tasks + #Feeders + #Transfer) * Cost per Task TC ($/unit) = perating Cost per Unit + Tota Capita Costs / Tota
O O l l o ume
l V l

NPV A nalysis Desc iption

Project Payment #1
r

$
Outflow

35,000 $
Inflow

12,000 $
New Flow

(35,000)

Net Present Value, or NPV, is an evaluation tool that considers the timed cash
. $ - $ . $ -

. $ - $ . $ -

flow, a discount rate, and returns a value in present dollars for the proposed
“ ” Project Payment #2 $ 25,000 $ . $ (25,000)

cash flow. For capital pro ects, the cash flow usually takes the form of an
j
. $ - $ . $ -

investment outlay at the start, followed by a series of returns, typically from cost
Project Payment #3 $ 35,000 $ . $ (35,000)

Project Payment # 4 $ 50,000 $ . $ (50,000)

savings. More sophisticated analysis might show multiple investments at the . $ - $ . $ -

beginning of the pro ect, representing the milestone payments to the supplier,
j
. $ - $ . $ -

M hl S vi gs i Y
followed by inflows and outflows over time for savings and operating costs.
ont y a n n ear 1 $ - $ 10,000 $ 10,000

M ont hl S vi gs i Y
y a n n ear 2 $ - $ 12,000 $ 12,000

N C sh Fl w
et a o $ 11 9
,000

Table 1 Dis u R co nt ate $ 3%

The time frame is set to two and a half years, with production starting halfway Net Pre s ent V lu
a e $ 11 9
,000

through the first year. Outflows are shown for the pro ect milestone payments,
j

and inflows are shown for cost savings derived from reduced labor. The total Desc iption r Outflow Inflow New Flow

system cost is set at 125,000, and a total savings of 119,000 for the first two
$ $
Project Payment #1 $ 125,000 $ . $ 125,000

A u l S vi g Y 4 4
years is shown at the bottom. Based on a payback analysis, this pro ect would fail
nn a a n - ear 1 $ - $ 11 ,000 $ 11 ,000

j
A u l S vi g Y
nn a a n - ear 2 $ - $ 1 68 ,000 $ 1 68 ,000

the two-year payback analysis, and might not be approved. A u l S vi g Y

nn a a n - ear 3 $ 25,000 $ 1 9 2,000 $ 1 9 2,000

A u l S vi g Y
nn a a n - ear 4 $ - $ 2 6 1,000 $ 2 6 1,000

Table 2 A u l S vi g Y
nn a a n - ear 5 $ 35,000 $ 2 4 0,000 $ 2 4 0,000

We stretch the time frame to five years and show an increase in savings over this
N C sh Fl w
et a o $ 8 35,000

timeframe. This increase might come from increases in hourly labor rates due to Dis u R co nt ate $ 3%

wage increases, overhead, or healthcare costs. In this case, the system looks Net Pre s ent V lu
a e $ 72 6 ,02 8

much more attractive, showing a total savings of 726,000 over five years,
$

including the time value of money.

14725 SE 36th Street Suite 300 Bellevue, WA 98006 | (425) 201-3713

Defining the Project
Project definition is a critical element in any successful automation project. Failure to define the project requirements succinctly and
accurately will result in unnecessary difficulty, higher costs, and delays as issues arise requiring corrective action.

Key Goals for Good Project Definition

Be Organized. Divide requirements into sections that represent the key features Be careful how you rely on standards. All specifications refer to industry

desired in the solution. Keep specifications relevant to these features grouped standards. However, standards are written in a very broad format, with multiple

together. Avoid specifying items that pertain to one sub-system in multiple interpretations. Include the standard, but be sure to state your specific

places. requirement.

Be specific and direct. Ask for what you want. If the supplier disagrees with Eliminate anything that doesn’t pertain. If you are reusing a specification,

something, they will voice this disagreement. using company boilerplate, or referencing company standards, exclude anything

that doesn’t pertain to the system at hand. Unnecessary inclusions add

Spell it out. Write your requirements in a bulleted list clearly stating 

complexity, making it more difficult to interpret your essential requirements.
each requirement.

Define the Scope Identify the Technical Requirements

The scope typically begins with a brief statement concerning what is to be done, The key to any project definition is the technical requirements deemed important
and how it is to be accomplished. Generally, this is stated as a list of phases or by the customer. These requirements should be a comprehensive view of the
top-level tasks, such as design, construction, development, and deployment. The features and parameters which the customer considers necessary for success. 

scope should also include a statement about installation and training to ensure Requirements are best presented as a bulleted list, organized by topic. 

that these items are addressed. Each statement should describe what is required, defining any relevant metrics
and their acceptable limits.

Describe the Deliverables

The deliverables section of your request for proposal should include all of the A list of deliverables typically includes the following :
physical and project items expected from the vendor. Mechanical drawings of consumable or perishable tooling. This might
include robot gripper fingers or part nests, which tend to wear with usage
This includes, but is not limited to, the following:
Assembly drawings with a bill of materials on each. The bill of materials
Reviews, meetings, or status reports
should indicate a supplier for any off-the -shelf components
Analysis, such as a process FMEA or risk assessment
Electrical and pneumatic schematics
Documentation, such as drawings, CAD models, or software source code
Source code for the PC, robot controller, or PLC
Tests, such as buy off or a process mockup
An operating manual, detailing how to start and stop the system 

Services, such as installation, installation support, production ramp-up, or
and perform some system troubleshooting.
training.

14725 SE 36th Street Suite 300 Bellevue, WA 98006 | (425) 201-3713

Se e l cting a Vendor
The successful custom automation integrator has experience with a wide range of technologies and industries, has worked at
varying production speeds, and can build systems ranging in size from semi-automatic lean task-specific stations to fully-
integrated automated production lines. But picking the right one for your unique project is vital.

Capabilities & Experiences The specialized integrator is best suited to difficult tasks within their area of
expertise but is weaker when the project requirements extend outside this
Capability and experience can come in two principal forms – depth or area.
breadth. An integrator may have great depth of experience in a particular
The generalized integrator is better suited for broad-ranging projects, with
technology, or they may have a broader experience in a number of technologies,
multiple technologies, and will be nearly equal to the specialized integrator
but no particular specialty.
within their area of expertise on all but the hardest tasks. This is because the
generalized integrator leverages the expertise of its own supplier network to
bring expertise to bear as necessary.

Capacity
From a manpower perspective, a project team ranges in size from one or two individuals to a full-size team of
engineers, technicians, and programmers. The team should be reflected in a work plan that enables the design
team to complete their tasks within the first 25% of the project timeline.

Tools
Machine and automation design is best done in a 3D 

solid-modeling environment.

Use of this design environment represents current best practice for creating

robust and successful systems. 

It also provides tools for communicating these designs to the customer 


for review and discussion. Suppliers who have not embraced the solid

modeling world are behind the times, and will generally provide technologies
In a similar vein
and solutions which are also behind the times.
Machine building typically
Machine building is a combination of activities from a variety of disciplines. requires the fabrication and
S ome of these activities can be successfully outsourced, such as welding and wiring of electrical panels, and
fabrication, while others are better kept in-house. field wiring of the final machinery.
These capabilities, and their
The efficient machine builder typically has the capability to machine most of
associated tools, should also 

their components in-house. They may send larger items out for machining but
should be able to meet most of their need through vertical integration. be included within the vertical
manufacturing realm of the
I t is best if this machining capability includes CNC machines, and even better builder.
if these CNC machines are linked directly to the CAD system, to create a
CAD-CAM process.

147 25 SE 36th Street Suite 300 Bellevue, WA 98006 | (4 25) 201-3713

Managing the Project

Custom automation is a valuable tool for cutting costs, improving quality, and boosting productivity. It’s one of the best

measures domestic manufacturers can use to remain competitive, provided they have product demand that justifies the cost.

The Client Team

The client team is critical to the success of any outsourced automation In some organizations, the client team is replaced with a sole project manager
project. The client team at a minimum should include a project manager and and a series of open project reviews. This project structure is much less efficient
knowledgeable representatives from each of the key stakeholder groups in the and can lead to sudden and chaotic changes in the project concept and
organization. This might include manufacturing or process engineering, direction. This happens, because most of the attendees are not versed in the
manufacturing, and quality engineering.

project requirements, or the supplier’s proposal, and tend to make assessments

based on a snap judgment of what they see before them. This can result in
For new products, a representative from product development or engineering sizable changes to the scope or requirements, and a loss of the work performed
should also be included. to date.

The Supplier Team Controls

Mechanical

Engineer Engineer
The supplier team is usually comprised of a project manager and any technical
contributors assigned to the team. For each technical discipline, there should be
one person identified as the lead contributor, who is responsible for continuity Pro ect
j

and quality for their area of expertise.


 Manager

This technical lead should be assigned to the project for the duration. 

Other resources may come and go as the project workload in each discipline Technician Programmer
changes through the development process.

Tools Don’t L e S eh l er
t tak o d s 


Milestones & Schedule

B e ate t t e a t
L o h P r y
A key project management tool is the project schedule. Generally, 

this should be constructed from a real and detailed estimate of the tasks to be A common problem in many engineering development projects is the
accomplished during the development process. This schedule 
 absence of select stakeholders at key points during the development
is generated by the Supplier Project Manager and should be made available to process, especially in the early conceptual stages of development. These
the entire development team, including the Client Team. stakeholders have valid interests in the performance, shape, and features
of the final project output, but fail to effectively participate in the review and
Task list
decision making process.
-

The task list is a simple tool for organizing and managing work on a near-term
Generally, this failure is just the unintentional result of busy schedules and
basis. The task list , maintained as a document or on a whiteboard, 

multiple priorities. owever, the effect on the development project can be
H
is a simple means for establishing and maintaining project momentum and
huge. e recommend that all stakeholders be required to ratify key
W
ensuring that issues are raised and handled in a timely fashion.


decisions and approve milestones through a written signoff.  

Task lists are best used for timelines of two weeks or less. They are also useful at
At the very least, requiring a written signoff at key milestones will focus
the end of the project, either during buy off or installation, to track and maintain
attention on the failure in the process after a failure occurs. t best, the
A
issues that need to be addressed to complete the project.
written signoff forces appropriate attention and diligence during the review
process, reducing the risk of a problem.

14725 SE 36 th treet uite

S S 300 B ellevue, WA 98006 | (425) 201-3713
Ensuring Success

Custom automation is a valuable tool for cutting costs, improving quality, and boosting productivity. It’s one of the best

measures domestic manufacturers can use to remain competitive, provided they have product demand that justifies the cost.

Trust, but Verify

Every unique automation system carries an element of technical risk. 


Most systems leverage a considerable body of prior art and off-the-shelf

equipment to help reduce this risk. However, every system includes one 


or more “experimental” elements that require particular care. These elements

may be a wholly new technology, or as is typically the case, a common

technology applied in a new way. 

In all cases, the intent should be to test early and often, and to reach a

knowledge point, good or bad, as quickly as possible. If the knowledge is good

then project can proceed as planned. If the knowledge is bad, then the problem

is identified early, and the corrective action will occur that much sooner, and with

that much less trouble.

“Ship No Wine Before It’s Time”

System buy off should always be conducted at the automation supplier’s Most machines ship with a small “punch list” of trivial items that need to be

site to the fullest extent possible. The client must verify that the system corrected before final acceptance. This list might include a missing safety

leaving the automation supplier meets or exceeds the specification. placard or a flow control for a cylinder that moves too quickly. The list

Shipping a machine before full compliance with all technical and safety should never include any software changes, electrical changes

requirements are demonstrated may lead to timely and expensive field or mechanical changes that require additional design work.

changes upon installation.

Road Testing Buy Off Checklist

Road testing generally involves running the machine over an extended period A buy off checklist is a good means of ensuring that each requirement of the

and monitoring performance. One needs to set reasonable goals for the system specification is met. It can also be used to ensure compliance with industry-

to demonstrate acceptable production rates, quality, and efficiency. We generally standard specifications. For best results, the buy off checklist should be

strive to achieve performance over a specified production run to demonstrate comprehensive, covering every item that will be checked during the buy off.

compliance.

It is also important that the checklist is provided to the supplier several weeks

Failure to achieve this goal requires correction of the problem, followed by a prior to buy off to ensure that the supplier has reviewed the equipment against

restart of the test from the beginning. A test is flawless when the system runs at the list, and made any changes necessary to bring the system into compliance. 

or above the specified production rate, produces no defects, and runs

The buy off list should only contain items pertaining to the pre-shipment buy off,

continuously without intervention, stoppages or other errors.

and should not contain items specific to the final installation, or to be performed

by the customer during installation.

14725 SE 36th Street Suite 300 Bellevue, WA 98006 | (425) 201-3713