TME351: Spare Parts and Storage

Management
Part II: Spare Parts Demand Forecasting
(Lecture 5 – 13rd October, 2018)
Dr. Sameer Al-Dahidi
Assistant Professor, Exchange Coordinator
Mechanical and Maintenance Engineering Department
German Jordanian University

Dr. Sameer Al-Dahidi

 Context of the Course

Maintenance in brief

Part I
Spare Parts

Spare Parts
Maintenance &
Classifications of Spare Parts

Dr. Sameer Al-Dahidi

SPs Demand Patterns

Forecasting Key Decisions

Forecasting Techniques
Part II

Forecasting

Forecast Error and TS

Spare Parts Demand

Forecasting as a Process

Inventory and Supply Chains

TME351

Inventory Costs

Exercises Sessions
Economic Order Quantity
Part III

Types of Inventory/Measures
TME351 Course Tree Plan

Inventory Control Systems

Managing Inventories

Special Inventory Models

Part IV
Spare Parts
Management
 Part II: Spare Parts Demand Forecasting

• Spare Parts Demand Patterns

• Forecasting Key Decisions

• Forecasting Techniques

• Forecast Error

• Tracking Signals

• Forecasting as a process

Dr. Sameer Al-Dahidi

 Part II: Spare Parts Demand Forecasting
Key Decisions on Making Forecasts

Before using forecasting techniques, a manager must make two decisions:

1) What to forecast?

2) What type of forecasting technique to select for different items?

Dr. Sameer Al-Dahidi

 Part II: Spare Parts Demand Forecasting
Key Decisions on Making Forecasts

Before using forecasting techniques, a manager must make two decisions:

1) What to forecast?
• Level of Aggregation
• Units of Measurements

2) What type of forecasting technique to select for different items?

Dr. Sameer Al-Dahidi

 Key Decisions on Making Forecasts
1) Deciding what to forecast

• Level of Aggregation:
 Few companies err by more than 5% when forecasting the annual total demand
for all their services or products.
 However, errors in forecasts for individual items and/or with “shorter time
periods” may be much higher.
 Recognizing this reality, many companies use a two-tier forecasting system.
o They first cluster (or roll up) several similar services or products in a process
called aggregation, making forecasts for families of services or goods that
have similar demand requirements and common processing, labor, materials
requirements.
o Next, they derive forecasts for individual items, which are sometimes called
Stock-Keeping Unit (SKU).

*SKU is an individual items of product that has an identifying code and is held in
inventory somewhere along the supply chain, such as in a distribution center.

Dr. Sameer Al-Dahidi

 Key Decisions on Making Forecasts
1) Deciding what to forecast

• Units of Measurements:
 Rather than using dollars as the initial unit of measurement, forecasts often begin
with service or product units, such as SKUs, express package to deliver, or
customers needing maintenance service or repairs for their cars.
 Forecasted units can then be translated to dollars by multiplying them by the unit
price.
 If accurate forecasting demand for a service or product is not possible in terms of
number of units, forecast the standard labor or machine-hours required of each of
the critical resources.

Dr. Sameer Al-Dahidi

 Key Decisions on Making Forecasts
2) Choosing the Type of Forecasting Technique

• Forecasting systems offer a variety of techniques, and no one of them is best for all
items and situation.
• The forecaster’s objective is to develop a useful forecast from the information at hand
with the technique that is appropriate for the different patterns of demand.
• Two general types of forecasting techniques are used:
 Judgment methods
 Quantitative methods:
 Causal methods
 Time series analysis
 Trend projection with regression

Dr. Sameer Al-Dahidi

 Part II: Spare Parts Demand Forecasting

• Spare Parts Demand Patterns

• Forecasting Key Decisions

• Forecasting Techniques

• Forecast Error

• Tracking Signals

• Forecasting as a process

Dr. Sameer Al-Dahidi

 Illustrative example for FORECASTING

Dr. Sameer Al-Dahidi

 Forecasting Techniques
Illustrative Example

• Let us assume that we have the demand of a certain product for the last 6 years from
2010-2015 as follows:
Year Demand
2014 25
2015 32
2016 24
2017 28
2018 26
2019 27
2020 ?

• A product could be any spare part required, machinery, an automobile, etc. and these
numbers could be at multiple of thousands, etc. For the sake of simplicity, we can
consider these numbers as reported from 2014-2019.
Dr. Sameer Al-Dahidi
 Forecasting Techniques
Illustrative Example

• Now, the objective is to predict (estimate) the demand for 2020 year.
• This demand estimation is of paramount important because this estimation is the
starting point for production planning, capacity building, etc.
• So, all of such activities start with the forecasting so it is worth to spend some time to
better understand how the forecasting can be performed.

Dr. Sameer Al-Dahidi

 Forecasting Techniques
Illustrative Example

• Now let us assume that we have available possible forecasts carried out by different
experts in the filed. Their estimations are stated below with their reasoning:

Forecast for 2020 Reasoning

𝑭𝟏 = 𝟐𝟕 Simple average of the whole demand from 2014-2019
𝑭𝟐 = 𝟐𝟔. 𝟐𝟓 Average of the latest 4 demand
𝑭𝟑 = 𝟐𝟖 Increase based on the last 2 values (Trend)
𝑭𝟒 = 𝟑𝟎 Based on other factors
𝑭𝟓 = 𝟐𝟔 Decrease and increase (Pattern)
𝑭𝟔 = 𝟐𝟕 Average of the latest 3 values
Weighted average using the latest 3 demands (with weights 3
𝑭𝟕 = 𝟐𝟔. 𝟖𝟑𝟑
(recent), 2, 1 (old))
𝑭𝟖 = 𝟐𝟔 Average of the whole demands after removing the outlier 32

Dr. Sameer Al-Dahidi

 Forecasting Techniques
1) Judgment methods

A forecasting method that translates the opinions of managers, experts opinions, and
maintenance technicians estimates into quantitative estimates.

• Forecasts from quantitative methods are possible only when there is a adequate
historical data, i.e., the history file.
• However, the history file may be nonexistence when a new product is introduced or
when technology is expected to change.
• The history file might exist but be less useful when certain events (such as rollouts or
special packages) are reflected in the past data, or when certain events are expected to
occur in the future.
• In some cases, judgment methods can also be used to modify forecasts that are
generated by quantitative methods.
• They may recognize that one or two quantitate methods have been performing
particularly well in recent periods. Adjustments certainly would be called for if the
forecaster has important contextual knowledge.

Dr. Sameer Al-Dahidi

 Forecasting Techniques
1) Judgment methods

A forecasting method that translates the opinions of managers, experts opinions, and
maintenance technicians estimates into quantitative estimates.

• Contextual Knowledge is knowledge that practitioners gain through experience, such as

cause-and-effect relationships, environmental signs, and organizational information that
may have an effect on the variable being forecast.
• Adjustments also could account for unusual circumstances, such as a new sales
promotion or unexpected international events.
• They could also have been used to remove the effect of special one-time events in the
history file before quantitative methods are applied.
• Four of the more successful judgment methods are as follows:
 Salesforce estimates
 Executive opinion
 Market research
 Delphi method.

Dr. Sameer Al-Dahidi

 Forecasting Techniques
2) Quantitative methods

2.1 Causal method: is a quantitative forecasting method that uses historical data on
independent variables (also called causal variables or predictors), such as promotional
campaigns, economic conditions, competitors’ actions, and the relation of viscosity of
product manufactured on a machine with demand of stirrer parts, to predict dependent
variable (also called response variable), that is the demand.

2.2 Time-series analysis: is a statistical approach that relies heavily on historical demand
data to project the future size of demand and recognizes trends and seasonal patterns.

2.3 Trend projection using regression: is a hybrid between a time-series analysis

technique and the causal method.

Dr. Sameer Al-Dahidi

 Quantitative Forecasting Techniques
2.1) Causal methods

• Are used when historical data are available and the relationship between the factor to be
forecasted and other external or internal factors (e.g., government actions or advertising
promotions) can be identified. For example, car repair parts is a function of weather
conditions, promoted item is a function of discount and advertisements, etc.
• These relationships are expressed in mathematical terms and can be complex.
• Causal methods are good for predicting turning points in demand and for preparing
long-range forecasts.
• Linear regression is here presented as one of the best known and most commonly used
causal methods.

Dr. Sameer Al-Dahidi

 2.1) Causal methods
Linear Regression

• A method in which one variable (dependent variable) is related to one or more

independent variables by a linear equation. In the simplest linear regression model,
the dependent variable is a function of only one independent variable and the
mathematical relationship is a straight line:

𝑌 = 𝑎 + 𝑏𝑋

• Dependent variable (𝑌), such as demand for door hinges, is the variable that the
manager wants to forecast.

• The independent variables (𝑋 ′ 𝑠), such as advertising expenditures, are assumed to

affect the dependent variable and thereby cause the results observed in the past.

Dr. Sameer Al-Dahidi

 2.1) Causal methods
Linear Regression

• Example shows how linear regression line relates to the data.

𝑿′ 𝒔 𝒀
Dependent variable

𝑋1 𝑌1

… …

𝑋𝑖 𝑌𝑖

𝑋𝑛 𝑌𝑛

𝑖 = 1, … , 𝑛 𝑑𝑎𝑡𝑎 𝑝𝑜𝑛𝑡𝑠

𝑋
Independent variable

Dr. Sameer Al-Dahidi

 2.1) Causal methods
Linear Regression

• Example shows how linear regression line relates to the data. In technical terms, the
regression line minimizes the squared deviations from the actual data.
𝑌

Regression equation:
Dependent variable

𝑌 = 𝑎 + 𝑏𝑋

𝑋
Independent variable

Dr. Sameer Al-Dahidi

 2.1) Causal methods
Linear Regression

• Example shows how linear regression line relates to the data. In technical terms, the
regression line minimizes the squared deviations from the actual data.
𝑌

Regression equation:
Dependent variable

𝑌 = 𝑎 + 𝑏𝑋

Value of X used to
estimate Y
𝑋
Independent variable

Dr. Sameer Al-Dahidi

 2.1) Causal methods
Linear Regression

• Example shows how linear regression line relates to the data. In technical terms, the
regression line minimizes the squared deviations from the actual data.
𝑌

Regression equation:
Dependent variable

𝑌 = 𝑎 + 𝑏𝑋
Estimate of Y from
regression equation

Value of X used to
estimate Y
𝑋
Independent variable

Dr. Sameer Al-Dahidi

 2.1) Causal methods
Linear Regression

• Example shows how linear regression line relates to the data. In technical terms, the
regression line minimizes the squared deviations from the actual data.
𝑌

Regression equation:
Dependent variable

𝑌 = 𝑎 + 𝑏𝑋

Deviation or error

Actual value of Y

𝑋
Independent variable

Dr. Sameer Al-Dahidi

 Linear Regression Revisions
How to find a and b of a regression line

σ𝑛𝑖=1 𝑌𝑖 σ𝑛𝑖=1 𝑋𝑖
𝒂= −𝒃
𝑛 𝑛
𝑛 σ𝑛𝑖=1 𝑋𝑖 . 𝑌𝑖 − σ𝑛𝑖=1 𝑋𝑖 σ𝑛𝑖=1 𝑌𝑖
𝒃= 2
𝑛 σ𝑛𝑖=1 𝑋𝑖2 − σ𝑛𝑖=1 𝑋𝑖

Dr. Sameer Al-Dahidi

 References

[1] NPTEL, Operations and Supply Chain Management by Prof. G. Srinivasan,

Department of Management Studies, IIT Madras.

[2] Spare Parts and Storage Management, Winter 2017/2018, Eng. Dina, German
Jordanian University.

[3] Search in www.google.com

[4] Text Book – Part 3 (Designing and Managing Supply Chains), Chapter 14
(Forecasting Demand).

Dr. Sameer Al-Dahidi

 Thank You
Dr. Sameer Al-Dahidi
Assistant Professor, Exchange Coordinator
Mechanical and Maintenance Engineering Department
GJU
sameer.aldahidi@gju.edu.jo

26
Dr. Sameer Al-Dahidi