A.

Suresh babu
College of Engineering, Guindy
 Unit I Manufacturing Planning and Control and
Forecasting 9

Introduction: Production Planning and Control-Limitations with

Traditional Production Planning and Control-Need and Evolution of
Manufacturing Planning and Control (MPC) System -Basic framework -
Demand Management in MPC System- Forecasting: Time Horizon,
Design of Forecasting Systems - Developing the Forecast Logic–
Qualitative methods: Delphi Technique, Market Research, Quantitative
methods -Time Series - Moving Averages, Exponential Smoothing
-Regression- Measure of Forecast Accuracy- Numerical Problems.

01/19/2021 Dr.ASB,CEG 2
 Challenges for a Manufacturing Engineer
Manufacturing engineers are required to achieve the following
objectives to be competitive in a global context.

Reduction in inventory
Reduce waste
Lower the cost of the product
Improve quality

01/19/2021 Dr.ASB,CEG 3
 Challenges for a Manufacturing Engineer

Increase flexibility in manufacturing to achieve immediate and rapid

response to:
Product changes

Production changes

Process change

Equipment change
Change of personnel
01/19/2021 Dr.ASB,CEG 4
 Evolution of CNC

• The manufacturing technology available during late 40's and early

50's could not meet the design and manufacturing challenges

arising out of the need to develop sophisticated aircraft and

satellite launch vehicles.

• During second world war this prompted the US Air Force to

approach MIT to develop suitable control systems, drives and

programming techniques for machine tools using electronic

control.

01/19/2021 Dr.ASB,CEG 5
01/19/2021 Dr.ASB,CEG 6
 Evolution of CNC
• The first major innovation in machine control is the Numerical
Control (NC), demonstrated at MIT in 1952.

• Early Numerical Control Systems were all basically hardwired

systems, since these were built with discrete systems or with later
first generation integrated chips.

• Mainframe computers were used to control a group of NC

machines by mid 60's.

• This arrangement was then called Direct Numerical Control (DNC)

as the computer bypassed the tape reader to transfer the
program data to the machine Dr.ASB,CEG
01/19/2021
controller. 7
 Evolution of CNC

• By late 60's mini computers were being commonly used to

control NC machines.

• At this stage NC became truly soft wired with the facilities of

mass program storage, offline editing and software logic control

and processing.

• This development is called Computer Numerical Control (CNC).

01/19/2021 Dr.ASB,CEG 8
 Evolution of CNC….
• Since 70's, numerical controllers are being designed around
microprocessors, resulting in compact CNC systems.

• A further development to this technology is the Distributed

numerical control (also called DNC) in which processing of NC
program is carried out in different computers operating at
different hierarchical levels - typically from mainframe host
computers to plant computers to the machine controller.

• Today the CNC systems are built around powerful 32 bit and 64
bit microprocessors.
01/19/2021 Dr.ASB,CEG 9
 Evolution of CNC….
• PC based systems are also becoming increasingly popular.

• CNC technology was adapted in the development of co-ordinate

measuring machine's (CMMs) which automated inspection.

01/19/2021 Dr.ASB,CEG 10
 Evolution of CNC….
• Robots were introduced to
automate several tasks like
machine loading, materials
handling, welding, painting and
assembly.

• All these developments led to

the evolution of flexible
manufacturing cells and flexible
manufacturing systems in late
70's.
01/19/2021 Dr.ASB,CEG 11
 Evolution of CAD
• Evolution of Computer Aided Design (CAD), on the other hand

was to cater to the geometric modeling needs of automobile and

aeronautical industries.

• The developments in computers, design workstations, graphic

cards, display devices and graphic input and output devices during

the last few decades have been phenomenal.

01/19/2021 Dr.ASB,CEG 12
 Evolution of CAD
• This coupled with the development of operating system with

graphic user interfaces and powerful interactive (user friendly)

software packages for modeling, drafting, analysis and

optimization provides the necessary tools to automate the design

process.

01/19/2021 Dr.ASB,CEG 13
 Evolution of CAD
• CAD in fact owes its development to the APT language project at

MIT in early 50's.

• Several clones of APT were introduced in 80's to automatically

develop NC codes from the geometric model of the component.

• Now, one can model, draft, analyze, simulate, modify, optimize

and create the NC code to manufacture a component and

simulate the machining operation sitting at a computer

workstation.

01/19/2021 Dr.ASB,CEG 14
 Merging of CAD and CNC resulted in CIM

• If we review the manufacturing scenario during 80's we will find

that the manufacturing is characterized by a few islands of
automation.

• In the case of design, the task is well automated. In the case of

manufacture, CNC machines, DNC systems, FMC, FMS etc
provide tightly controlled automation systems.

• Manufacturing engineers also started using computers for such

tasks like inventory control, demand forecasting, production
planning and control etc.
01/19/2021 Dr.ASB,CEG 15
 Merging of CAD and CNC resulted in CIM

• Unless all the segments of manufacturing are integrated,

permitting the transfer of data across various functional
modules.

• This realization led to the concept of Computer Integrated

Manufacturing.

• Thus the implementation of CIM required the development of

whole lot of computer technologies related to hardware and
software.

• Yet the full potential of computerization could not be obtained.

01/19/2021 Dr.ASB,CEG 16
 Production Planning and Control

A Manufacturing Planning and Control system is

a methodology designed to manage efficiently
the flow of material,
the utilization of people and equipment,
and to respond to customer
requirements
by utilizing the capacity of suppliers, that of
internal facilities, and in some cases that of
customers to meet customer demand.

01/19/2021 Dr.ASB,CEG 17
 Definition

Plan the use of

Mfg Planning and people
Control System Efficient use
Manage the of capacity
efficient flow of and that of
materials suppliers.

Driven by customer Possibly even

demand and other customers.
Customer requirements

01/19/2021 Dr.ASB,CEG 18
 Evolution of Production and Inventory Control

1770 Domestic for own use handicraft production- Local

region cottage industries - Middle man
1770-1800 Steam engine & textile machinery -Industrial
revolution
1800 Formal organization - Foreman and a gang of workers -
Line type organization
1878-1890 Scientific management movement F.W.Taylor- Midvale
company
1915 Ford W.Harris - Westing house corporation- Economic
order quantity

01/19/2021 Dr.ASB,CEG 19
 Evolution of Production and Inventory Control

1928 T.C.Fry,Bell Telephone company -Use of Statistics-

inventory level

1934 R.H.Wilson- Order points

1940 World War II post war backlog in industries- Use of
operation research in PIC

1950 New techniques in forecasting , Master schedule-

adjustment, EOQ Computers-Introduction
1954 Computers-Accounting

01/19/2021 Dr.ASB,CEG 20
 Evolution of Production and Inventory Control

1957 Managing projects-CPM - Dupond, American

Production & Inventory control society,-APICS
dictionary
1959-1971 MRP system, micro computer, scale down system of
MRP
1980 MRP-II (CIPIC),Production planning, capacity
planning, shop floor control

01/19/2021 Dr.ASB,CEG 21
 Objectives

• Maximum customer service

• Minimum inventory investment

• Maximum utilization

01/19/2021 Dr.ASB,CEG 22
 A Context for Thinking about
Manufacturing Planning and Control (MPC)

What are processes?

Processes define how we do things!
Step 1 Step 2 Step 3 ……

So, what is the role of MPC?

The MPC system tells the processes what to work on.
Plans the resources that processes use (i.e. people, equipment,
material).
Plans for what processes will be doing in the future.
01/19/2021 Dr.ASB,CEG 23
 Key Areas of Influence on MPC system design

Degree of Internationalization

01/19/2021 Dr.ASB,CEG 24
 Key Areas of Influence on MPC system design
Role of the customer in the system

• Increasing use of Information Technology.

01/19/2021 Dr.ASB,CEG 25
 Key Areas of Influence on MPC system design
Increasing use of Information Technology

01/19/2021 Dr.ASB,CEG 26
 Functions
• Planning
– Production
– Inventory
– Capacity
• Authorization
– Production
– Procurement
• Control
– Production
– Inventory
– Capacity
• Storage and movement of materials
01/19/2021 Dr.ASB,CEG 27
 Information Processing in Manufacturing
Business Functions
1. Order Entry
2. Marketing
3. Sales
4. Billing Customer
5. Sales Forecasting

Product Design
1. Product Development
2. Engineering Drawings
3. Bills of materials

Manufacturing Planning
1. Process planning
2. Master schedule
3. Requirement
planning
4. Capacity planning
Manufacturing control
1. Shop Floor Control
2. Quality Control
3. Inventory Control

Factory operations

01/19/2021 Dr.ASB,CEG 28
 Traditional PIC Organization

01/19/2021 Dr.ASB,CEG 29
 Material Management Organization

01/19/2021 Dr.ASB,CEG 30
 Typical Production planning system
Engineering Design
Production decisions

Master Production
Master Production Engineering
Schedule and
SMchedule
Manufacturing
Data Base

Materials requirement Capacity Planning

planning

Purchasing

Inventory Management Shop floor Control

ation
oper
Fact
ory

s
Raw materials
Finished
Products

01/19/2021 Dr.ASB,CEG 31
Cycle of activities in a traditional production planning and control systems

Sales and
Forecasting Markettin
g

Production
planning
Design
Engineeri
Master ng
scheduling Process
planning

Requirements
planning Estimating

Cust
om
ers
Purchasin
g Machine loading

and scheduling

Dispatching

Inventory control

Production and Assembly

Shipping and Ware
Out side housing
suppliers Expediting

Quality control

01/19/2021 Dr.ASB,CEG 32
Functions of Traditional Production Planning and Control
1 Forecasting
 The forecasting function is concerned with projecting or
predicting the future Demands .
 
Long range forecasts:
Five years or more
Plant construction and equipment acquisition.
 
Intermediate range forecasts
one or two years
plan for long lead-time material and components
 
Short-term forecasts
Three to six month futures.
Decisions on personnel, purchasing, and production
01/19/2021
schedulingDr.ASB,CEG 33
Functions of Traditional Production Planning and Control

 2 Production Planning
 It's objective is to establish general production levels
for products group over the next year or so. It is based
on the sales forecast and is used to raise or lower the
inventories, stabilize production over the planning
horizon, and allow the launching of new products into
the company's product line.
 Aggregate production planning is a function that
precedes the detailed Master Production Schedule
(MPS).
01/19/2021 Dr.ASB,CEG 34
Functions of Traditional Production Planning and Control
3 Process Planning
Process planning involves determining the
sequence of manufacturing operations
required producing a certain product and/or
its components. Process Planning has been
traditionally been carried out manually,
prepared be hand and called `route sheet'.

 Route sheet is a listing of the operations and

machine tools through which the part or

product must be routed.

01/19/2021 Dr.ASB,CEG 35
Functions of Traditional Production Planning and Control
 
 4 Estimating
 
For purpose to determine prices, predicting costs and
preparing schedules, the firm will determine estimates
of the manufacturing lead time and production costs
for its products.
 
The manufacturing lead-time is the total time required
to process a work part throughout the factory.
 
The production costs are the sum of the material costs,
labor costs and applicable overhead costs needed to
produce the part. These estimates of lead-time and
costs are based on data contained in the route sheets,
purchasing files, and accounting record.
01/19/2021 Dr.ASB,CEG 36
Functions of Traditional Production Planning and Control
5 Master Scheduling
 
The aggregate production plan must be translated into a master schedule
which specifies how many units of each product are to be delivered and when.
 
In turn, this master schedule must be converted into purchase orders for raw
materials, orders for components from outside vendors, and production
schedules for parts made in the shop.
 
These events must be timed and coordinated to allow delivery of the final
product according to the master schedule.
 
Specifically, the master schedule or master production schedule is listing of the
products to be produced, when they are to be delivered, and in what
quantities.
 
The scheduling period in the MPS are typically months, weeks, or dates.
 
The MPS must be consistent with the plant's production capacity. It should not
list more quantities of product than the plant is capable of producing with its
given resources of machine and labor. 

01/19/2021 Dr.ASB,CEG 37
Functions of Traditional Production Planning and Control
6 Requirement Planning
Based on MPS, the individual component and
subassemblies that make up each product
must be planned. Raw material must be
ordered to make the various components.
Purchased parts must be ordered. And all
these items must be planned so that the
components and assemblies are available
when needed. This whole task is called
requirement planning (RP). 

01/19/2021 Dr.ASB,CEG 38
Functions of Traditional Production Planning and Control
 
 7 Purchasing
 The firm will elect to manufacture some components
for its products in its own plants. Other components
will be purchased. Deciding between these
alternatives is the familiar "make or buy" decision.

 For the components made in-house, raw material

have to be acquired. Ordering the raw material is the

function of the purchasing department.

 Materials will be ordered and the receipt of these

items will be scheduled according to the timetable

defined during the requirement planning procedure.

01/19/2021 Dr.ASB,CEG 39
Functions of Traditional Production Planning and Control
8 Scheduling (Machine loading, Dispatching,
Expediting)
 Also based on the requirement planning activity is production scheduling. This
involves the assignment of start dates and due dates for the component to be
processed throughout the factory. Several factors make scheduling complex.
 
First, the number of individual parts and orders to be scheduled may run into the
thousands.
 
Second, each part has its own individual process routing to be followed. Some part
may have to be routed through dozens of separate machines.
 
Third, the number of machines in the shop is limited, and the machines are
different. They perform different operations and have different features and
capacities.
 
The total number of jobs to be processed through the factory will typically exceed
the number of machines by a substantial margin. Accordingly, each machine, or
work center, will have a queue of jobs waiting to be processed.
Allocating the jobs to work center is referred to as machine loading. Allocating the
01/19/2021 Dr.ASB,CEG
jobs to the entire shop is called shop loading. 40
Functions of Traditional Production Planning and Control
 9 Quality Control
 
The quality control department is responsible for
assuring that assuring that the quality of the product
and its component meets the standard specified by
the designer.
 
This function must be accomplished at the various
points throughout the manufacturing cycle.
Materials and parts purchased from outside
suppliers must be inspected when they are received.
 
Parts fabricated inside the company must be
inspected, usually several times during data
processing. Final inspection of the finished product
is performed to test its overall functional and
appearance quality.

01/19/2021 Dr.ASB,CEG 41
Functions of Traditional Production Planning and Control

 10. Inventory Control

 
The purpose of the inventory control is to ensure that enough products
of each type are available to satisfy customer demand. Competing with
this objective is the desire that the company's financial investment in
inventory be kept at a minimum.
 
Inventory control interfaces with production control since there must
be coordination between the various products sales production and
inventory level. Inventory control is often included within the
production control department.
 
The inventory control function applies not only to the company’s final
product. It Laos applies to raw material, purchased components and
WIP (work in progress) within the factory.
 
In each case, planning and control are required to achieve a balance
between the danger of too little inventory) with possible stock outs)
and the expense of too much inventory.

01/19/2021 Dr.ASB,CEG 42
 Problems with Traditional Production Planning and
Control

Plant capacity problems. Production falls behind schedule due to a lack of labor
and equipment. This results in excessive overtime, delays in meeting delivery
schedules, customer complaints, backordering, and other similar problems.
Suboptimal production scheduling. The wrong jobs are scheduled because of a lack
of clear order priorities, inefficient scheduling rules, and the ever-changing status of
jobs in the shop. As a consequence, production runs are interrupted by jobs whose
priorities have suddenly increased, machine setups are increased, and jobs that are
on schedule fall behind.
Long manufacturing lead times. In an attempt to compensate for problems 1 and 2,
production planners allow extra time to produce an order. The shop becomes
overloaded, order priorities become confused, and the result is excessively long
manufacturing lead times.
Inefficient inventory control. At the same time that total inventories are too high
for raw materials, work-in-progress, and finished products, there are stockouts that
occur on individual items needed for production. High total inventories mean high
carrying costs, while raw material stockouts mean delays in meeting production
01/19/2021 Dr.ASB,CEG 43
schedules.
Problems with Traditional Production Planning and Control
Low work center utilization. This problem results in part from poor
scheduling (excessive product changeovers and job interruptions), and
from other factors over which plant management has limited control
(e.g., equipment breakdowns, strikes, reduced demand for products).
Process planning not followed. This is the situation in which the reqular
planned routing is superseded by an ad hoc process sequence. It occurs,
for instance, because of bottlenecks at work centers in the planned
sequence. The consequences are longer setups, improper tooling, and
less efficient processes.
Errors in engineering and manufacturing records. Bills of materials are
not current, route sheets are not up to date with respect to the latest
engineering changes, inventory records are inaccurate, and production
piece counts are incorrect.
Quality problems. Quality defects are encountered in manufactured
components and assembled products, resulting in rework or scrapped
parts, thus causing delays in the shipping
01/19/2021 Dr.ASB,CEG schedule. 44
 Computer integrated production management system
• Various factors working over the last several decades to cause
the evolution of a more modern and effective approach to the
problem of production planning and control towards
computerization were:
• Increase in the level of production professionalism
• Development of better production planning
• Increase in the complexity of both the products manufactured
and the markets to buy these.
• Development of the techniques of the operations research

01/19/2021 Dr.ASB,CEG 45
Computer integrated production management system
In an ideal system the computerised system would overcome these problems
and the systems will also be integrated through the principle functions of the
computerised production management system such as:

Master Production Scheduling – This is a list of what needs to be

manufactured, when they are to be delivered, and in what quantities,
developed from customer orders and demand forecasting. It is basically a
production plan that inputs to the material requirements planning.

Material Requirements Planning (MRP) – a computerised procedure to help

determine when to order parts and raw materials. It can also be used to
reschedule orders due to changes in demand and priority.

Capacity Planning. This element differs from MRP in that it concentrates with
planning for production resources such as labour and equipment rather than
materials and components. It could be defined as the amount of product that
the production facility can produce in a specified time frame.
01/19/2021 Dr.ASB,CEG 46
 Computer integrated production management system

However, it would be a mistake to think that this is the end of the computers impact on
production planning, as it actually has a role to play in many of the manufacturing
functions which interface with production planning such as:
  
Purchasing – this department places the orders that are required by the MRP system
and the capacity system. The function also has the responsibility for qualifying vendors
and maintaining sources of supply.
 
Shop floor control – this area monitors the progress of the orders through the factory,
reporting to management to ensure effective control.
 
Engineering / manufacturing database – this consists of all the data, such as product
designs, bills of materials, process plans and so on. Each are needed to make the
components and assemble the products. They also help with the planning calculations
for the MRP and capacity planning systems.
 
Inventory / warehouse management – this is the responsibility for keeping investment
in stocks to a minimum without disrupting operations.
01/19/2021 Dr.ASB,CEG 47
Computer Integrated Production Planning and Control
System

01/19/2021 Dr.ASB,CEG 48
 Typical Support Activities

• Long Term

• Intermediate Term and

• Short Term.

01/19/2021 Dr.ASB,CEG 49
 Long Term

The system is responsible for providing information to make

decisions on

– The appropriate amount of capacity - including supplier

capacity - to meet the market demands of the future.

– Provide the appropriate mix of human resource

capabilities, technology, and geographic locations.

01/19/2021 Dr.ASB,CEG 50
 Intermediate Term

The fundamental issue is matching supply and demand in

terms of both volume and product mix.
– Planning for the right logistics.
– Providing customers with information on correct quantities
and location to meet market needs.
– Planning of capacity to determine employment levels,
budgets, overtime and subcontracting needs, etc.

01/19/2021 Dr.ASB,CEG 51
 Short Term
There is a need for detailed scheduling of resources to meet
production requirements.

– Involves time, people, material, equipment and facilities.

– Involves people working on the right things.

– Involves tracking the use of resources and execution

results.

– Provide problem-solving support.

01/19/2021 Dr.ASB,CEG 52
 Performance Indicators

• Output results
• Equipment utilization
• Cost associated with different departments, products, labor
utilization, and project conditions.
• Measures of customer satisfaction such as late deliveries,
product returns, quantity and quality errors.

01/19/2021 Dr.ASB,CEG 53
 Costs and Benefits of MPC systems

• The costs of an MPC are substantial but so can be the Rewards .

• Common to see ERP system costs measured at 5-10% of cost of
goods sold.

01/19/2021 Dr.ASB,CEG 54
 An MPC System Framework

• Front End
– Set of activities and systems for overall direction setting
(Demand Management, Sales & Operations Planning,
Resource Planning and MPS)
• Engine

– Material Requirements Plans (MRP): Systems for detailed

material and capacity planning.
• Back End
– Depicts the MPC execution systems
01/19/2021 Dr.ASB,CEG 55
 Resource Sales and operations Demand
Planning planning management

Enterprise Resource Planning (ERP) System

Master production
scheduling Front End

Detailed capacity Detailed material

planning planning
Engine
Material
Requirements
Plans (MRP)

Shop-floor Supplier Back end

systems systems

Manufacturing Planning and Control System

01/19/2021 Dr.ASB,CEG 56
 Front End

• Demand Management

• Sales & Operations Planning

• Resource Planning

• Production Schedule (MPS)

01/19/2021 Dr.ASB,CEG 57
 Resource Sales and operations Demand
Planning planning management

Enterprise Resource Planning (ERP) System

Master production
Scheduling (MPS) Front End

Front End
SetDetailed capacity and Detailed
of activities systemsmaterial
for overall direction setting
planning planning
(Demand Management, Sales & Operations Engine Planning, Resource
Planning and MPS)
Material and
capacity plans

Shop-floor Supplier Back end

systems systems

Manufacturing Planning and Control System

01/19/2021 Dr.ASB,CEG 58
 Engine

• Detailed Capacity Planning (DCP)

• Detailed Material Planning (DMP)

• Material Requirements Plans (MRP):

01/19/2021 Dr.ASB,CEG 59
 Resource Sales and operations Demand
Planning planning management

Enterprise Resource Planning (ERP) System

Engine Master production
Systems for detail material and capacity planning.
scheduling Front End

Detailed capacity Detailed material

planning planning
Engine

Material and
capacity plans

Shop-floor Supplier Back end

systems systems

Manufacturing Planning and Control System

01/19/2021 Dr.ASB,CEG 60
 Back End (execution systems)

• Supplier systems

• Shop-floor systems

01/19/2021 Dr.ASB,CEG 61
 Resource Sales and operations Demand
Planning planning management

Enterprise Resource Planning (ERP) System

Master production
scheduling Front End

Detailed capacity Detailed material

planning planning
Engine

Back End Material and

capacity plans
MPC execution systems – Track progress and collect data.

Shop-floor Supplier Back end

systems systems

Manufacturing Planning and Control System

01/19/2021 Dr.ASB,CEG 62
 Matching MPC System with the Firm needs
(A DYNAMIC PROCESS)
• As competitive conditions, customer expectations, supplier capabilities, and
internal needs change the MPC should change.
• The trend is towards on-line data access and systems .
• Physical changes influences MPC:
– Outsourcing mfg (manufacturing).
– Hollowing out of corporation ( etc.
• Providing information at the level where decision are made, in appropriate time
and frames, directly from the shop floor
• Firms competing in product variety strength MPS and DMP
• Firms competing on delivery speed need to improve execution.
01/19/2021 Dr.ASB,CEG 63
 MPC Classification Schema
Number of Subparts

Project
MRP Managment

Just-in-time
Repetitive

Flow

Seconds Minutes Days Weeks Months

Time between successive units

01/19/2021 Dr.ASB,CEG 64
 MPC Classification Schema

Flow (chemical plants, food, etc.)

Number of Subparts

Project

MRP

Just-in-time

Repetitive

Flow

Seconds Minutes Days Weeks Months

Time between successive units

01/19/2021 Dr.ASB,CEG 65
 MPC Classification Schema

Repetitive Mfg. (automobiles, high

volume consumer products (i.e. TVs)
Number of Subparts

Project

MRP

Just-in-time

Repetitive

Flow

Seconds Minutes Days Weeks Months

Time between successive units

01/19/2021 Dr.ASB,CEG 66
 MPC Classification Schema
Just-in-Time systems (feedback management for
shorter life cycle products – e.g. fashion items). Focus
on minimizing inventory in the supply chain.
Number of Subparts

Project

MRP

Just-in-time

Repetitive

Flow

Seconds Minutes Days Weeks Months

Time between successive units

01/19/2021 Dr.ASB,CEG 67
 MPC Classification Schema

MRP (feedforward management for management

of complicated longer life products, i.e.
machines, electronic equipment).
Number of Subparts

Project

MRP

Just-in-time

Repetitive

Flow

Seconds Minutes Days Weeks Months

Time between successive units

01/19/2021 Dr.ASB,CEG 68
 MPC Classification Schema

Project (unique; long term), buildings, one of a kind

(satellites).
Number of Subparts

Project

MRP

Just-in-time

Repetitive

Flow

Seconds Minutes Days Weeks Months

Time between successive units

01/19/2021 Dr.ASB,CEG 69
 Principles

• The framework for the MPC is general and

all 3 phases must be performed but specific
applications must reflect particular firm
conditions and objectives.

• The MPC must coordinate the planning and

control efforts.

• MPC systems should support the strategy

and tactics pursued by the company.

01/19/2021 Dr.ASB,CEG 70
 Principles

• Different Manufacturing processes and

Supply Chains need different designs of the
MPC. … one size does not fit all.

• The MPC must evolve to meet changing

requirements in the market, technology,
products, and processes.

Major problem – the conflict between the

flexible and changing needs of companies
and the inflexibility of ERP systems.

01/19/2021 Dr.ASB,CEG 71
 Principles
• An effective MPC can contribute to competitive performance
by lowering costs and providing greater responsiveness to the
market.

• In firms that have an integrated ERP system and database, the

MPC system should integrate with and support cross-functional
planning through the ERP system.

01/19/2021 Dr.ASB,CEG 72
01/19/2021 Dr.ASB,CEG 73
01/19/2021 Dr.ASB,CEG 74
UNIT I PRODUCTION PLANNING AND CONTROL AND
FORECASTING: 9

• Introduction :Production Planning and Control-Traditional

Production Planning and Control - Problems with Traditional
Production Planning and Control-Computer-Integrated
Production Management System-Engineering and
manufacturing data base –Forecasting - Qualitative methods:
Delphi technique, Market research, Intrinsic methods-Time
series-moving averages-exponential smoothing- Extrinsic
methods-regression-forecast errors-numerical problems

01/19/2021 Dr.ASB,CEG 75
 Outline

• What is forecasting?
• Types of forecasts
• Time-Series forecasting
– Naïve
– Moving Average
– Exponential Smoothing
– Regression
• Good forecasts
“There are two kind of forecasters: those who don´t know and those who don´t know they don
´t know”
-John Kenneth Galbraith (1993)
 What is Forecasting?

 Process of predicting a future

event based on historical data
 Educated Guessing
 Underlying basis of
all business decisions
 Production
 Inventory
 Personnel
 Facilities

01/19/2021 Dr.ASB,CEG 77
 Why do we need to forecast?

In general, forecasts are almost always wrong. So,

Throughout the day we forecast very different things
such as weather, traffic, stock market, state of our
company from different perspectives.

Virtually every business attempt is based on

forecasting. Not all of them are derived from
sophisticated methods. However, “Best" educated
guesses about future are more valuable for purpose of
Planning than no forecasts and hence no planning.
01/19/2021 Dr.ASB,CEG 78
 Importance of Forecasting in OM

Departments throughout the organization depend on

forecasts to formulate and execute their plans.

Finance needs forecasts to project cash flows and capital

requirements.

Human resources need forecasts to anticipate hiring

needs.

Production needs forecasts to plan production levels,

workforce, material requirements, inventories, etc.
01/19/2021 Dr.ASB,CEG 79
 Importance of Forecasting in OM

Demand is not the only variable of interest to

forecasters.

Manufacturers also forecast worker absenteeism,

machine availability, material costs, transportation and
production lead times, etc.

Besides demand, service providers are also interested in

forecasts of population, of other demographic variables,
of weather, etc.

01/19/2021 Dr.ASB,CEG 80
 Famous long-term technological forecasts

•This 'telephone' has too many shortcomings to be seriously considered as

a means of communication. The device is inherently of no value to us."
Western Union internal memo, 1876

•"The wireless music box has no imaginable commercial value. Who would
pay for a message sent to nobody in particular?"
David Sarnoff's associates in response to his urgings for
investment in the radio in the 1920s

•"I think there is a world market for maybe five computers."

Thomas Watson, chairman of IBM, 1943

•"There is no reason anyone would want a computer in their home."

Ken Olson, president, chairman and founder of DEC, 1977
Characteristic requirements of good Forecasts

1. Accuracy

2. Cost

3. Response

4. Simplicity
 Constraints for Forecasts

1. Data

2. Time

3. Expertise

4. Funds

“ A good forecaster is not smarter than everyone else, he merely has his
ignorance better organized ”
-Anonymous
 Principles of Forecasting

They are always involves some error (actual demand –

forecast demand)
The longer the forecast horizon the less accurate the
forecast will be
Group forecasts are more accurate than individual
forecasts.
A good forecast says something about its likely error size
Wherever possible calculate demand rather than forecast
 Types of Forecasts by Time Horizon

Quantitative
• Short-range forecast
methods
– Usually < 3 months
• Job scheduling, worker assignments
Detailed
• Medium-range forecast use of
– 3 months to 2 years system

• Sales/production planning
• Long-range forecast
– > 2 years
• New product planning Design
of system
Qualitative
01/19/2021 Dr.ASB,CEG 85
Methods
 Designing a forecasting system
The following questions should be answered before designing a
forecasting system
1.What aggregate product, options items or stock keeping units
are to be forecast?
2.What geographic area should the forecast include?
3.What is the forecast horizon?
4.How should the forecast horizon be divided into periods?
5.How frequently should the forecast be updated?
6.How much accuracy is required?
 Forecasting Process
1. Identify the 2. Collect 3. Plot data
purpose of historical data and identify
patterns
forecast

4. Select a forecast
6. Check forecast 5. Develop/compute model that seems
accuracy with one or forecast for period of appropriate for data
more measures historical data

7.
8b. Select new forecast
Is accuracy model or adjust
of forecast parameters of existing
acceptable model
?

9. Adjust forecast
8a. Forecast based on additional 10. Monitor results
over planning qualitative and measure forecast
horizon information and accuracy
insight
 Forecasting During the Life Cycle

Introduction Growth Maturity Decline

Qualitative models Quantitative models

- Executive judgment
- Time series analysis
- Market research
-Survey of sales force - Regression analysis
-Delphi method
Sales

Time
01/19/2021 Dr.ASB,CEG 88
 Qualitative Forecasting Methods

Qualitative
Forecasting

Executive Models
Sales Force Market Research Delphi Method
Judgement Composite /Survey

Smoothing

01/19/2021 Dr.ASB,CEG 89
 Qualitative Methods

Briefly, the qualitative methods are:

Executive Judgment: Opinion of a group of high level experts or

managers is pooled

Sales Force Composite: Each regional salesperson provides his/her

sales estimates. Those forecasts are then reviewed to make sure
they are realistic. All regional forecasts are then pooled at the
district and national levels to obtain an overall forecast.

Market Research/Survey: Solicits input from customers pertaining to

their future purchasing plans. It involves the use of questionnaires,
consumer panels and tests of new products and services..

01/19/2021 Dr.ASB,CEG 90
 Qualitative Methods

Delphi Method: As opposed to regular panels where the individuals involved are
in direct communication, this method eliminates the effects of group potential
dominance of the most vocal members. The group involves individuals from
inside as well as outside the organization.
Typically, the procedure consists of the following steps:
Each expert in the group makes his/her own forecasts in form of statements
 The coordinator collects all group statements and summarizes them
 The coordinator provides this summary and gives another set of questions
to each
group member including feedback as to the input of other experts.
 The above steps are repeated until a consensus is reached.

01/19/2021 Dr.ASB,CEG 91
 Quantitative Forecasting Methods

Quantitative
Forecasting

Time Series Regression

Models Models

2. Moving 3. Exponential
1. Naive
Average Smoothing
a) simple a) level
b) weighted b) trend
c) seasonality
01/19/2021 Dr.ASB,CEG 92
 Quantitative Forecasting Methods

Quantitative
Forecasting

Time Series Regression

Models Models

2. Moving 3. Exponential
1. Naive
Average Smoothing
a) simple a) level
b) weighted b) trend
c) seasonality
01/19/2021 Dr.ASB,CEG 93
 Time Series Models
• Try to predict the future based on past data

– Assume that factors influencing the past will continue to

influence the future

01/19/2021 Dr.ASB,CEG 94
 Time Series Models: Components

Random Trend

Composite
Seasonal

01/19/2021 Dr.ASB,CEG 95
 Product Demand over Time
Demand for product or service

Year Year Year Year

1 2 3 4

01/19/2021 Dr.ASB,CEG 96
 Product Demand over Time
Trend component
Seasonal peaks
Demand for product or service

Actual demand
Random line
variation
Year Year Year Year
1 2 3 4
Now let’s look at some time series approaches to forecasting…
Borrowed from Heizer/Render - Principles of Operations Management, 5e, and Operations Management, 7e

01/19/2021 Dr.ASB,CEG 97
 Quantitative Forecasting Methods
Quantitative
Time Series
Models

Models

2. Moving 3. Exponential
1. Naive
Average Smoothing
a) simple a) level
b) weighted b) trend
c) seasonality

01/19/2021 Dr.ASB,CEG 98
 Naive Approach

 Demand in next period is the same as demand

in most recent period
 May sales = 48 → June forecast = 48

 Usually not good

01/19/2021 Dr.ASB,CEG 99
 Moving Average
•  Average several periods of data
• Dampen, smooth out changes
• Use when demand is stable with no trend or seasonal
pattern

D t i
M t 1  Ft  i 1
n

• Where
n =number of periods in the moving average
Di =demand in period i
t =Forecast period
01/19/2021 Dr.ASB,CEG 100
 Simple Moving Average
• Only a few periods of demand is considered for
evaluating the average
• Use when demand is neither growing nor declining
rapidly
• Use when demand is stable with no trend or
seasonal pattern
• Dampen, smooth out random fluctuation for
forecasting

01/19/2021 Dr.ASB,CEG 101

 Simple Moving Average-Example
3-Week Moving Average
n
MONTH DEMAND
D t i
Jan 120 M t 1  i 1
n
Feb 90
90  110  130
Mar 100 
3
Apr 75
Orders for Nov  110
May 110
June 50 5-Week Moving Average
July 75 n

Aug 130 D t i

Sept 110
M t 1  i 1
n
Oct 90
90  110  130  75  50

5
Orders for Nov  91
01/19/2021 Dr.ASB,CEG 102
 Simple Moving Average-Example
MA
MONTH ORDERS F3 F5
3 Week
Jan 120 – –
Feb 90 – –
Mar 100 103.3 – –
Apr 75 88.3 103.3 –
May 110 95.0 88.3 –
June 50 78.3 95.0 99.0
July 75 78.3 78.3 85.0
Aug 130 85.0 78.3 82.0
Sept 110 105.0 85.0 88.0
Oct 90 110.0 105.0 95.0
Nov – 110.0 91.0

01/19/2021 Dr.ASB,CEG 103

 Simple Moving Average-Example

3 Week Double Moving Average

D i
90  110  130
M oct  i 1
  110
3 3

M i
85  105  110
DM oct  i 3
  100
3 3

Orders for Nov  (( 2  110  100 )  (110  100 ))  130

01/19/2021 Dr.ASB,CEG 104

 Double Moving Average-Example

MA DM DM
MONTH ORDERS F3 F5
3 Week 3 Week F3
Jan 120 – –
Feb 90 – –
Mar 100 103.3 – –
Apr 75 88.3 103.3 –
May 110 95.0 88.3 – 95.5
June 50 78.3 95.0 99.0 87.2 94
July 75 78.3 78.3 85.0 83.9 60.5
Aug 130 85.0 78.3 82.0 80.5 67.1
Sept 110 105.0 85.0 88.0 89.4 94
Oct 90 110.0 105.0 95.0 100 136.2
Nov – 110.0 91.0 93.9 130

01/19/2021 Dr.ASB,CEG 105

 Smoothing Effects for given periods
Actual
140
120
100
80
Orders

60
40
20
0
1 2 3 4 5 6 7 8 9 10 11

Months

• It is important to select best period for the moving average

 Smoothing Effects for given periods
Actual
140 3 Week
120
100
80
Orders

60
40
20
0
1 2 3 4 5 6 7 8 9 10 11

Months

• It is important to select best period for the moving average

It can be seen from the above figure.
• Shorter the period produces more oscillation but closer following of the trend.
 Smoothing Effects for given periods
Actual 5 Week
140
120
100
80
Orders

60
40
20
0
1 2 3 4 5 6 7 8 9 10 11

Months

• It is important to select best period for the moving average

It can be seen from the above figure.
• Longer the period, greater the random element smoothened but there is
lagging in trend effect
 Smoothing Effects for given periods
Actual
160
140
120
100
80
Orders

60
40
3 Week Double Moving
20
0
1 2 3 4 5 6 7 8 9 10 11

Months
 Smoothing Effects for given periods
160
Actual 5 Week
140
3 Week
120

100
Orders

80

60

40 3 Week Double Moving

20

0
1 2 3 4 5 6 7 8 9 10 11

Months

• It is important to select best period for the moving average

It can be seen from the above figure.
• Shorter the period produces more oscillation but closer following of the trend.
• Longer the period, greater the random element smoothened but there is
lagging in trend effect
 Weighted Moving Average
• Adjusts moving average method to more closely reflect data
fluctuations

• Allows any weights to be placed to each element based on trial

and error or by past experience or by considering seasonal factor

• Sum of all the weights should be equal to one

WMA n  i 1 Wi D i
n

Where

Wi = weight for period i, between 0 and 100 percent

 Wi = 1.00
01/19/2021 Dr.ASB,CEG 111
 Weighted Moving Average

MONTH WEIGHT DATA

August 17% 130

September 33% 110

October 50% 90

November forecast
3
WMA3   Wi Di
i 1

= (0.50)(90) + (0.33)(110) + (0.17)(130)

= 103.4 orders
 Exponential Smoothing

01/19/2021 Dr.ASB,CEG 113

 Simple Exponential Smoothing

Averaging method
Random fluctuations around an average value
known as base (S)
Weights most recent data more strongly
(Weights decreased exponentially as the data
becomes older)
Reacts more to recent changes
Widely used, accurate method
 Simple Exponential Smoothing
Forecast for next period,

S t =  D t + (1 - ) S t -1

F t = St-1

Where,
S t = base value for period t
S t –1 = previous to Base value t
F t = forecast value of period t
D t = actual demand of period t
 = Weighting factor, smoothing constant 0.0  1.0

If  = 0.20, then F t = 0.20 D t + 0.80 S t-1

 
If  = 0, then F t = 0 D t + 1 S t-1= S t-1

Forecast does not reflect recent data

 
If  = 1, then F t = 1 D t + 0 S t-1 = D t

Forecast based only on most recent data

01/19/2021 Dr.ASB,CEG 115
 Exponential Smoothing

• How to choose α
– depends on the emphasis you want to place on the most
recent data

• Increasing α makes forecast more sensitive to recent data

01/19/2021 Dr.ASB,CEG 116

 Forecast Effects of
Smoothing Constant 

Ft+1 = F +  (Dt - Ft)

t

or Ft+1 =  Dt + (1- ) Dt - 1 + (1- )2Dt - 2 + ...

w1 w2 w3
Weights

=
Prior Period 2 periods ago 3 periods ago
 (1 - ) (1 - )2

= 0.10 10% 9% 8.1%

= 0.90 90% 9% 0.9%

01/19/2021 Dr.ASB,CEG 117
 Simple Exponential Smoothing Example
S =D since it is not given
0 1,

S1 = D1 + (1 - )S0
PERIOD MONTH DEMAND
1 Jan 37
= (0.30)(37) + (0.70)(37)

2 Feb 40 = 37=F2
3 Mar 41 S2 = D2 + (1 - ) S1
4 Apr 37
= (0.30)(40) + (0.70)(37)
5 May 45
= 37.9=F3
6 Jun 50
7 Jul 43
S12 = D12 + (1 - ) S11
8 Aug 47
= (0.30)(54) + (0.70)(50.84)
9 Sep 56
= 51.79= F13
10 Oct 52
11 Nov 55
12 Dec 54

01/19/2021 Dr.ASB,CEG 118

 Simple Exponential Smoothing Example
PERIOD MONTH DEMAND BASE, St FORECAST, Ft

 = 0.3  = 0.5  = 0.3  = 0.5

37.00
1 Jan 37 37 – -
38.50 37.00
2 Feb 40 37.90 37.00
39.75 38.50
3 Mar 41 38.83 37.90
38.37 39.75
4 Apr 37 38.28 38.83
41.68 38.37
5 May 45 40.29 38.28
45.84 41.68
6 Jun 50 43.20 40.29
44.42 45.84
7 Jul 43 43.14 43.20
45.71 44.42
8 Aug 47 44.30 43.14
50.85 45.71
9 Sep 56 47.81 44.30
51.42 50.85
10 Oct 52 49.06 47.81
53.21 51.42
11 Nov 55 50.84 49.06
53.61 53.21
12 Dec 54 51.79 50.84
53.61
13 Jan – -- 51.79

01/19/2021 Dr.ASB,CEG 119

 Simple Exponential Smoothing Example
The following graph shows, we can see that the higher
value of  follows more closer to the trend changes than
lesser value.

01/19/2021 Dr.ASB,CEG 120

 Exponential Smoothing (continued)

• We looked at using exponential smoothing to forecast

demand with only random variations

Ft+1 = Ft + a (Dt - Ft)

Ft+1 = Ft + a Dt – a Ft

Ft+1 = a Dt + (1-a) Ft
What if demand varies due to randomness and trend?

What if we have trend and seasonality in the data?

01/19/2021 Dr.ASB,CEG 121

 Trend

Trend causes the Forecast to lag

Assumptions:
• Demand is known
• Demand is Linear that is the demand is increasing by T units
every period

T
De
ma
nd T

1 2 3
Time

01/19/2021 Dr.ASB,CEG 122

 Moving average with trend
• If we use n periods for moving average Forecast lags by (n+1) / 2 periods
• Usually the trend is considered as the difference between the current demand and
previous demand
• Calculate moving average forecast:

D t i
M t 1  Ft ( MA)  i 1
n

• Calculate moving average of the Trend:

n

T t i
Tt  i 1
n
(n  1)
Forecast, FITt  Ft ( MA)   Tt
2
 Moving average with trend Example

• Moving average with trend Example

• Calculate moving average forecast:

D3  D2  D1
M 41  F4 ( MA)   103.3
3
• Calculate moving average of the Trend :
T4  T3  T2
T5    15
n

(n  1)
Forecast, FITt  Ft ( MA)   Tt  88.3  2  (15)  58.3
2
 Moving average with trend Example
MA
MONTH ORDERS Ft (MA) Trend Tt (MA) Ft
3 Week
-
Jan 120 – – -
- -
Feb 90 – -30
- -
Mar 100 103.3 – 10
- -
Apr 75 88.3 103.3 -25

May 110 95.0 88.3 35 -15 58.3

June 50 78.3 95.0 -60 6.67 108.34
July 75 78.3 78.3 25 -16.67 44.96
Aug 130 85.0 78.3 55 0 78.3
Sept 110 105.0 85.0 -20 6.67 98.34
Oct 90 110.0 105.0 -20 20 145
Nov – -- 110.0 -- 5 120
 Exponential Smoothing with trend

St  αD t  (1  α)(S t 1  Tt 1 )

Tt  β(S t  S t 1 )  (1  β)Tt 1

FIT t1  S t  Tt

01/19/2021 Dr.ASB,CEG 126

 Exponential Smoothing with trend
PERIOD MONTH DEMAND BASE SMOOTHED ADJUSTED SIMPLE
FORECAST EXP.SMOO
St TREND FIT  
  Tt    
1 Jan 37 37.0 0.0    
2 Feb 40 37.3 0.1 37.0 37.0
3 Mar 41 37.8 0.2 37.4 37.3
4 Apr 37 37.9 0.2 37.9 37.7
5 May 45 38.7 0.4 38.0 37.6
6 Jun 50 40.2 0.7 39.1 38.3
7 Jul 43 41.1 0.8 40.9 39.5
8 Aug 47 42.4 0.9 41.9 39.9
9 Sep 56 44.6 1.3 43.3 40.6
10 Oct 52 46.5 1.5 45.9 42.1
11 Nov 55 48.7 1.7 48.0 43.1
12 Dec 54 50.7 1.8 50.4 44.3
13 Jan       52.5 45.3

01/19/2021 Dr.ASB,CEG 127

 Exponential Smoothing with trend

60

50

40

30
DEMAND
ADJUSTED FORECAST
20 SIMPLE EXP.SMOO

10

01/19/2021 Dr.ASB,CEG 128

 Exponential Smoothing with seasonal demand

Dt
St    (1   )( S t 1 )
I t c

Dt
It    (1   ) I t c
St

Ft 1  S t  I t  c

01/19/2021 Dr.ASB,CEG 129

Exponential Smoothing with seasonal demand –method 1
Base Ratio Forecast
2011 0            
  1 53   0.70  
  2 83   1.10  
  3 95   1.25  
  4 72 303 75.75 72 0.95  
2012 1 50 71.84 0.70 50.38
  2 75 70.82 1.08 78.71
  3 102 73.97 1.29 88.82
  4 66 293 73.25 72.61 0.94 70.31
2013 1 55 74.45 0.71 50.73
  2 81 74.52 1.09 80.75
  3 93 73.76 1.28 96.24
  4 76 305 76.25 75.94 0.96 69.19
2014 1       53.97
  2       82.43
  3       97.38
01/19/2021
  4   303.8 75.94
Dr.ASB,CEG
    72.66 130
 Exponential Smoothing with seasonal demand

120

100

80

60 demand
Forecast

40

20

0
1 2 3 4 5 6 7 8 9 10 11 12

01/19/2021 Dr.ASB,CEG 131

Exponential Smoothing with seasonal demand –method2
Base Ratio Forecast
2011 0            
  1 53 0.70  
  2 83 1.10  
  3 95 1.25  
  4 72 303 303 75.75 0.95  
2012 1 50   51.25
  2 75   80.26
  3 102   91.86
  4 66 293 300 73.25   69.62
2013 1 55   53.35
  2 81   83.55
  3 93   95.63
  4 76 305 301.5 76.25   72.48
2014 1     52.74
  2     82.59
  3     94.53
  4     301.5 75.38   71.64
01/19/2021 Dr.ASB,CEG 132
 Exponential Smoothing with seasonal demand

120

100

80

DEMAND
60
METHOD 2
METHOD 1

40

20

0
1 2 3 4 5 6 7 8 9 10 11 12

01/19/2021 Dr.ASB,CEG 133

 Exponential Smoothing with trend and seasonality
component

Tt   ( St  St 1 )  (1   )Tt 1

Dt
It    (1   ) I t c
St
Dt
St    (1   )( S t 1  Tt 1 )
I t c

Ft 1  ( S t  Tt )  I t  c 1

01/19/2021 Dr.ASB,CEG 134

 Exponential Smoothing with trend and seasonality
component
Year Sl.No Actual Total SI Base Trend Ratio Forecast
2011 1 53     0.70  
  2 83     1.10  
  3 95     1.25  
  4 72 303 75.75 72 0 0.95  
2012 1 50 71.84 -0.03 0.70 50.38
  2 75 70.80 -0.23 1.08 78.68
  3 102 73.80 0.41 1.29 88.50
  4 66 293 73.25 72.78 0.13 0.94 70.53
2013 1 55 74.56 0.46 0.71 50.93
  2 81 74.59 0.37 1.09 81.38
  3 93 73.80 0.14 1.28 96.90
  4 76 305 76.25 75.98 0.55 0.96 69.31
2014 1         54.36
  2         83.05
  3         98.18
  4   306.1 76.53       73.18

01/19/2021 Dr.ASB,CEG 135

 Exponential Smoothing with trend and seasonality component

120

100

80

Demand
60 EXP WITH SD AND TREND
METHOD1
METHOD2

40

20

0
1 2 3 4 5 6 7 8 9 10 11 12

01/19/2021 Dr.ASB,CEG 136

 Regression Analysis as a Method for Forecasting
Regression analysis takes
advantage of the relationship
between two variables. Demand
is then forecasted based on the
knowledge of this relationship
and for the given value of the
related variable.

Ex: Sale of Tires (Y), Sale of Autos

(X) are obviously related

If we analyze the past data of

these two variables and establish
a relationship between them, we
may use that relationship to
forecast the sales of tires given
the sales of automobiles.

The simplest form of the

relationship is, of course, linear,
hence it is referred to as a
regression line.
01/19/2021 Dr.ASB,CEG 137
 Linear Trend Line (Least Squares)

y  a  bx
 Where
a = intercept (at period 0)
b = slope of the line

a  y  bx b
 xy  n x y
 x  nx 2 2

x = the time period

y = forecast for demand for period x
n = number of periods

x
 x
 mean of the xValues y
 y
 mean of the y Values
n n
01/19/2021 Dr.ASB,CEG 138
 Linear Trend Line (Least Squares)
78
X Y xy x2 x   6.5
12
(PERIOD) (DEMAND)
1 73 73 1 557
y  46.42
2 40 80 4 12
3
4
41
37
123
148
9
16 b
 xy  n x y
 x  nx 2 2
5 45 225 25
6 50 300 36 3903 (12)(6.5)(46.42)
b
7 43 301 49 650  12(6.5)2
8 47 376 64
= 1.72
9 56 504 81
10 52 520 100 a  46.42  1.72(6.5)  35.2
11 55 605 121 y = 35.2 + 1.72x
12 54 648 144 Forecast for period 13
78 557 3903 650 y = 35.2 + 1.72(13)
01/19/2021 Dr.ASB,CEG y = 57.56 units 139
 Linear Trend Line (Least Squares)

01/19/2021 Dr.ASB,CEG 140

 Numerical problems

Month Advertising Sales

January 3 1

February 4 2

March 2 1

April 5 3

May 4 2

June 2 1

July    
01/19/2021 Dr.ASB,CEG 141
 MonthAdvertising Sales X 2 XY
January 3 1 9.00 3.00
February 4 2 16.00 8.00
March 2 1 4.00 2.00
April 5 3 25.00 15.00
May 4 2 16.00 8.00
June 2 1 4.00 2.00
July

TOTAL 20 10 74 38

01/19/2021 Dr.ASB,CEG 142

 To Use a Forecasting Method

• Collect historical data

• Select a model
– Moving average methods
• Select n (number of periods)
• For weighted moving average: select weights
– Exponential smoothing
• Select 

• Selections should produce a good forecast

01/19/2021
…but what is a good forecast?
Dr.ASB,CEG 143
 A Good Forecast

Has a small error

Error = Demand - Forecast

01/19/2021 Dr.ASB,CEG 144

 Measures of Forecast Error
a. MAD = Mean Absolute Deviation
et
n

 D t - Ft
MAD = t =1
n

b. MSE = Mean Squared Error

n

 t t
D - F  2

MSE = t =1
n

c. RMSE = Root Mean Squared Error

RMSE = MSE

 Ideal values =0 (i.e., no forecasting

01/19/2021 Dr.ASB,CEG error) 145
 n

MAD Example
D t - Ft = 40 =10
MAD = t =1
4
n

What
What isis the
the MAD
MAD value
value given
given the
the forecast
forecast
values
values in
in the
the table
table below?
below?
At Ft
Month Sales Forecast |Dt – Ft|
1 220 n/a
2 250 255 5
3 210 205 5
4 300 320 20
5 325 315 10
n
01/19/2021 Dr.ASB,CEG  D - F = 40
t =1
t t 146
 = 550 =137.5
MSE/RMSE Example
4

What
What isis the
the MSE
MSE value?
value? RMSE = √137.5
=11.73
Dt Ft
Month Sales Forecast |Dt – Ft| (Dt – Ft)2
1 220 n/a
2 250 255 5 25
3 210 205 5 25
4 300 320 20 400
5 325 315 10 100
01/19/2021 Dr.ASB,CEG
= 147550
 Measures of Error

1. Mean Absolute Deviation

t Dt Ft et |et| et 2 (MAD)
Jan 120 100 20 20 400 n

 et
84
Feb 90 106 -16 256 MAD  1 = 14
n
16
6
Mar 101 102 -1 1 1
2a. Mean Squared Error (MSE)
April 91 101 -10 10 100 n

May 115 98

 te  2

17 17 289
MSE  1 1,446
n = 241
June 83 103 -20 20 400 6
2b. Root Mean Squared Error
-10 84 1,446 (RMSE)
An accurate forecasting system will have small MAD, MSE and RMSE; ideally RMSE  MSE
equal to zero. A large error may indicate that either the forecasting method
used or the parameters such as α used in the method are wrong.
= SQRT(241)
Note: In the above, n is the number of periods, which is 6 in our example
01/19/2021 Dr.ASB,CEG =15.52 148
 Forecast Bias or Control

• How can we tell if a forecast has a positive or negative

bias?

Reasons for out-of-control forecasts

Change in trend

Appearance of cycle

Weather changes

Promotions

Competition

Politics
01/19/2021 Dr.ASB,CEG 149
30
 Tracking Signal

• Compute each period

• Compare to control limits
• Forecast is in control if within limits

 D  F  E
Tracking Signal  t
t

MAD MAD

Use control limits of + / - 2 to + / - 5 MAD

01/19/2021 Dr.ASB,CEG 150

 Tracking Signal
PERIOD DEMAND FORECAST, ERROR E = MAD TRACKING
Dt Ft Dt - Ft (Dt - Ft) SIGNAL

1 37 37.00 – – – –
2 40 37.00 3.00 3.00 3.00 1.00
3 41 37.90 3.10 6.10 3.05 2.00
4 37 38.83 -1.83 4.27 2.64 1.62
5 45 38.28 6.72 10.99 3.66 3.00
6 50 40.29 9.69 20.68 4.87 4.25
7 43 43.20 -0.20 20.48 4.09 5.01
8 47 43.14 3.86 24.34 4.06 6.00
9 56 44.30 11.70 36.04 5.01 7.19
10 52 47.81 4.19 40.23 4.92 8.18
11 55 49.06 5.94 46.17 5.02 9.20
12 54 50.84 3.15 49.32 4.85 10.17
01/19/2021 Dr.ASB,CEG 151
 Tracking Signal

01/19/2021 Dr.ASB,CEG 152

 General Guiding Principles for Forecasting
1. Forecasts are more accurate for larger groups of items.
2. Forecasts are more accurate for shorter periods of time.
3. Every forecast should include an estimate of error.
4. Before applying any forecasting method, the total system
should be understood.
5. Before applying any forecasting method, the method should
be tested and evaluated.
6. Be aware of people; they can prove you wrong very easily in
forecasting

01/19/2021 Dr.ASB,CEG 153

Those who have knowledge, don't predict. 
Those who predict, don't have knowledge.
01/19/2021
-Lao Tzu, 6th Century BC.154
Dr.ASB,CEG