MODULE DMQ40403 PRODUCTION PLANNING AND CONTROL

MODULE
DMQ40403

PRODUCTION PLANNING
AND CONTROL
JUNE 2021

1

CHAPTER 1 : OVERVIEW OF PRODUCTION
PLANNING & CONTROL

INTRODUCTION

This chapter introduces the nature of planning and control as it has evolved and is in use in
many organisation today, and also discusses the use and implementation of the fundamental
principles of planning and control systems. Virtually every organisation (large, small,
manufacturing, service, for profit or not for profit) has as its central function the production of
some defined output from its processes. In order for that organisation to be effective and
efficient in serving its customers, the managers of the organisation must understand and
apply certain fundamental principles of planning for the production of the output and also
controlling the process producing the output as it is being produced. The chapter is to
identify and explain those fundamental principles. While the planning and control
approaches discussed in this subject are most commonly used in manufacturing companies,
many are used or have been adapted for use in services companies. Those differences in
operations leading to different uses are discussed, as are several of the environmental
issues that heavily influence the design and use of the appraoches to planning and control
that are selected.

LEARNING OBJECTIVES

The objectives of this unit are to :
1. explain the important of production planning and control in manufacturing and service

operations.
2. explain the customer influence in design of production environment system.
3. explain the five (5) types categories of process in manufacturing operation.
4. explain the concept of order winner and qualifier in business environment.
5. explain the SWOT analysis and the usage in business operation.
6. discuss the business environment issues and the common solution approaches used

by industrialist.

2

1.1 MANUFACTURING VERSUS SERVICE OPERATIONS

While the major focus for this subject is manufacturing, the same principles also
apply (in many cases) to service organisations. Service organisations are, of course,
those organisations whose primary outputs are not manufactured goods, but instead
sevices to individuals. Legal services, accounting services, banking, insurance, and
haircutting are all examples of “production” outputs that are services. There are
clearly some major differences between a service and manufacturing environment,
and these differences do impact the formality and approach taken in the application
of these principles, but often the principles do still apply. This subject approaches the
explaination of the principles in their most formal and structured application, which
tend to reflect the manufacturing environment. Where applications can be applied in
service settings, an attempt is made to describe those application as well. To that
extent, this subject applies to both manufacturing and service operations. It is
interesting to note in this discussion that as service organisations become larger and
have many “branches”, such as banks, that some services (particularly the “home
offices” of banks, insurance companies, etc.) have been to organise to take
advantage of some of the efficiencies of a typical manufacturing environment. These
cases are sometimes called “quasimanufacturing” organisations.

To some extent the service organisation’s approach to planning and control is more
dificult to manage, for at least four major reasons. It is these four issues that
generally provide the major influence on the way that planning and control
approaches are designed for service organisations:

a. Timing: In service organisations there is often little time between the recognition
of demand and the expected delivery of the process output. Customers enter
some service establishments and expect almost instantaneous delivery of the
output. Service organisations often attempt to control this to some some extent,
especially if the capacity to deliver the service is relatively fixed and/or very
costly. Appointments and reservations in some service establishments are
examples of how they attempt to control the demand for process output.

b. Customer Contact: Related to the issue of timing is the fact that the customer in a
service environment is often much more involved in the design of the “product” or
output of the experience. In addition, the contact point is often the person who will
be delivering the service. In that respect the service worker can be thought of a
sales person and an operations worker.

3

c. Quality: A key dimension of quality in service organisations is that much of the
quality may be intangible, making it much more difficult to effectively measure.

d. Inventory: “Pure” service organisations (those that have virtually no physical
goods involved in their output) often do not have the luxury of inventorying their
output. It is imposible, for example, to inventory a haircut. Many people in
manufacturing may be taken aback by the image of inventory as a luxury, given
that they are often pressed for inventory reduction, but in fact inventory in the
perspective of manufacturing planning can be thought of as “store capacity”.
Essentially, inventory (especially finished goods) can be viewed as the
application of the organisation’s capacity prior to the actual demand for that
output.

1.2 CUSTOMER INFLUENCE IN DESIGN: PRODUCTION
ENVIRONMENTAL CHOICES

The design of the planning and control system will be impacted by several factors in
addition to the points mentioned above. Among the most critical of these factors are
the volume an dvariety of the expected output, and those issues in turn tend to be
largely driven by the maount of influence the customer has in the design of the
product or service delivered to them from organisation’s processes. In some cases
the issue of customer design influence is a part of the organisation’s basic strategy,
but in some cases it is a reaction to market drivers. Many automobiles, for example,
are purchased as finished goods from a dealer’s lot primarily because the customers
do not wish to wait for an automobile that is ordered with the exact options they want.
That extent of customer influence tends to be described by the following categories,
listed here in order of influence, from less to more:

a. Make-To-Stock (MTS): As the name implies, these are products that are
completely made into their final form and stocked as finished goods. The
collective customer base may have some influence on the overall design in the
early product design phase, but an individual customer has essentially only one
decision to make once the product is made to purchase or not to purchase.
Again, these purchase patterns can influence overall product design changes, but
not usually in the case of an individual customer. Examples of these products are

4

very common, as found in virtuall all retail stores such as hardware, clothing,
office supplies, and so on.

b. Assemble To Order (ATO): In this case the customer has some more influence
on the design, in that they can often select various options from predesigned
subassemblies. The producer will then assemble these options into the final
product for the cutomer. As in the case of the MTS, the collective customer base
can influence the overall design of the options and final products, but the
individual customer can only select from the specified options. Automobiles and
personal computers are good examples of these types of products. If a customer
orders an automobile from dealer, for example, they can often choose from
variety of colours, body stlyes, engines, transmissions, and other “pure” options,
such as cruise control.

In some industries this approach is sometimes called Package to order, in that it
is the pakaging that is customer specified. In products such as breakfast cereal or
baking products (flour, baking soda, etc.), the product does not change, but can
be placed in several diffreent sizes and types of packages according to customer
need. A service example of ATO may be in some restuarants, where the
customer can specify their choice of side dishes for their meal. They may have
little option as to how those side dishes are prepared, but do have some say in
which ones they select.

c. Make-To-Order (MTO): This environment allows the customer to specify the
exact design of the final or service, as long as they use standard raw materials
and components. An example might be a specialty furniture maker or a bakery. In
the bakery, for example, a customer may specify a special cake be produced for
an occasion such as a birthday or anniversary. They have many design options
for the cake pans, cake flavourings, and so on.

d. Engineer-To-Order (ETO): In this case the customer has almost complete say in
the design of the product or service. They are often not even limited to the use of
standard components or raw material, but can have the producer deliver
something designed “from scratch.”

5

1.3 PROCESS CATEGORIES

The nature of the customer influence issue described above not only impacts the
design of the product or service, but also has a profound impact on the design of the
process used to deliver the product or service. There are essentially five categories
given to describe the process used in production, although in practice there are
several combinations of these basic types. The five categories typically given are:

a. Project: A project-based process typically assumes a one-of-a-kind production
output, such as building a new building or developing a new software application.
Projects are typically large in scope and will often be managed by teams of
individuals brought together for this one-time activity based on their particular
skills.

b. Job Process: Job processes (job shop processes) are typically designed for
flexibility. The equipment is often general purpose, meaning it can be used for
many different production requirements. The skill in delivering the production as
specified by the customer is generally focused on the workers, who tend to be
highly skilled in a job shop process. This environment is generally focused for
production requirements, as may be found in an ETO or MTO design
environment. The high variety of design requires the flexible processes and
higher skills of the workforce. Work in these environments will often move in a
very “jumbled” fashion because of the high variability in designs for each job.
Again because of the variability in design and work requirements information
linkages tend to be informal and loose. An example is general-purpose machine
shop or specialty bakery or caterer.

c. Batch or Intermittent Processing: Many of the production facilities in the world
today fall into this “middle of the road” category. The equipment tends to be more
specialised than the equipment in the job shops, but still flexible enough to
produce some variety in design. As more of the “skill” to produce the product
rests in the more specialised equipment, the workers do not usually need to be
quite as skilled as the workers in the job shops. Often these organisations are
organised with homogeneous groupings of worker skills and machines, forcing
the work to move from area to area as it is being processed. The catagory is
often called batch since products are often made in discrete batches. For
example, a batch process may make several hundred of one model of product,
taking many hours before they switch the setup to produce another MTO and

6

some MTS, but this environment is usally well suited to the ATO environment.
There are many examples of products built in this environment, including clothing,
bicycles, furniture, and so on.

d. Repetitive or Flow Processing: As the name implies, this type of process facility
tends to be used for a very large volume of a very narrow range of designs. The
equipment tends to be highly specialised and expensive, requiring little labour,
and the labour that is used tends to be unskilled. The expense of the special
equipment is placed into the overhead cost category, allowing the relatively fixed
cost to be spread over a large volume. This makes the cost per item lower,
making it price competitive. Repetitive processing is typically used for make to
stock (MTS) designs, such as refrigerators and other appliances.

e. Continuous: As with project processing, this type of process is at the far extreme
of the processing types, again making it focused on highly specialised
applications. The equipment is very specialised, and little labour tends to
beneeded. High volume chemical processes and petroleum refining will fall into
this category.

Table 1.1: Summary Of Process Categories

Job process Batch Repetitive

Equipment General purpose Semi-specialised Highly-specialised
Labour skills
Managerial Highly skilled Semi skilled Low skills
approach Efficiency-keep
Technical problem
Volume output Team leadership the process
per design moving
Variety of solver
designs
produced Low Medium High
Design
environment High Medium Low
Flow of work
ETO, MTO MTO, ATO, MTS ATO, MTS

Variable, jumbled More defined Highly defined and
fixed

7

While these are the common types, it should be noted that some products are
produced in “hybrid” operation, which can be thought of as combinations of these
common types. For example, some chemicals might be produced in a continuous
process, but then packaged in a batch environment. Table 1.1 summarises some of
the key points and differences between the middle three types of processes: job
process, batch, and repetitive.

In addition, there are several implications for planning and control that need to be
highly specialised and different across these types of processing environments.
Virtually all aspects of planning and control will be impacted depending on the type of
production environment.

An easy way to illustrate the differences in volume and variety relating to various
process types was developed several years ago by Robert Hayes and Steven
Wheelright, often called the Hayes-Wheelright Product / Process Matrix. As can be
seen in axample of the matrix in Figure 1.1, the horizontal axis shows the range of
processes, from those with general-purpose machines with variable flow to those
with fixed flow. The diagonal shows the optimal type of processing that is usually
used for each type of product.

High Variety, Product Low variety,
Low Volume mix high volume

Erratic flow, Job Shop Additional
loose linkages Batch Cost

Process Additional Line Flow
Pattern Cost Continuous

Continuous,
linked, rigid

flow

Figure 1.1: The Hayes-Wheelwright Matrix

8

It should be noted that producing a product or service off the diagonal is not
impossible, just often not wise from a business perspective. It is not that one cannot
produce off the diagonal, but more that one should not. An example may illustrate.
Take the example of a quarter-pound beefburger produced in a fast-food restuarant.
That would fall into the lower right-hand quadrant of the matrix, in that it is a low-
variety, high-volume product typically produced in a fairly rigid, repetitive process in a
fast-food restuarant. Now the question could be asked if a fancy, gourmet restuarant
could also produce such a beefburger. Clearly they would have both the equipment
and skills to produce such a product but to do so would place them in the upper right
hand quadrant. The additional cost in this case is represented by the highly skilled
and expensive labour would be better utilised to produce the product, but could not
compete well in the price-sensitive market that typifies such a standard high-volume
product.

To explain the other “off-diagonal” (lower left-hand part of the matrix), we could ask
the question “can a typical fast food restuarant produce a fancy prime rib dinner?”
The answer is possibly yes, but to do so would clearly take an investment in
equipment and training for the employees. Therefore, it may be posiible, but not
without extensive extra costs.

1.4 ORDER WINNER AND QUALIFIERS

Another aspect of the business environment that will impact the design and
management of the planning and control system is the market dirvers for the product
or service. To start this discussion, it first must be recognised that there are several
dimensions by which customers in the market may evaluate the desirability of buying
a certain product or service from a given producer. Some of the more important
dimensions of competition include:

a. Price: Usually related to cost of product or service. There are two primary types of
price categories:
i. Standard price, such as a catalog price.
ii. Custom pricing, usually negotiated.

9

b. Quality: There are two major aspects to consider.
i. Tangible quality, including those aspects for which specific measures can
often be developed. These can include standard quality measures such
as conformance, reliability, and durability.
ii. Intangible quality, including those aspects that may prove of value to the
customer, yet are difficult to specifically measure. They may include such
aspects as reputation (brand), aeshetics, responsiveness, and customer
service.

c. Delivery: Again, there are two major aspects:
i. Speed: how quickly can the product or service be delivered?
ii. Reliability: once a promise for delivery is made, is it kept?

d. Flexibility: Two major issues must be considered:
i. Volume: can the producer easily produce a wide range of product
volumes?
ii. Variety: can the producer easily produce a wide range of product designs
and/or options?

It should be noted that these four major dimensions on this list are major issues for
the operation function within the organisation. There are issues that tend to be
heavily shared by nonproduction functional areas of responsibility, such as marketing
and engineering. A prime example is the issue of intangible quality, many aspects of
which are often the responsibility of functions other than direct production.

It also should be noted that it is virtually imposibble for any one producer to be the
“best” in the market for all these dimensions of competition at the same time. As part
of the development of the operations strategy of the organisation, the producer must
determine which of these dimensions represent order winners and which are order
qualifiers for their market as defined by the corporate startegy.

a. Order Qualifiers: Order qualifiers represent the dimension by which a potential
customer determines which suppliers of a product or service meet certain criteria
to be considered for receiving the final order from the customer. Qualifiers only
allow consideration, and meeting the order-qualifying criteria does not
neccessarily mean the supplier will be successful winning the order. Not meeting
the criteria, on the other hand, will almost ensure the order will not be won.

10

b. Order Winners: Once potential uppliers have been evaluated as to their order-
qualifying criteria, the final successful supplier for the product or service is
selected based on the order-winning criteria in the mind of the customer.

As an example, suppose a person is in the market for a basic colour television. They
may first check out producers that have a reputation for quality and reliability (order
qualifiers). They may then look at sample pictures and products from the producers
that qualify on the quality and reliability perspective, and further reduce the possible
products based on the features and basic look of the television (another qualifier).
Finally, they may actually purchase the television from the possible qualifying
products based on price (the order winner).

1.5 SWOT ANALYSIS

SWOT Analysis, or sometimes known as the TOWS Matrix, is a strategic planning
tool used to evaluate the Strengths, Weaknesses, Opportunities, and Threats
involved in a project or in a business venture or in any other situation of an
organization or individual requiring a decision in pursuit of an objective. It involves
monitoring the marketing environment internal and external to the organization or
individual.

If a clear objective has been identified, SWOT analysis can be used to help in the
pursuit of that objective. In this case, SWOTs are (see figure):

1. Strengths: attributes of the organization that are helpful to achieving the
objective.

2. Weaknesses: attributes of the organization that are harmful to achieving the
objective.

3. Opportunities: external conditions that are helpful to achieving the objective.
4. Threats: external conditions that are harmful to achieving the objective.

11

INTERNAL HELPFUL HARMFUL
to achieving the to achieving the
(attribute of the
organisation) objective objective

EXTERNAL STRENGTHS WEAKNESSES

(Attribute of the OPPORTUNITIES THREATS
environment)

Figure 1.2: SWOT analysis Matrix

Ideally a cross-functional team or a task force that represents a broad range of
perspectives should carry out the SWOT analysis. For example, a SWOT team may
include an accountant, a salesperson, an executive manager and an engineer.

The aim of any SWOT analysis is to identify the key internal and external factors that
are important to achieving the objective. SWOT analysis groups key pieces of
information into two main categories:

1. Internal factors - The strengths and weaknesses internal to the organization.
2. External factors - The opportunities and threats presented by the external

environment.

The internal factors may be viewed as strengths or weaknesses depending upon
their impact on the organization's objectives. What may represent strengths with
respect to one objective may be weaknesses for another objective. The factors may
include all of the 4P's; as well as personnel, finance, manufacturing capabilities, and
so on. The external factors may include macroeconomic matters, technological
change, legislation, and socio-cultural changes, as well as changes in the
marketplace or competitive position. The results are often presented in the form of a
matrix.

SWOT analysis is just one method of categorization and has its own weaknesses.
For example, it may tend to persuade companies to compile lists rather than think
about what is really important in achieving objectives. It also presents the resulting

12

lists uncritically and without clear prioritization so that, for example, weak
opportunities may appear to balance strong threats.

It is prudent not to eliminate too quickly any candidate SWOT entry. The importance
of individual SWOTs will be revealed by the value of the strategies it generates. A
SWOT item that produces valuable strategies is important. A SWOT item that
generates no strategies is not important

1.6 TOWS MATRIX

The TOWS Matrix is derived from the SWOT Analysis model, which stands for the
internal Strengths and Weaknesses of an organization and the external Opportunities
and Threats that the business is confronted with.
The acronym TOWS is a variant of this and was developed by the American
international business professor Heinz Weirich.
The TOWS Matrix is not just meant for the highest levels of management in an
organization. It can be a very useful tool for departments (i.e. a marketing or sales
team) or for individual employees on an operational level. Once it’s employees or a
department’s strengths are known, these can be improved further to become even
better. The TOWS Matrix emphasizes the external environment.

It starts by analyzing external opportunities and threats. Up next are the internal
strengths and weaknesses, which will subsequently be linked to the external
analysis. And this is where it goes a step beyond the traditional SWOT Analysis;
strategic tactics emerge by opposing S-O (Strengths-Opportunities), W-
O (Weaknesses-Opportunities), S-T (Strengths-Threats) and W-T (Weaknesses-
Threats).

A next step in the analysis helps when thinking about the option they want to pursue.
Here the external opportunities and threats are compared to the internal strengths
and weaknesses to help identify strategic options:

1. Internal Strengths and External Opportunities (S-O) – how can they use the
strengths to benefit from existing external opportunities?

2. Internal Strengths and External Threats (S-T) – how can they benefit from
their strengths to avoid or lessen (potential) external threats?

3. Internal Weaknesses and External Opportunities (W-O) – how can they use
opportunities to overcome the organization’s internal weaknesses?

4. Internal Weaknesses and External Threats (W-T) – how can they minimize
weaknesses and thus avoid potential threats?

13

Figure 1.3: TOWS Matrix

Figure 1.4: Example of TOWS Matrix

14

The above-mentioned factors can be linked to each other, leading to strategies:

S-O – How can the organization employ the expertise of its own professionals to
respond to the needs of vocational education centers? By partnering up, the
organization can convince the vocational education centers that there is enough
capacity, knowledge and experience to train young people to independent
professionals at all levels of vocational education.

S-T – How can the organization use its skilled staff to compete with cheaper workers
employed by competitors? A smart approach for the organization would be to
communicate to the outside world that their staff has accredited diplomas and that it’s
important for housing co-operatives to comply with legal requirements and safety
standards.

W-O – How can partnerships with vocational education centers help the organization
to improve itself and put more effort into customer acquisition? By presenting itself as
an accredited apprenticeship provider, the organization will put itself on the market
again and its shows that adapt to changing times and wants to offer different kinds of
maintenance to businesses and housing co-operatives.

W-T – How can the organization better position itself in the market and thus reduce
the threat posed by competitors? By presenting itself as an accredited apprenticeship
provider, the organization can claim that they are a serious competitor and can
possibly offer maintenance services by apprentices at reduced rates, with the work
still being done by an accredited company.

1.7 BUSINESS ENVIRONMENT ISSUES

A major message from the discussion above is that it is critical for a producer to
understand their market(s) and design their systems to at least meet the minimum
level of criteria for the order qualifiers in their market, but at the same time strive to
be the best in those dimensions that represent order winners. While that may appear
on the surface to be a fairly basic and simple approach, there are a number of
complicating issues. They include:

a. Customer “Learning”: Competitors often attempt to approach the market in the
same way (emphasizing the same competitive dimensions) as each other, but
from time to time a competitor may attempt to gain market share by emphasizing

15

they ate the “best” at it. As this happens, the customer expectation may also
change. For example, if delivery speed is an order winner, as producers change
their system to improve delivery time, continually “raising the bar” for all
companies in the market.

b. Competitor Moves: Some competitor moves may disqualify order winners, turning
them into qualifiers, and thereby establishing new order winners. For example,
suppose an order winner in the market has been price. The competitors have
been working hard to cut costs, thereby allowing lower prices to be charged.
Suppose all the competitors have developed cost controls to now charge almost
equal prices to the point where customers perceive very little difference. In such a
market, they may become sensitive to another order winner, such as delivery
speed. If all competitors have roughly the same price, but one has a much faster
delivery, then the order winner may now become delivery speed, leaving price as
a qualifier. Effective marketing and advertising plans can also sometimes change
customer perceptions as to what is important as an order-winning dimension.

c. Multiple Markets: It is likely that most companies have numerous products or
services serving numerous markets. In such cases, there may be many different
order qualifiers in many different markets and all may be subjected to the
changes described in the first two points. The effective producer needs to be
aware of and continually monitor all the markets and the company planning and
control systems need to effectively support all.

d. Product Design Changes: New products and changes in product design,
especially as technology impacts customer expectations, will also often change
order winners and qualifiers. A good example of this is how Internet technology
has altered customer perception of how to purchase many goods and services.

SUMMARY

In this chapter we have studied that

1. There are several of the key environmental and organisational drivers that are used
by managers to most effectively design and manage the planning and control
systems used by their companies.

2. The issues of organisational output (manufacturing vs. service), as well as the
amount of customer influence in the design of the product or service.

16

3. The categories of processing options, ranging from projects used for unique output
with very low volume to flow production used for very high outputs of standard
products.

4. The dimension of competition by which the customer makes their purchasing
decision (the order winner) from companies who have attained a basic level of
performance in order qualifying criteria.

5. The dynamic nature of customer behaviour and process change based on customer
and technological issues.

REFERENCES

1. Operation Management: Flexible Version. Seventh Edition. Heizer, J. & Render, B.
Pearson Prentice Hall 2005.

2. Restoring Our Competition Edge: Competing Through Manufacturing. Hayes, R. H.,
and Wheelwright, S. C., New York: John Wiley 1984.

3. Manufacturing Strategy. Hill, T. New York: Irwin McGraw-Hill 2000.
4. Manufacturing Planning And Control Systems. Vollmann, T. E., Berry, W. L. and

Whyback, D. C. New York: Irwin McGraw-Hill 1997.

17

FORECASTING

CHAPTER 2

FORECASTING

INTRODUCTION

This chapter explains the Forecasting methods and techniques. The techniques to be used
may vary according to type of data that have been acquired for forecasting. However, the
accuracy of each forecasting technique can be determined through forecasting errors
analsysis. Each section is provided with simple example with similar data, so that students
will be able to differentiate the forecasting techniques accordingly.

LEARNING OBJECTIVES

The learning objectives of this chapter are;
1. Students are able to use appropriate forecasting techniques according to acquired

data.
2. Student are able to inspect the accuracy of the forecasting made through the

forecasting errors analysis.

KKTM KUANTAN DMQ40403

18

FORECASTING

2.1 FORECAST

In general, Forecasting is the process of
estimation in unknown situations. Prediction is a
similar, but more general term, and usually refers
to estimation of time series, cross-sectional or
longitudinal data.

Forecasting in production is a process of judging how much production is
required to meet estimated sales in a particular forecasting period.
Considerations include previous sales, the general state of the economy, consumer
preferences, and competitive products. Production forecasting decisions affect
budgetary and scheduling decisions.

In more recent years, Forecasting has evolved into the practice of Demand Planning
in every day business forecasting for manufacturing companies. The discipline of
demand planning, also sometimes referred to as supply chain forecasting, embraces
both statistical forecasting and consensus process. Applications for forecasting
include;

 Inventory control/production planning
Forecasting the demand for a product enables us to control the stock of raw
materials and finished goods, plan the production schedule, etc.

 Investment policy
Forecasting financial information such as interest rates, exchange rates, share
prices, the price of gold, etc.

 Economic policy
Forecasting economic information such as the growth in the economy,
unemployment, the inflation rate, etc is vital both to government and business in
planning for the future.

KKTM KUANTAN DMQ40403

19

FORECASTING

2.2 TYPES OF FORECAST

One way of classifying forecasting problems is
to consider the timescale involved in the
forecast i.e. how far forward into the future we
are trying to forecast.

One way of classifying forecasting problems is to consider the timescale involved in
the forecast i.e. how far forward into the future we are trying to forecast. Short,
medium and long-term are the usual categories but the actual meaning of each
will vary according to the situation that is being studied, e.g. in forecasting energy
demand in order to construct power stations 5-10 years would be short-term and 50
years would be long-term, whilst in forecasting consumer demand in many
business situations up to 6 months would be short-term and over a couple of
years long-term. The table below shows the timescale associated with business
decisions;

Table 2.1 : Time scale Forecasing

Timescale Type of decision Examples
Operating
Short-term Tactical Inventory control
Up to 3-6 months Strategic Production planning, distribution

Medium-term Leasing of plant and equipment
3-6 months - 2 years Employment changes

Long-term Research and development
Above 2 years Acquisitions and mergers
Product changes

The basic reason for the above classification is that different forecasting
methods apply in each situation, e.g. a forecasting method that is appropriate for
forecasting sales next month (a short-term forecast) would probably be an
inappropriate method for forecasting sales in five years time (a long-term forecast). In
particular note here that the use of numbers (data) to which quantitative techniques
are applied typically varies from very high for short-term forecasting to very low for
long-term forecasting when we are dealing with business situations.

KKTM KUANTAN DMQ40403

20

FORECASTING

2.3 FORECASTING TECHNIQUES

There are two broad categories of forecasting techniques: quantitative methods
and qualitative methods. Quantitative methods are based on algorithms of varying
complexity, while qualitative methods are based on educated guessing. Forecasting
methods can be classified into several different categories:

i. Qualitative methods - where there is no formal mathematical model, often
because the data available is not thought to be representative of the future (long-
term forecasting).

ii. Quantitative methods - come in two main types: time-series methods and
explanatory methods.

iii. Regression methods - an extension of linear regression where a variable is
thought to be linearly related to a number of other independent variables.

iv. Multiple equation methods - where there are a number of dependent variables
that interact with each other through a series of equations (as in economic
models).

v. Time series methods - where we have a single variable that changes with time
and whose future values are related in some way to its past values. Time-series
methods make forecasts based purely on historical patterns in the data. Say you
want to forecast site visitors over the next few weeks. Time-series methods only
use historical site visit data to make that forecast.

2.3.1 Qualitative methods
Methods of this type are primarily used in situations where there is judged to be

no relevant past data (numbers) on which a forecast can be based and typically
concern long-term forecasting. One approach of this kind is the Delphi technique. It
involves asking a body of experts to arrive at a consensus opinion as to what the
future holds.

Underlying the idea of using experts is the belief that their view of the future will

be better than that of non-experts (such as people chosen at random in the street).

This brings us to our first key point, we are interested in the difference between the

original forecast and the final outcome, i.e. in forecast error. We need to use the most

appropriate (best) forecasting method, even if we know that (historically) it does not

give accurate forecasts. DMQ40403

KKTM KUANTAN

21

FORECASTING

2.3.2 Regression methods
The regression method is useful for time series that have a trend component.

You have probably already met linear regression where a straight line of the form
Y = a + bX is fitted to data. It is possible to extend the method to deal with more than
one independent variable X. Suppose we have k independent variables X1, X2,...., Xk
then we can fit the regression line;

Y = a + b1X1 + b2X2 + ... + bkXk

This extension to the basic linear regression technique is known as multiple
regression. Plainly knowing the regression line enables us to forecast Y given values
for the Xi i=1,2,...,k.

Example 2.1
Calculate the forecasting for week 11.

Week Number of Housing Number of Yards
Starts of Concrete Ordered

x y

1 11 225
2 15 250
3 22 336
4 19 310
5 17 325
6 26 463
7 18 249
8 18 267
9 29 379
10 16 300

Solution Number of Number Yards of
Housing starts Concrete Ordered
Week
xy xy x² y²
1 11 225 2475 121 50625
2 15 250 3750 225 62500
3 22 336 7392 484 112896
4
5 19 310 5890 361 96100
6 17 325 5525 289 105625
7 26 463 12038 676 214369
8 18 249 4482 324 62001
9
10 18 267 4806 324 71289
Total 29 379 10991 841 143641
16 300 4800 256 90000
191 3104 62149 3901 1009046

KKTM KUANTAN DMQ40403

22

FORECASTING

Solution

The average for;
X = 191/10 = 19.10
Y = 3104/10 = 310.40

Where,

b = ∑xy – nXY = (62149) – (10)(19.10)(310.40)

∑x² -nX² (3901) – (10)(19.10)²

b = 11.3191

And,
a = Y - bX = 310.40 – 11.3191(19.10)

a = 94.2052

Therefore, forecast for week 11,
Y = a + b1X1 + b2X2 + ... + bkXk

= 94.2052 + 11.3191(25)
= 337

2.3.3 Time series method
Methods of this type are concerned with a variable that changes with time and

which can be said to depend only upon the current time and the previous values that it
took (i.e. not dependent on any other variables or external factors). If Yt is the value of
the variable at time t then the equation for Yt is,

Yt = f(Yt-1, Yt-2, ..., Y0, t)

As an example, the value of the variable at time t is purely some function of its
previous values and time, no other variables/factors are of relevance. The purpose of
time series analysis is to discover the nature of the function f and hence allow us
to forecast values for Yt.

Time series methods are especially good for short-term forecasting where,
within reason, the past behaviour of a particular variable is a good indicator of its future
behaviour, at least in the short-term. The typical example here is short-term demand
forecasting. Note the difference between demand and sales - demand is what
customers want - sales is what we sell, and the two may be different. In graphical
terms the plot of Yt against t is as shown below.

KKTM KUANTAN DMQ40403

23

FORECASTING

The purpose of the analysis is to discern some relationship between the Yt
values observed so far in order to enable us to forecast future Yt values. We shall
deal with two techniques for time series analysis in detail and briefly mention a more
sophisticated method.

2.4 TECHNIQUE OF AVERAGING

2.4.1 Simple Moving Average (SMA)
One, very simple, method for time series forecasting is to take a moving

average (also known as weighted moving average). The moving average (mt) over
the last L periods ending in period t is calculated by taking the average of the values
for the periods t-L+1, t-L+2, t-L+3, ..., t-1, t so that;

mt = [Yt-L+1 + Yt-L+2 + Yt-L+3 + ... + Yt-1 + Yt]/L

To forecast using the moving average we say that the forecast for all periods
beyond t is just mt (although we usually only forecast for one period ahead, updating
the moving average as the actual observation for that period becomes available).

KKTM KUANTAN DMQ40403

24

FORECASTING

Example 2.2
The demand for a product for 6 months is shown below. Calculate the three month
moving average for each month and forecast the demand for month 7.

Month 123456

Demand (100's) 42 41 43 38 35 37

Solution:
Now we cannot calculate a three month moving average until we have at least 3
observations - i.e. it is only possible to calculate such an average from month 3
onward. The moving average for month 3 is given by;

m3 = (42 + 41 + 43)/3 = 42
and the moving average for the other months is given by;

m4 = (41 + 43 + 38)/3 = 40.7
m5 = (43 + 38 + 35)/3 = 38.7
m6 = (38 + 35 + 37)/3 = 36.7
We use m6 as the forecast for month 7.
Hence the demand forecast for month 7 is 3670 units.

2.4.2 Central Moving Average (CMA)
For a number of applications it is advantageous to avoid the shifting induced

by using only 'past' data. Hence a central moving average can be computed, using
both 'past' and 'future' data. The 'future' data in this case are not predictions, but
merely data obtained after the time at which the average is to be computed.
Weighted and exponential moving averages can also be computed centrally.

Example 2.3
This example analyzes annual sales data (in thousands of dollars) 1931 to 1960 shows
at below table. The data set is listed in Newbold. Calculate a centered 5-point moving
average. Given;

SAMPLE 1 30 years
READ SALES / BYVAR
1806 1644 1814 1770 1518 1103 1266 1473 1423 1767
2161 2336 2602 2518 2637 2177 1920 1910 1984 1787
1689 1866 1896 1684 1633 1657 1569 1390 1387 1289

KKTM KUANTAN DMQ40403

25

FORECASTING

Solution
The results for a centered 5-point moving average are listed in the Table in the column
MOVING-AVE.
Step 1
Consider a time series with observed values X1, X2, ..., XN. A centered 5-point moving
average is obtained as:

for t = 3, ..., N-2
Step 2
The number of periods used in calculating the moving average is specified with the NMA
= option on the SMOOTH command. The simple exponential smoothing method is based
on a weighted average of current and past observations, with most weight to the current
observation and declining weights to past observations. This gives the formula for the
smoothed series as;

where w is a smoothing constant with a value in the range [0,1]. The value for w is
specified with the WEIGHT = option on the SMOOTH command.

Step 3
By using excel,
i. Construct a Table as follow.
ii. Key-in the sales amount.
ii. Compute the following formula accordingly for calculating centered 5-point moving

average;
for t = 3, ..., N-2

KKTM KUANTAN DMQ40403

26

FORECASTING

Table 3.2 : Results of centered 5-point moving average

CENTRAL MOVING AVERAGES – PERIODS = 5 NSPAN = 1 WEIGHT = 0.600

OBSERVATION SALES MOVING-AVE SEAS&IRREG SA(SALES)

1 1806.0 ------- ------- 1806.0

2 1644.0 ------- ------- 1644.0

3 1814.0 1710.4 1.0606 1814.0

4 1770.0 1569.8 1.1275 1770.0

5 1518.0 1494.2 1.0159 1518.0

6 1103.0 1426.0 0.77349 1103.0

7 1266.0 1356.6 0.93322 1266.0

8 1473.0 1406.4 1.0474 1473.0

9 1423.0 1618.0 0.87948 1423.0

10 1767.0 1832.0 0.96452 1767.0

11 2161.0 2057.8 1.0502 2161.0

12 2336.0 2276.8 1.0260 2336.0

13 2602.0 2450.8 1.0617 2602.0

14 2518.0 2454.0 1.0261 2518.0

15 2637.0 2370.8 1.1123 2637.0

16 2177.0 2232.4 0.97518 2177.0

17 1920.0 2125.6 0.90327 1920.0

18 1910.0 1955.6 0.97668 1910.0

19 1984.0 1858.0 1.0678 1984.0

20 1787.0 1847.2 0.96741 1787.0

21 1689.0 1844.4 0.91574 1689.0

22 1866.0 1784.4 1.0457 1866.0

23 1896.0 1753.6 1.0812 1896.0

24 1684.0 1747.2 0.96383 1684.0

25 1633.0 1687.8 0.96753 1633.0

26 1657.0 1586.6 1.0444 1657.0

27 1569.0 1527.2 1.0274 1569.0

28 1390.0 1458.4 0.95310 1390.0

29 1387.0 ------- ------- 1387.0

30 1289.0 ------- ------- 1289.0

1 SEASONAL FACTORS

1 1.0000

A graph of the sales data is shown below;

KKTM KUANTAN DMQ40403

27

FORECASTING

The next graph shows the series smoothed by moving averages;

2.4.3 Weighted Moving Average (WMA)
A weighted average is any average that has multiplying factors to give

different weights to different data points. But in technical analysis a weighted
moving average (WMA) has the specific meaning of weights which decrease
arithmetically. In an n-day WMA the latest day has weight n, the second latest n-1,
etc, down to zero. In practice the weighting factors are often chosen to give more
weight to the most recent terms in the time series and less weight to older data.

When calculating the WMA across successive values, it can be noted that the
difference between the numerators of WMAM + 1 and WMAM is;

.

If we denote the sum by TotalM,

Then,

TotalM + 1 = TotalM + pM + 1 − pM − n + 1

NumeratorM + 1 = NumeratorM + npM + 1 − TotalM

KKTM KUANTAN DMQ40403

28

FORECASTING

WMA weights n=15

The denominator is a triangle number, and can be easily computed as
The graph at the right shows how the weights decrease, from highest weight for the
most recent days, down to zero. It can be compared to the weights in the exponential
moving average which follows.

2.4.4 Exponential Moving Average (EMA)
An Exponential Moving Average (EMA), sometimes also called an

Exponentially Weighted Moving Average (EWMA), applies weighting factors which
decrease exponentially. The weighting for each day decreases exponentially, giving
much more importance to recent observations while still not discarding older
observations entirely. The graph below shows an example of the weight decrease.

EMA weights N=15

The degree of weighing decrease is expressed as a constant smoothing
factor α, a number between 0 and 1. α may be expressed as a percentage, so a
smoothing factor of 10% is equivalent to α=0.1. Alternatively, α may be expressed in
terms of N time periods, where;

KKTM KUANTAN DMQ40403

29

FORECASTING

For example, N=19 is equivalent to α=0.1.
The simplest form of exponential smoothing is given by the formulas

Where α is the smoothing factor, and 0 < α < 1.

In other words, the smoothed statistic st is a simple weighted average of the
latest observation xt and the previous smoothed statistic st−1. Simple exponential
smoothing is easily applied, and it produces a smoothed statistic as soon as two
observations are available. Values of α close to unity have less of a smoothing effect
and give greater weight to recent changes in the data, while values of α closer to
zero have a greater smoothing effect and are less responsive to recent changes.
There is no formally correct procedure for choosing α.

2.4 CHOOSING A FORECASTING TECHNIQUE

One problem with forecasting is simple - how good is it?

Example 2.4
Use Simple Moving Average and Exponential Moving Average for below data. Which
type of forecasting would give less forecast error?

Month 123456

Demand (100's) 42 41 43 38 35 37

Solution
A) Simple Moving

 Produce a demand forecast for month 7 using a two month moving average.
This would give the following:
m2 = (42 + 41)/2 = 41.5
m3 = (41 + 43)/2 = 42
m4 = (43 + 38)/2 = 40.5
m5 = (38 + 35)/2 = 36.5
m6 = (35 + 37)/2 = 36

KKTM KUANTAN DMQ40403

30

FORECASTING

Would this forecast (m6 = 3600 units) be better than our current demand forecast
of 3670 units? Rather than attempt to guess which forecast is better we can
approach the problem logically.

 In an attempt to decide how good a forecast is we have the following logic.
Consider the three month moving average given above and pretend for a moment
that we had only demand data for the first three months, then we would calculate
the moving average for month 3 (m3) as 42 (see above). This would be our
forecast for month 4. But in month 4 the outcome is actually 38, so we have a
difference (error) defined by:
 error = forecast-outcome = 42-38 = 4

Note here that we could equally well define error as outcome-forecast. That would just
change the sign of the errors, not their absolute values. Indeed note here that if you inspect
the package output you will see that it does just that.

In month 4 we have a forecast for month 5 of m4 = 40.7 but an outcome for month 5
of 35 leading to an error of 40.7-35 = 5.7. In month 5 we have a forecast for month
6 of m5 = 38.7 but an outcome for month 6 of 37 leading to an error of
38.7-37 = 1.7.

 Hence we can construct the table below:

Month 1 234567

Demand (100's) 42 41 43 38 35 37 ?

Forecast - - - m3 m4 m5 m6

- - - 42 40.7 38.7 36.7

Error - - - 4 5.7 1.7 ?

 Constructing the same table for the two month moving average we have:

Month 1 234567

Demand (100's) 42 41 43 38 35 37 ?

Forecast - - m2 m3 m4 m5 m6

- - 41.5 42 40.5 36.5 36

Error - - -1.5 4 5.5 -0.5 ?

 Comparing these two tables we can see that the error terms give us a measure of
how good the forecasting methods (two or three month moving average) would
have been had we used them to forecast one period (month) ahead on the
historical data that we have. Plainly, in the real world, we are hardly likely to get a
situation where all the errors are zero. It is genuinely difficult to look at (as in this
case) two series of error terms and compare them.

KKTM KUANTAN DMQ40403

31

FORECASTING

 It is much easier if we take some function of the error terms, i.e. reduce each
series to a single (easily grasped) number. One suitable function for deciding
how accurate a forecasting method has been is;
 average squared error

The logic here is that by squaring errors we remove the sign (+ or -) and
discriminate against large errors (being resigned to small errors but being
adverse to large errors). Ideally average squared error should be zero
(i.e. a perfect forecast). In any event we prefer the forecasting method that
gives the lowest average squared error. We have that for the three month
moving average;

 average squared error = [4² + 5.7² + 1.7²]/3 = 17.13
and for the two month moving average;

 average squared error = [(-1.5)² + 4² + 5.5² + (-0.5)²]/4 = 12.19

 The lower of these two figures is associated with the two month moving average
and so we prefer that forecasting method (and hence prefer the forecast of
3600 for month 7 produced by the two month moving average). Average
squared error is known technically as the mean squared deviation (MSD)
or mean squared error (MSE).

Note here that we have actually done more than distinguish between two different forecasts
(i.e. between two month and three month moving average). We now have a criteria for
distinguishing between forecasts, however they are generated - namely we prefer the forecast
generated by the technique with the lowest MSD (historically the most accurate forecasting
technique on the data had we applied it consistently across time).

The disadvantages of Simple Moving Average
One disadvantage of using moving averages for forecasting is that in calculating the
average all the observations are given equal weight (namely 1/L), whereas we would
expect the more recent observations to be a better indicator of the future (and
accordingly ought to be given greater weight). Also in moving averages we only use
recent observations, perhaps we should take into account all previous observations.

KKTM KUANTAN DMQ40403

32

FORECASTING

Solution
B) Exponential smoothing

 Define a constant µ where 0 <= µ <= 1 then the (single) exponentially
smoothed moving average for period t (Mt say) is given by,
Mt = µYt + µ(1- µ)Yt-1 + µ(1- µ)²Yt-2 + µ(1- µ)³Yt-3 + ...
So you can see here that the exponentially smoothed moving average takes
into account all of the previous observations, compare the moving average
above where only a few of the previous observations were taken into account.
The above equation is difficult to use numerically but note that,
Mt = µYt + (1- µ)[µYt-1 + µ(1- µ)Yt-2 + µ(1- µ)²Yt-3 + ...]
i.e. Mt = µYt + (1- µ)Mt-1

 Hence the exponentially smoothed moving average for period t is a linear
combination of the current value (Yt) and the previous exponentially smoothed
moving average (Mt-1). The constant µ is called the smoothing constant and
the value of µ reflects the weight given to the current observation (Yt) in
calculating the exponentially smoothed moving average Mt for period t (which is
the forecast for period t+1).

For example, if µ = 0.2 then this indicates that 20% of the weight in
generating forecasts is assigned to the most recent observation and the
remaining 80% to previous observations.

Note here that Mt = µYt + (1- µ)Mt-1 can also be written Mt = Mt-1 - µ(Mt-1 - Yt) or current
forecast = previous forecast - µ(error in previous forecast) so exponential smoothing
can be viewed as a forecast continually updated by the forecast error just made.

 For the demand data given in the previous section calculate the exponentially
smoothed moving average for values of the smoothing constant µ = 0.2 and 0.9.
We have the following for µ = 0.2,

M1 = Y1 = 42 (we always start with M1 = Y1)
M2 = 0.2Y2 + 0.8M1 = 0.2(41) + 0.8(42) = 41.80
M3 = 0.2Y3 + 0.8M2 = 0.2(43) + 0.8(41.80) = 42.04
M4 = 0.2Y4 + 0.8M3 = 0.2(38) + 0.8(42.04) = 41.23
M5 = 0.2Y5 + 0.8M4 = 0.2(35) + 0.8(41.23) = 39.98
M6 = 0.2Y6 + 0.8M5 = 0.2(37) + 0.8(39.98) = 39.38

Note here that it is usually sufficient to just work to two or three decimal places when
doing exponential smoothing. We use M6 as the forecast for month 7, i.e. the forecast
for month 7 is 3938 units.

KKTM KUANTAN DMQ40403

33

FORECASTING

 We have the following for µ = 0.9,
M1 = Y1 = 42
M2 = 0.9Y2 + 0.1M1 = 0.9(41) + 0.1(42) = 41.10
M3 = 0.9Y3 + 0.1M2 = 0.9(43) + 0.1(41.10) = 42.81
M4 = 0.9Y4 + 0.1M3 = 0.9(38) + 0.1(42.81) = 38.48
M5 = 0.9Y5 + 0.1M4 = 0.9(35) + 0.1(38.48) = 35.35
M6 = 0.9Y6 + 0.1M5 = 0.9(37) + 0.1(35.35) = 36.84
As before M6 is the forecast for month 7, i.e. 3684 units.

 In order to decide the best value of µ (from the two values of 0.2 and 0.9
considered) we choose the value associated with the lowest MSD (as above for
moving averages).
For µ=0.2 we have that,
 MSD = [(42-41)²+(41.80-43)²+(42.04-38)²+(41.23-35)²+ (39.98- 37)²]/5
= 13.29

 For µ=0.9 we have that,
 MSD = [(42-41)²+(41.10-43)²+(42.81-38)²+(38.48-35)²+(35.35- 37)²]/5
= 8.52

Note here that these MSD values agree (to within rounding errors) with the MSD
values given in the package output above.

 Hence, in this case, µ=0.9 appears to give better forecasts than µ=0.2 as it
has a smaller value of MSD. Above we used MSD to reduce a series of error
terms to an easily grasped single number. In fact functions other than MSD
such as,
 MAD (mean absolute deviation) = average | error |

and
 bias (mean error) = average error

Also know as Cumulative Forecast Error exist which can also be used to
reduce a series of error terms to a single number so as to judge how good a
forecast is.

KKTM KUANTAN DMQ40403

34

FORECASTING

 In fact methods are available which enable the optimal value of the smoothing
constant (i.e. the value of µ which minimises the chosen criteria of forecast
accuracy, such as mean squared deviation (MSD) to be easily determined. This
can be seen below where the package has calculated that the value of µ which
minimises MSD is µ=0.86 (approximately).

Note here that the package can be used to plot both the data and the forecasts as
generated by the method chosen. Below we show this for the output above (associated
with the value of µ which minimises MSD of 0.86.

 To illustrate the change in MAD, bias and MSD as µ changes we graph below
MAD and bias against the smoothing constant µ,

 Below we graph the value of the forecast against µ. One particular point to
note is that, for this example, for a relatively wide range of values for µ the
forecast is stable (e.g. for 0.60 <= µ <= 1.00 the forecast lies between 36.75
and 37.00). This can be seen below - the curve is "flat" for high µ values.

KKTM KUANTAN DMQ40403

35

FORECASTING

Note here that the above graphs imply that in finding a good value for the smoothing
constant it is not usually necessary to calculate to a very high degree of accuracy (e.g. not
to within 0.001 for example).

Comments:
Exponential Moving smoothing (or, more accurately, single exponential
smoothing) gives greater weight to more recent observations and takes into
account all previous observations

2.6 TIME SERIES FORECASTING ERRORS

We cannot expect a time-series forecast to be perfect; it will surely and
always have prediction errors. Predict defines some statistics based on the error
terms;

et = Xt - X't

This is the difference between the actual time-series Xt and the forecast X't
and will be useful to analyze and summarize the accuracy of the forecasts. The
forecast errors are determined by following;

i. The Cumulative Forecast Error is the sum of all prediction errors;
CFE = ∑ et

ii. The Mean Error is the arithmetic average of all prediction errors;

ME = 1/n ∑ et

KKTM KUANTAN DMQ40403

36

FORECASTING

iii. The Mean Squared Error is the arithmetic mean of the sum of the squares of
the prediction errors. This error measure is popular and corrects the
'canceling out' effects of the previous two error measures;
MSE = 1/n ∑ e2t

iv. The Root Mean Squared Error is the square root of the MSE:
RMSE = √ MSE

v. The Standard Deviation is as the name implies the standard deviation of the
prediction errors.

vi. The Mean Absolute Deviation is another popular error measure that
corrects the 'canceling out' effects by averaging the absolute value of the
errors:
MAD = 1/n ∑ |et|

vii. The Mean Absolute Percent Error is a very popular measure that corrects
the 'canceling out' effects and also keeps into account the different scales at
which this measure can be computed and thus can be used to compare
different prediction;
MAPE = 100/n ∑ et / Xt

How much accuracy can we expect from a forecasting system? How much
does this accuracy (or inaccuracy) cost to you or your company? In general a MAPE
of 10% is considered very good, a MAPE in the range 20% - 30% or even higher is
quite common. How much will you or your company save if the MAPE reduces say
from 25% to 20%? Inaccurate forecasts will increase the need to keep stock in your
inventory system and will reduce the service level to the customer. Inaccurate
forecasts mean poor stock trading decisions and wrong timings. The cost can thus be
very high and it is worth the effort to insure that forecasts are as accurate as
possible.

KKTM KUANTAN DMQ40403

37

FORECASTING

SUMMARY

In this chapter we have studied that Forecasting could be calculated through two mehtods
which is Qualitative method or Quantitative method. In quantitative method, there are four
types of Forecasting technique;
1. Simple Moving Average
2. Centered Moving Average
3. Weight Moving Average
4. Exponential Moving Average.
The accuracy of Forcasting result from each technique can be determined through
Forecasting errors analysis.

REFERENCES

1. Nigel, S., Stuart, C., and Robert, J., 2004, Operations Management, 4th Edition,
Prentice Hall.

2. John, N., 2002, Introduction to Operations Management, 2nd Edition, Prentice Hall.
3. Hopp, W.J. and Spearman, M.L., 1995, Factory Physics, Foundation of Manufacturing

Management, Irwin Publisher.
4. www.forecastingprinciples.com

KKTM KUANTAN DMQ40403

38

CHAPTER 3 : SALES AND OPERATIONS PLANNING

INTRODUCTION

The strategic plans of the company and the more specific business plans derived from the
strategic plans will specify the product and service mix that the company will pursue, and will
also indicate planned changes in market penetration, market approaches, and other critical
aspects of the business. Strategic and business plans tend to be too general to specify
resource needs and the timing of those needs, however, and also tend to be too general to
adequately coordinate action plans and resource needs of several of the key functions of the
organisation, including Operations, Marketing / Sales, Finance, Information Technology, and
Human Resources.

Much of the most detail planning of resources, including the type of resources, the quantity
of resources, and the timing of those resources, is accomplished by Sales and Operations
Planning (S&OP). This planning activity tends to go by several names, depending on the
business and the type of production in which that business is involved. Other common
names that have been used in the past include aggregate planning, production planning, and
in the case of operations focused more directly in services, staffing planning. Sales and
operation planning (S&OP) is chosen as the preferred name in this subject primarily because
it more effectively indicates the trade-offs that are typically involved with those two major
functions in an organisation. In fact, production planning is not realy an adequate sysnonym
for S&OP, since it will be shown that developing a production plan is merely a portion of the
S&OP process in the organisation. Other functions (such as Human Resources, Information
Technology, and Finance) tend to be extremely important in the process, but primarily
because they represent opportunities and / or contratints in the ability of the organisation to
deliver startegic action plans.

39

LEARNING OBJECTIVES

The objectives of this unit are to :
1. explain the purpose of sales and operation planning which influence the future

production activity.
2. discuss the main idea in designing of sales and operation planning
3. discuss the several approaches can be used in sales and operation planning
4. discuss the balacing resources in sale and operations planning
5. discuss the qualitative issues that occured in sale and operation planning.

3.1 THE PURPOSE

The S&OP activity is seldom used for the actual scheduling of production activity.
Instead the primarily purpose is to plan for and coordinate resources, including type,
quantity and timing. The time horizon for the S&OP tends, then, to be dictated by
how long in the future the organisation needs to have an estimate of resource needs
in order to act appropriately to secure those resources. Manufacturing equipment
such as specialised machine tools often take well more than a year to design and
build, implying organisations using such equipment need to have plans that reach
that far into the future.

The same may be true with certain people with unique skilles, either because it may
take a long time to identify and recruit those people, or in some cases there may be
extensive training programs. Finance departments also need to know when certain
quantities of funds may be needed in order to plan financing and / or investment
activities. More specifically, the S&OP tends to be major source for the planning of:

a. Inventory levels
b. Cash flow
c. Human resource needs

i. Number of people
ii. Skill levels
iii. Timing of need
iv. Trianing programs
d. Capital needs

40

e. Production output
f. Facilities planning
g. Sale and marketing activities

i. Sales promotions
ii. Advertising
iii. Pricing
iv. New product introductions
v. Expansion of markets

In other cases, especially in pure service organisations requiring only people, the
time horizon in some cases may be much shorter, especially if the people are
relatively easy to obtain and / or train. Some of these organisations may have such
flexibility that the S&OP activity is not formalised at all, but is essentially driven on an
ad-hoc basis by general manager or operations manager.

The major objectives of the S&OP are:
a. Support and measure the business plan
b. Support the customer
c. Ensure that plans are realistic
d. Manage change effectively
e. Manage finished good inventory and / or backlog better to support customer
service
f. Control costs
g. Measure performance
h. Build teamwork

3.2 DESIGN OF SALES AND OPERATION PLANNING

Often products and / or services are aggregated along “family” lines in the S&OP
(thus the origin of the term “aggregate planning”). The key determinant is grouping
together products or services that the function of the activity is to plan resources. For
example, an organisation may make several different styles of tables, perhaps with
different wood and different finishes. From a sales and marketing perspective, they
may represent different products for different types of customers, but if they utilise
the same resources (e.g., people and equipment) they may be logically grouped
together into a product family from the perspective of the S&OP. While a common

41

method of aggregation is product families, some organisations use revenues or even
labor hours as the unit of analysis.

There is at least one other important reason for the aggregation. The primary source
of demand estimates that drive the development of the plan is forecasting. Forecast
tend to be more accurate when developed in aggregate as opposed to plans for
specific products or services. These forecasts need to be developed and then
coordinated with strategic plans that could provide significant influence on the actual
demand.

Examples of plans that could impact demand include:
a. Advertising campaigns
b. Promotions
c. Pricing changes
d. Strategic moves into new markets
e. Moves against competition
f. Development of new products
g. New usesfor existing products

Clearly these plans need to be carefully coordinated with whatever resources are
needed to accomplish the plans. This coordination is a major function of the S&OP.

There are other issues in the design that need to be addressed, including the
aggregation of time. For example, it is preferable to examine “buckets” of time that
represent a week, a month, or some other unit of time. Again, the answer lies in a
basic trade-off between the level of detail that is useful for planning and the amount
of effort necessary to obtain information. The general rule is to aggregate as much as
possible to the point where useful resource plans can be made. Aggregation of time
and production units allow for ease of plan development and tend to be more
accurate in the aggregate, but aggregation should not be done to tha point where
useful information is lost. The correct amount of aggregation is highly edpendent on
the type of product or service, the nature of the customers being served, and the
processes being used to deliver the product or service.

42

3.3 APPROACHES TO SALES AND OPERATIONS PLANNING

The primary focus of developing S&OP is to establish decisions about sales volumes,
customer service goals, rates of production,inventory levels, and order backlogs. In
order to accomplish this process, it is important for Sales, Marketing, Operations,
Finance and Product Development to all work together, guided by the strategic plan
and vision for the future of the organisation.

Once the strategic planning process is completed in an organisation, it is generally
used to make a business plan, which is ussually expressed in financial terms. Since
many of the decisions made in the S&OP will impact the financial plans, it is
important that these two planning processes are reconciled for agreement. Since in
most organisations the business plan is “owned” by top management, it is clear that
those top managers must also be involvement is important. Their involvement
provides a clear “message” to all in the organisation that the process and outcome of
the process are important activities and therefore the resulting plans should be
followed.

3.3.1 THE MAKE TO STOCK VIEW OF AN S&OP

The diagram in Figure 2.1 provides a simple example of what the output from an
S&OP process may look like.

There are several issues that should be noted on this sample. First, note that the
sales history for the last 3 months shows that overall there were 11,000 more units
sold than was called for in the plan and the production was 4,000 less than called for
in the plan. This meant that over the 3 months the inventory would have dropped by
15,000 units, since they would have used finished goods inventory to satisfy
customer requirements. You can easily see the month to month calculations as well.
For example, in the first month on the chart (august), sales were 314,000 while
production was only 303,000 units a difference of 11,000 units. That is what brought
the planned inventory down from 150,000 units to 139,000 units.

43

Family: Standard Polybobs Unit Of Measure: 1,000 units
Target Finished Inventory: 15 days on hand

SALES HISTORY ND J FMAMJ J
ASO 320 300 310 310 310 320 320 320 320
Forecast 300 310 300
314 302 305
Actual
sales 14 - 8 5

Difference: 6 11
Month

Cumulative

PRODUCTION A S O ND J FMAMJ J
325 310 310 310 310 320 320 320 320
Production 300 310 300

plan

Actual 303 305 298

Production

Difference: 3 -5 -2

Month

Cumulative 3 -2 -4

INVENTORY A S O ND J FMAMJ J
140 150 150 150 150 150 150 150 150
Plan 150 150 150

Actual 139 142 135

Days on 14 14 13.5

hand

Figure 3.1: Sample S&OP for Make To Stock

You can also see how they plan to make up for the shortfall in their target level of 15
days (15,000 units) in inventory. In November they plan to produce 5,000 more than
expected sales, and then produce 10,000 more than sales for December. By the end
of December they should then be back on target.

Of course, these plans need to be reviewed and revised at the end of each month,
since neither sales nor production are likely to exactly equal projections. In addition,
conditions, policies, and other business plans may change. In this sense, the S&OP
can be thought of as a dynamic plan, rolling through time to reflect conditions at the
time. As each month passes, in fact, most companies will add an additional month to
the end of the plan to maintain the same time horizon.

44

3.3.2 THE MAKE TO ORDER VIEW OF AN S&OP

When the product is a make to order product, there is generally no finished goods
inventory. The orders are taken and then production starts to that order. The orders
that exist awaiting production are typically called a backlog. That produces a slight
variation in the S&OP, illustrated in Figure 2.2.

Family: Deluxe Polybobs Unit Of Measure: Each
Target Finished Inventory: 3 weeks

SALES HISTORY ND J FMAM J J
ASO 80 80 80 80 80 80 80 80 80
Forecast 80 80 80
Actual 82 79 81
sales
Difference: 2 -1 1
Month
Cumulative 12

PRODUCTION A S O N DJ FMAMJ J
80 80 80 80 80 80 80 80 80
Production 80 80 80

plan

Actual 80 81 81

Production

Difference: 011

Month

Cumulative 12

ORDER AS O ND J F M AM J J
BACKLOG 60 60 60 60 60 60 60 60 60
60 60 60 33 3 3 3 3 3 3 3
Plan 64 62 62
33 3
Actual 62
Days on
hand

Figure 3.2: Sample S&OP for Make To Order

45

3.4 STRATEGIES SALES AND OPERATIONS PLANNING

While we now may have a “view” of the production requirements and how they
“match” with sales, we need to develop more specific plans as to how the plan will
most effectively and efficiently be accomplished. The rest of the chapter focuses on
some of the approaches and trade-offs that can be used to more specifically look at
production planning, keeping in mind that the major focus should be on planning
resources.

3.4.1 SOME TECHNIQUES

Mathematically, there are several approaches used to develop plans. In the past
some companies would often try to put the major information into mathematical
algorithms to search for an optimal combination of products to maximise an objective
function, often defined in terms of profitability. While that approach is still taken in
some environments where capacity and outputs are well defined and not too complex
(such as in some process industries such as chemicals), many companies opt for
different approaches for several reasons:

a. Environments are often too complex to capture all the major variables and
conditions adequately without making the model very difficult to set up, solve
and manage.

b. When simplifying assumptions are made to make the mathematical model
manageable, the simpler model often does not adequately reflect the
environment itself.

c. Many managers have not been adequately trained in modeling techniques to
the point where they can understand how to manage the process.

A second approach is to simulate the production environment in a computer
simulation, allowing rapid and effective solutions to scenarios that can be input into
the program. This approach is gaining in popularity as fast and efficient computers
become plentiful and inexpensive, and as the simulation programming packages
become more powerful and easier to use. While it is often difficult to initially build the
simulation model, once built the approach can be quite effective in developing “what
if” approaches to the planning process.

46

The third popular approach is really a subset of the second. It involves simulating the
demand on an production resource environment in the form of a computer
spreadsheet. As with a simulation of the entire production environment, once the
spreadsheet format has been established it becomes relatively easy to investigate
various approaches in a “what if” format. A major difference in both the computer
model simulation and the spreadsheet appraoch is that they usually do not give an
optimal solution merely a rapid and fairly simple approach to search for a satisfactory
solution for the various combinations of conditions being input. This third approach,
while not generally yielding an optimal solution, is heavily used because of the ease
of use and the widespread knowledge and acceptance of spraedsheet software.

3.4.2 TRADE-OFF APPROACHES

The general objective of developing a good S&OP is to find the “best” alternative to
align resources to meet the expected demand under certain operating conditions.
Often “best” means an attempt to maximize organisation profits, but other conditions
can also be established to define “best” in the context of the organisation’s strategic
plan. Examples of other conditions may be:

a. Attempting to meet all expected customer demand
b. Attempting to minimize inventory investment
c. Attempting to minimize the adverse impact on people, often experienced with

volatility in the workforce caused by frequent layoffs.

It is frequently impossible to establish perfect conditions, so these tradeoff criteria are
important to understand as the plan is being developed. For example, under certain
expected demand conditions very high levels of customer service may only be
possible by having a very large inventory investment, sometimes to the point of
having a negative impact on profitability for that period. In such a case the
organisation has to make a conscious decision as to whether it is better to allow
customer service to fall or to accept the negative financial implications of attempting
to meet customer demand. If the decision criteria used to make the final decision are
established before the development of the plan, it often leads to a much smoother
plan development with smaller probability for functional “battles” based on
functionally focused criteria. In this context it must be noted that the S&OP planning
process is for the entire business, not just for any one function.

47

There are three general categories of approaches used. They are:

a. Level. As the name implies, a level planning approach establishes a level set
of resources and implies the demand will fluctuate around those available
resources, or in some cases, attempts to alter the demand patterns
themselves to more effectively match the resource level established. This
approach tends to be more common and certainly more appealing in
environments where resources are difficult or expensive to alter. This also
tends to be the approach used in many “lean production” environments.

Examples include:
i. Professional services, such as doctors and dentists. Professional
services tend to use appoinments to alter and smooth demand
patterns around the availability of the relatively expensive and difficult-
to-alter resource represented by the doctors and dentists themselves.

ii. Hotel and airlines. In both cases the resources room and seats,
respectively are again expensive and difficult to alter in quantity.
Appointments under a different name (reservations) are again used to
alter demand patterns. In addition, pricing strategies (weekend rates
and super-saver tickets, for example) are used to once again alter
demand patterns to smooth out the demand closer to the resource
availability. Many restaurants and automobile repair facilities also fit
into this same category.

iii. Some manufacturing processes have similar characteristics. Some
chemical processes cannot be turned off without expensive and time-
consuming startup activities and they additionally cannot be sped up
or slowed down. An example is making certain glass products in large
volume. The glass furnace may need to be continually run, as shutting
it down implies cleaning out the entire furnace and starting it over. The
one "luxury" that manufacturing processes have over the two previous
examples is the ability to inventory the output as an alternative to
altering the demand.

48

Level strategies have appeal from the perspective that they tend to provide
highly stable production environments, but tend to put more pressure on the
sales and marketing activities to alter the demand patterns in appropriate
ways if the normal market demand does not tend to be level in a non
influenced environment. The only alternative is to build an inventory in low
demand times and use it when demand is high.

b. Chase. This approach represents the other extreme, in that demand is not
altered, but resources are. In fact in a "pure" chase environment the
resources are continually being raised or lowered to meet the demand as it
fluctuates under normal market conditions. As the approach is the opposite
of level, so too are the typical characteristics of the environments where
chase strategies may be appealing or, in some cases, the only alternative.
These tend to be environments where demand is difficult or impossible to
alter and where attractively simple and/or inexpensive methods to alter the
resource base are available. Examples of such environments include:

i. "Mid-tier" suppliers of manufactured products. The demand for such
products are often resulting from customers two or more levels down
the supply chain, making it difficult if not impossible to alter demand.
For example, a supplier of light bulbs for automobiles is reacting to
demand from the automobile manufacturers, which in turn are
reacting to the consumers of the automobiles. The light bulb
manufacturer has little ability to influence the demand for automobiles
themselves.

ii. Service industries where demand is difficult to predict and equally
difficult to alter. Some examples include:

 Grocery stores and banks, where demand is often not
recognized until the customer actually walks in and declares
what they want.

 Professional tax accounting services, which must attempt to
deliver much of their service during the tax "season" with little
opportunity to alter the demand pattern.

 Some "process" industries such as electric utilities.

49

c. Combination. This approach is by far the 'most common approach. As the
name implies, companies using this approach will "mix and match," altering
demand and resources in such a way to maximize performance to their
established criteria, including profit, inventory investment, and the impact on
people.
Graphically, the differences in the three approaches can be illustrated in
Figures 2.3, 2.4, 2.5, 2.6, and 2.7.

Figure 3.3: A Demand Pattern with Chase Production

Figure 3.4: A Demand Pattern with Level Strategy

50