PROCESS SELECTION, DESIGN, AND ANALYSIS

(Source: David Alan Collier, James Evans, and William Lindsay. Operations Management and
Total Quality Management. Cengage Learning Asia Pte Ltd (Philippines Branch). 2020.)

PROCESS CHOICE DECISIONS

Firms generally produce either in response to customer orders and demand or in anticipation of
them. This leads to three major types of goods and services: custom, option-oriented, and standard.
• Custom, or make-to-order, goods and services are generally produced and delivered as one
of a kind or in small quantities, and are designed to meet specific customers’ specifications.
Examples include ships, Internet sites, weddings, taxi service, estate plans, buildings, and
surgery. Because custom goods and services are produced on demand, the customer must
wait for them, often for a long-time because the goods or services must be designed,
created, and delivered.
• Options, or assemble-to order, goods and services are configurations of standard parts,
subassemblies, or services that can be selected by customers from a limited set. Common
examples are Dell computers, Subway sandwiches, machine tools, and travel agent
services. Although the customer chooses how the goods and services are configured, any
unique technical specifications or requirements cannot generally be accommodate.
• Standard, or make-to-stock, goods and services are made according to a fixed design, and
the customer has no options from which to choose. Appliances, shoes, sporting goods,
credit cards, online Web-based courses, and bus service are some examples. Standard
goods are made in anticipation of customer demand and stocked in inventory, and therefore
are usually available, although in some cases the proper color or size might be out of stock.

We note that manufacturing systems often use the terms make-to-order, assemble-to-order, and
make-to-stock to describe the types of systems used to manufacture goods. The terminology is not
as standardized in service industries, although the concepts are similar.

Four principal types of processes are used to produce goods and services:
1. Projects.
2. Job shop processes.
3. Flow shop processes
4. Continuous flow processes.

Projects are large-scale, customized initiatives that consist of many smaller tasks and activities that
must be coordinated to finish on time and within budget. Some examples of projects are legal
defense preparation, construction, and software development. Projects are often used for custom
goods and services, and occasionally for standardized products such as “market homes” that are
built from a standard design.

Job shop processes are organized around particular types of general-purpose equipment that are
flexible and capable of customizing work for individual customers. Job Shops produce a wide
variety of goods and services, often in small quantities. Thus, they are often used for custom or
option type products. In job shops, customer orders are generally processed in batches, and

1|P age
different orders may require a different sequence of processing steps and movement to different
work areas.

Flow shop processes are organized around a fixed sequence of activities and process steps, such
as an assembly line, to produce a limited variety of similar goods or services. An assembly line is
a common example of a flow shop process. Many large-volume, option-oriented, and standard
goods and services are produced in flow shop settings, Some common examples are automobiles,
appliances, insurance policies, checking account statements, and hospital laboratories. Flow shops
tend to use highly productive, specialized equipment and computer software.

Continuous flow processes create highly standardized goods or services, usually around the clock
in very high volumes. Examples of continuous flow processes are automated car washes, paper
and steel mills, paint factories, and many electronic, information-intensive services such as credit
card authorizations and security systems. The sequence of work tasks is very rigid and the
processes use highly specialized, automated equipment that is often controlled by computers with
minimal human oversight.

Exhibit below summarizes these different process types and their characteristics.

A product life cycle is a characterization of product growth, maturity, and decline over time. It is
important to understand product life cycle because when goods and services change and mature,
so must the processes and value chains that create and deliver them.

The traditional product life cycle (PLC) generally consists of four phases–introduction, growth,
maturity, and decline and turnaround. A product’s life cycle has important implications in terms
of process design and choice.

2|P age
 • For example,new products with low sales volume might be produced in a job shop process;
however, as sales grows and volume increases, a flow shop process might be more efficient.
As another example, a firm might introduce a standard product that is produced with a flow
shop process, but as the market matures, the product might become more customized. In
this case, a job shop process might be more advantageous.

What often happens in many firms is that product strategies change, but managers do not make the
necessary changes in the process to reflect the new product characteristics.

Two approaches to help understand the relationships between product characteristics for goods
and services and process choice decisions are the product-process matrix and service-positioning
matrix

THE PRODUCT-PROCESS MATRIX

The product-process matrix was first proposed by Hayes and Wheelwright and is shown in the
exhibit below. The product-process matrix is a model that describes the alignment of process
choice with the characteristics The most appropriate match between type of product and type of
process occurs along the diagonal in the product-process matrix. As one moves down the diagonal,
the emphasis on both product and process structure shifts from low volume and high flexibility to
higher volumes and more standardization. If product and process characteristics are not well
matched, the firm will be unable to achieve its competitive priorities effectively.

3|P age
For example, consider a firm that manufactures only a few products with high volumes and low
customization using a flow shop process structure. This process choice best matches the product
characteristics. However, suppose that as time goes on and customer needs evolve, marketing and
engineering functions develop more product options and add new products to the mix. This results
in a large number and variety of products to make, lower volumes, and increased customization.
The firm finds itself “off the diagonal” and in the lower left-hand corner of the matrix (denoted
by Position A in Exhibit 5.2). There is a mismatch between product characteristics and process
choice. If the firm continues to use the flow shop process, it may find itself struggling to meet
delivery promises and incur unnecessary costs because of low efficiencies.

On the other hand, by selectively and consciously positioning a business off the diagonal of the
product-process matrix (often called a “positioning strategy”), a company can differentiate itself
from its competitors. However, it must be careful not to get too far off the diagonal, or it must have
a market where high prices absorb any operational inefficiencies.

4|P age
 • For example, Rolls-Royce produces a small line of automobiles, using a process similar to
a job rather than the traditional flow shop of other automobile manufacturers. Each car
requires about 900 hours of labor. For Rolls-Royce this strategy has worked, but its target
market is willing to pay premium prices for premium quality and features.

The theory of the product-process matrix has been challenged by some who suggest that advanced
manufacturing technologies may allow firms to be successful even when they position themselves
off the diagonal. These new technologies provide manufacturers with the capability to be highly
flexible and produce lower volumes of products in greater varieties at lower costs. Therefore, off-
diagonal positioning strategies are becoming more and more viable for many organizations and
allow for “mass-customization” strategies and capabilities.

THE SERVICE-POSITIONING MATRIX

The product-process matrix does not transfer well to service businesses and processes. The
relationship between volume and process is not found in many service businesses.
• For example, to meet increased volume, service businesses such as retail outlets, banks,
and hotels have historically added capacity in the form of new stores, bench banks, and
hotels to meet demand but they do not change their processes. These limitations are
resolved by introducing the service-positioning matrix. To better understand it, we first
discuss the concept of a pathway in a service-delivery system.

A pathway is a unique route through a service system. Pathways can be customer driven or
provider driven, depending on the level of control that the service firm wants to ensure. Pathways
can be physical in nature, as in walking around Disney World or a golf course;procedural, as in
initiating a transaction via the telephone with a brokerage firm; or purely mental and virtual, as in
doing an Internet search.
• Customer-routed services are those that offer customers broad freedom to select the
pathways that are best suited for their immediate needs and wants from many possible
pathways through the service delivery system. The customer decides what path to take
through the service-delivery system with only minimal guidance from management.
Searching the Internet to purchase an item or visiting a park are examples.
• Provider-routed services constrain customers to follow a very small number of possible
and pre-defined pathways through the service system. An automatic teller machine (ATM)
is an example. A limited number of pathways exist–for example, getting cash, making a
deposit, checking an account balance, and moving money from one account to another.
Mailing and processing a package using the U.S. Postal Service, Federal Express, or UPS
is another example of a provider-routed service.

Designs for customer-routed services require a solid understanding of the features that can delight
customers, as well as methods to educate customers about the variety of pathways that may exist
and how to select and navigate through them.

5|P age
The service-positioning matrix (SPM), as shown in the exhibit above, is roughly analogous to the
product-process matrix for manufacturing. The SPM focuses on the service-encounter level and
helps management design a service system that best meets the technical and behavioral needs of
customers. The position along the horizontal axis is described by the sequence of service
encounters. The service-encounter activity sequence consists of all the process steps and associated
service encounters necessary to complete a service transaction and fulfill a customer’s wants and
needs. It depends on two things:

1. The degree of customer discretion, freedom, and decision-making power in selecting the
service-encounter activity sequence. Customers may want the opportunity to design their
own unique service-encounter activity sequence, in any order they choose.
2. The degree of repeatability of the service-encounter activity sequence. Service-encounter
repeatability refers to the frequency that a specific service-encounter activity sequence is
used by customers. Service-encounter repeatability provides a measure analogous to
product volume for goods-producing firms.

The more unique the service encounter, the less repeatable it is.
• A high degree of repeatability encourages standardized process and equipment design and
dedicated service channels, and results in lower costs and improved efficiency.
• A low degree of repeatability encourages more customization and more flexible equipment
and process designs, and typically results in higher relative cost per transaction and lower
efficiency.

The position along the vertical axis of the SPM reflects the number of pathways built into the
service system design by management. That is, the designers or management predefine exactly

6|P age
how many pathways will be possible for the customer to select, ranging from one to an infinite
number of pathways.

The SPM is similar to the product-process matrix in that it suggests that the nature of the
customer’s desired service-encounter activity sequence should lead to the most appropriate service
system design and that superior performance results by generally staying along the diagonal of the
matrix. Like the product-process matrix, organizations that venture too far off the diagonal create
a mismatch between service system characteristics and desired activity sequence characteristics.
As we move down the diagonal of the SPM, the service-encounter activity sequence becomes less
unique and more repeatable with fewer pathways. Like the product-process matrix, the midrange
portion of the matrix contains a broad range of intermediate design choices.

PROCESS DESIGN

The goal of process design is to create the right combination of equipment, labor, software, work
methods, and environment to produce and deliver goods and services that satisfy both internal and
external customer requirements.
• Process design can have a significant impact on cost (and hence profitability), flexibility
(the ability to produce the right types and amounts of products as customer demand or
preferences change), and the quality of the output.

We can think about work at four hierarchical levels:

1. Task
2. Activity
3. Process
4. Value chain

A task is a specific unit of work required to create an output.

• Examples are inserting a circuit board into an iPad subassembly or typing the address on
an invoice.
An activity is a group of tasks needed to create and deliver an intermediate or final output.
• Examples include all the tasks necessary to build an iPad; for example, connecting the
battery and assembling the cover pieces, or inputting all the information correctly on an
invoice, such as the items ordered, prices, discounts, and so on.
An example of a process would be manufacturing an iPad or fulfilling a customer order.
The value chain for an iPad would include acquiring the materials and components, manufacturing
and assembly, distribution, retail sales, and face-to-face and Web-based customer support.

7|P age
The exhibit above shows an example for the production of antacid tablets. The value chain shows
an aggregate view focused on the goods-producing processes (supporting services such as
engineering, shipping, accounts payable, advertising, and retailing are not shown). The next third
level in the hierarchy of work is at the production process level where tablets are made. The third
level focuses on the mixing workstation (or work activities) where the ingredients are unloaded
into mixers. the mixer must be set up for each batch and cleaned for the next batch because many
different flavors, such as peppermint, strawberry-banana, cherry, and mandarin orange, are
produced using the same mixers. The fourth and final level in the work hierarchy is the flavoring
tasks, which are defined as three tasks, each with specific procedures, standard times per task, and
labor requirements. These three tasks could be broken down into even more detail if required.

Process and Value Stream Mapping

Understanding process design objectives focuses on answering the question: What is the process
intended to accomplish? An example process objective might be “to create and deliver the output
to the customer in 48 hours. “ Another key question to consider is: What are the critical customer
and organizational requirements that must be achieved?

Designing a goods-producing or service-providing process requires six major activities:

1. Define the purpose and objectives of the process,
2. Create a detailed process or value stream map that describes how the process is currently
performed (sometimes called a current state or baseline map). Of course, if you are
designing an entirely new process, this step is skipped.

8|P age
 3. Evaluate alternative process designs. That is, create process or value stream maps
(sometimes called future state maps) that describe how the process can best achieve
customer and organizational objectives.
4. Identify and define appropriate performance measures for the process.
5. Select the appropriate equipment and technology
6. Develop an implementation plan to introduce the new or revised process design.

A process map (flowchart) describes the sequence of all process activities and tasks necessary to
create and deliver a desired output or outcome. It documents how work either is or should be
accomplished, and how the transformation process creates value. We usually first develop a
“baseline” map of how the current process operates in order to understand it and identify
improvements for redesign.

Process maps delineate the boundaries of a process. A process boundary is the beginning or end
of a process. The advantages of a clearly defined process boundary are that it makes it easier to
obtain senior management support, assign process ownership to individuals or teams, identify key
interfaces with internal or external customers, and identify where performance measurements
should be taken. Thus, each of the levels in the exhibit (Antacid) represents a process map defining
different process boundaries.

Typical symbols used for process maps are the following:

The exhibit below shows a flowchart for an automobile repair process. Process maps clearly
delineate the process boundaries.

9|P age
In service applications, flowcharts generally highlight the points of contact with the customer and
are often called service blue-prints or service maps. Such flowcharts often show the separation
between the back office and the front office with a “line of customer visibility,” such as the one
shown in the exhibit above.

Non-value-added activities such as transferring materials between two nonadjacent workstations,

waiting for service, or requiring multiple approvals for a low-cost electronic transaction simply
lengthen processing time, increase costs,and often increase customer frustration. Eliminating non-
value-added activities in a process design is one of the most important responsibilities of
operations managers. This is often accomplished using value stream mapping, a variant of more
generic process mapping.

The value stream refers to all value-added activities involved in designing, producing, and
delivering goods and services to customers. A value stream map (VSM) shows the process flows
in a manner similar to an ordinary process map; the difference lies in that value stream maps

10 | P a g e
highlight value-added versus non-value added activities and include costs associated with work
activities for both value- and non-value added activities.

To illustrate this, consider a process map for the order fulfillment process in a restaurant, as shown
in the exhibit below. From the times on the process map, the “service standard” order posting and
fulfillment time is an average of 30 minutes per order (5+1+4+12+3+5). The restaurant’s service
guarantee requires that if the order posting and fulfillment time is more than 40 minutes, the
customer’s order is free of charge.

The chef’s time is valued at $30 per hour, oven operation at $10 per hour, precooking order waiting
time at $5 per hour, and post-cooking order waiting time at $60 per hour. The $60 estimate reflects
the cost of poor quality for a dinner waiting too long that might be delivered to the customer late
(and cold).

The exhibit below illustrates a value stream map for the order posting and fulfillment process in
the exhibit above. Exhibit 5.7 is one of many formats for value stream mapping. Here, non-value-
added time is 33.3 percent (10/30 minutes) of the total order posting and fulfillment time, and the
non-value-added cost is 31.7 percent ($5.417/$17.087) of total cost. Suppose that a process
improvement incorporates wireless technology to transmit food orders to the kitchen and notify
the waiter from 10 minutes to 4 minutes on the front and back ends of the process. Hence, the total
processing time is reduced from 30 to 24 minutes (a 20 percent improvement). Costs are reduced
by $3.25 with a three-minute wait reduction on the front and back ends of the process. Therefore,
cost per order goes from$17.087 to $13.837 (a 19 percent improvement). Increasing the speed of
this part of the restaurant delivery process may also allow for a higher seat turnover during peak
demand periods, thus helping to increase total revenue and contribute to profit and overhead.

11 | P a g e
PROCESS ANALYSIS AND IMPROVEMENT

Few processes are designed from scratch. Many process design activities involve redesigning an
existing process to improve performance. Management strategies to improve process designs
usually focus on one or more of the following:
• increasing revenue by improving process efficiency in creating goods and services and
delivery of the customer benefit package;
• increasing agility by improving flexibility and response to changes in demand and
customer expectations;
• increasing product and/or service quality by reducing defects, mistakes, failures, or service
upsets;
• decreasing cost through better technology or elimination of non-value-added activities;

12 | P a g e
 • decreasing process flow time by reducing waiting time or speeding up movement through
the process and value chain; and
• decreasing the carbon footprint of the task, activity, process, and/or value chain.

Process and value stream maps are the foundation for improvement activities. Typical questions
that need to be evaluated during process analysis include:
• Are the steps in the process arranged in logical sequence?
• Do all steps add value? Can some steps be eliminated, and should others be added in order
to improve quality or operational performance? Can some be combined? Should some be
reordered?
• Are capacities of each step in balance; that is, do bottlenecks exist for which customers
will incur excessive waiting time?
• What skills, equipment, and tools are required to each step of the process? Should some
steps be automated?
• At which points in the system (sometimes called process fail points) might errors occur
that would result in customer dissatisfaction, and how might these errors be corrected?
• At which point or points in the process should performance be measured? What are
appropriate measures?
• Where interaction with the customer occurs, what procedures, behaviors, and guidelines
should employees follow that will present a positive image?
• What is the impact of the process on sustainability? Can we qualify the carbon footprint of
the current process?

Sometimes, processes grow so complex that it is easier to start from a “clean sheet” rather than try
to improve incrementally. Reengineering has been defined as “the fundamental rethinking and
radical redesign of business processes to achieve dramatic improvements in critical, contemporary
measures of performance, such as cost, quality, service, and speed.”

Reengineering was spawned by the revolution in information technology and involves asking basic
questions about business processes: Why do we do it? Why is it done this way? Such questioning
often uncovers obsolete, erroneous, or inappropriate assumptions. Radical redesign involves
tossing out existing procedures and reinventing the process, not just incrementally improving it.
The goal is to achieve quantum leaps in performance. All processes and functional areas participate
in reengineering efforts, each requiring knowledge and skills in operations management.

PROCESS DESIGN AND RESOURCE UTILIZATION

Idle machines, trucks, people, computers, warehouse space, and other resources used in a process
simply drain away potential profit. Utilization is the fraction of time a workstation or individual is
busy over the long run. It is difficult to achieve 100 percent utilization.
• For example, utilization in most job shops ranges from 65 to 90 percent. In flow shops, it
might be between 80 and 95 percent, and for most continuous flow processes, above 95
percent. Job shops require frequent machine changeovers and delays, whereas flow shops
and continuous flow processes keep equipment more fully utilized. Service facilities have
a greater range of resource utilization. Movie theaters, for example, average 5 to 20 percent

13 | P a g e
 utilization when seat utilization is computed over the entire week. Similar ranges apply to
hotels, airlines, and other services.

Two ways of computing resource utilization are:

Utilization = Resources Used/ Resources Available

Utilization = Demand Rate/ Service Rate x Number of Servers

The first equation, the measurement base (time, units, etc.) must be the same in the numerator and
denominator. For a process design to be feasible, the calculated utilization over the long run cannot
exceed 100 percent. However, over short periods of time, it is quite possible that demand for a
resource will exceed its availability. If a manager knows any three of the four variables in the
second equation, then the fourth can be easily found.

The two equations can provide useful insight for evaluating alternative process designs. The
exhibit above provides an analysis of the utilization of the restaurant order posting and fulfillment
process in the exhibit (restaurant order). Using the second equation, the resource utilization for
work activity #3, assuming only one chef and two ovens, is computed as:

(20 orders/hour) ([5 orders/ hour)

(1 chef) = 4.0 or 400 percent

As we noted earlier, whenever the utilization is calculated to be greater than 100 percent, the work
will endlessly pile up before the workstations. This is clearly a poor process design, and more
resources needed.

A logical question to consider is how many chefs are needed to bring the utilization down below
100 percent at work activity #3? Because the chef is the most skilled and highest-paid employee,
it would make sense to design the process so that the chef would have the highest labor utilization

14 | P a g e
rate (although 100 percent would probably not be practical). This can be found by solving the
second equation as follows:

(20 orders/hour) [(5 orders/ hour)(x chefs) = 1.00

(20 orders/hour) x 1.00 x x = 20 orders/hour
or x = 4.00 chefs

With four chefs, the resource utilizations are recomputed in the exhibit above. We see that the oven
is still a problem, with a calculated 167 percent utilization. To determine how many ovens to have
for ma 100 percent utilization, we solve the equation:

(20 orders/hour) [(5 orders/ hour)(x chefs) = 1.00

(20 orders/hour) x 1.00 x x = 20 orders/hour
or Y = 3.33 chefs

Rounding this up to 4, actual oven utilization would now be 83 percent (see Exhibit below for the
final results).

The exhibit below shows a simplified flowchart of the order fulfillment process along with the
output rates from exhibit above that can be achieved for each work activity. The average number
of entities completed per unit time–the output rate–from a process is called throughput.

15 | P a g e
Throughput might be measured as parts per day, transactions per minute, or customers per hour,
depending on the context.

A logical question to consider is what throughput can be achieved for the entire process. A
bottleneck is the work activity that effectively limits the throughput of the entire process. A
bottleneck is like the weakest link of a chain; the process in exhibit above can never produce more
than 20 orders/hour–the output rate of work activity #3 (assuming all five activities must be
completed). Identifying and breaking process bottlenecks is an important part of process design
and improvement, and will increase the speed of the process, reduce waiting and work-in-process
inventory, and use resources more efficiently.

Little’s Law

At any moment, people, orders, jobs, documents, money, and other entities that flow through
processes are in various stages of completion and may be waiting in queues. Flow time, or cycle
time, is the average time it takes to complete one cycle of a process. It makes sense that the flow
time will depend not only on the actual time to perform the tasks required but also on how many
other entities are in the work-in-process stage.

Little’s Law is a simple formula that explains the relationship among flow time (T), throughput ®,
and work-in-process (WIP):

Work-in-process = Throughput X Flow Time

or

WIP = R x T

Little’s Law provides a simple way of evaluating average process performance. If we know any
two of the three variables, we can compute the third using Little’s Law . Little’s Law can be applied
to many different types of manufacturing and service operations. (See the accompanying Solved
Problems.)

Suppose that a voting facility processes an average of 50 people per hour (thoughput) and that, on
average, it takes 10 minutes for each person to complete the voting process (flow time). Using the
equation, we can compute the average number of voters in the process (work in process):

16 | P a g e
Work in process (WIP) = R x T
= 50 voters/hr x (10 minutes/60 minutes per hour)
= 8.33 voters

It is important to understand that Little’s Law is based on simple averages for all variables. Such
an analysis serves as a good baseline for understanding process performance on an aggregate basis,
but it does not take into account any randomness in arrivals or service times, or different
probability distributions.

17 | P a g e