Download as pdf or txt
Download as pdf or txt
You are on page 1of 43

http://www.sciedu.ca/journal/index.

php/bmr/article/viewFile/2713/1574
Because forecasts are relied upon as if they are accurate ..

•  59% of respondents believed that sales forecasts were only


“somewhat accurate”
•  67% of respondents report an absence of accountability for
forecast accuracy!
•  Dynamic external, dynamic internal and ‘manipulative
internal factors’
•  Unintended consequences .. (of commission compensation
structures)
“show me how I’m measured
and I’ll show you how I act.”

•  investigating how sales compensation systems can lead to


lower firm performance through inaccurate and
manipulative forecasts by the sales force

•  agency theory models the relationship between one who


assigns responsibilities (the principal) and one who fulfils
them (the agent, which in this case is the sales person).
Conflict or goal incongruence arises from the contract that
governs this relationship.
“show me how I’m measured
and I’ll show you how I act.”
•  Sales based commission compensation structures (including
bonuses) based on sales revenue will predict excess
inventory

•  Sales representatives will increase forecasts for new


product introductions if they believe their existing product
portfolios are inadequate

•  Sales forecast audits will reduce excess inventory

- We will see later how some of these inflated forecasts arise


from how we structure relationships across the supply chain!
Key issue:
“demand for fashion products cannot be forecast!”
“Too much” and “too
little” costs
*  Co = overage cost
*  The cost of ordering one more unit than what you would have ordered had
you known demand.
*  In other words, suppose you had left over inventory (i.e., you over
ordered). Co is the increase in profit you would have enjoyed had you
ordered one fewer unit.
*  Cu = underage cost
*  The cost of ordering one fewer unit than what you would have ordered had
you known demand.
*  In other words, suppose you had lost sales (i.e., you under ordered). Cu is
the increase in profit you would have enjoyed had you ordered one more
unit.

13
In the fashion industry, products are usually characterized by
long replenishment lead times, short selling seasons and
nearly unpredictable demand and therefore, inaccurate
forecasts.

[Good summary of industry trends, echoing earlier case
material]

.. no largely recognized link between demand in fashion and a
specific attribute or pattern. Only Bartezzaghi [31] hazards a
guess on the basis of correlation as a cause of lumpiness.

Correlation may be, for example, due to imitation in fashion,
which will lead to sudden peaks in demand.
A pattern-mining approach – looking at shape, and ‘hazarding
a guess that it may be predictable?

Demand Demand

Interval 1 Interval 2

Period
requirements

Average
Requirements


-  Hence Average inter-Demand Interval (ADI)
-  Coefficient of variation (CV) of period requirements divided
by average requirements
A pattern-mining approach –
looking at shape, and ‘hazarding
a guess that it may be
predictable?











.
A pattern-mining approach –
looking at shape, and ‘hazarding
a guess that it may be
predictable?


A matrix that allows type of
demand to be used to suggest
type of forecasting method
(about which the authors offer
little insight)
Key insights?
•  “it is logical to presume that the main attribute of demand in
the fashion industry is in fact lumpiness”
•  “Zara has just demonstrated that colour is a more important
feature than model or type of clothes”
Conclusion?
SUPPLY CHAIN
MANAGEMENT

DEMAND MANAGEMENT
AIMS:

*  To know about different types of demand


*  How demand can be presented?
*  To have clear view about demand error and its
consequences
*  To have clear view about the position of Demand
Management in Supply Chain Management
*  To know more about the function of sales and
demand forecasting
Supplier A
Supplier A
Supplier A
Supplier A
Supplier A
Goods & Services
Customer

FLOW

INFORMATION

Supply Chain
Management

Demand
Supply chain
chain

Procurement Demand
Management Management
Value Chain
FIRM INFRASTRUCTURE
Support Activities

HUMAN RESOURCE MANAGEMENT


TECHNOLOGY DEVELOPMENT
PROCUREMENT

Supply Chain Demand Chain

SER
VIC
INBOUND OUTBOUND MARKETING & ES
OPERATIONS
LOGISTICS LOGISTICS SALES

Primary Activities
Redrawn and modified from (PORTER, 1985, p. 38)
Demand management

The main responsibility of any demand management


section is to manage and to supervise the activities in
the demand value chain

The major three activities by demand management is
as follows:
*  Demand management
*  Demand planning
*  Sales forecasting management
(Mentzer and Moon, 2005)
Demand types

Demand
types

Independent Derived Dependent


Demand Demand Demand
Demand types

*  Independent demand:

“Any supply chain has only one point of independent demand:


the amount of product demanded (by time and location) by the
end-use customer of the supply chain. Whether this end-use
customer is a consumer shopping in a retail establishment or
online (B2C), or a business buying products for consumption
in the process of conducting its business operations (B2B),
these end-use customers determine the true demand for the
product that will flow through the supply chain”
(Mentzer and Moon, 2005, p. 3)
Demand types

*  Derived demand:

“The second type of supply chain demand is called
derived demand, because it is not the independent
demand of the end-use customer, but rather a
demand that is derived from what other companies in
the supply chain do to meet their demand from their
immediate customer (i.e., the company that orders
from them).”
(Mentzer and Moon, 2005, p. 3)
Demand types

*  Dependent demand:
“Dependent demand is the demand for the
component parts that go into a product. […],
this is usually demand that is dependent upon
the demand for the product in which it is a
component. “
(Mentzer and Moon, 2005, p. 3)
“ It is important to note that only one company in any given
supply chain is directly impacted by independent demand. The
rest of the companies in the supply chain are impacted by
derived and/or dependent demand”
(Mentzer and Moon, 2005, p. 4)
Supplier A
Supplier A
Supplier A
Supplier A
Supplier A Independent
COMPANY Demand
Customer

Supplier A
Supplier A
Supplier A
Supplier A
Supplier B
Demand Error

*  Traditional supply chains:


*  Chain of end customers, retailers, wholesalers
Manufactures and suppliers
1,000 units 1,100 units 1,331units 1,464units
1,210 units

Retailer Wholesaler Manufacturer Supplier

Customer
10% error 10% error 10% error 10% error

Redrawn from (Merter and Moon, 2005, p. 5)

*  Network style supply chain


*  Behavioural Issues – what could we get away with,
who suffers?
Coordinated Demand

Demand error inventory (safety stock) was 1105, now 400!

1,000 units 1,100 units 1,100 units 1,100 units


1,100 units

Retailer Wholesaler Manufacturer Supplier

Customer
10% error 0% error 0% error 0% error

Redrawn from (Merter and Moon, 2005, p. 5)


*  Behavioural Issues – what could we get away with, who
suffers (how does the Retailer notice)?
*  What are the risks to companies in the Supply Chain?
Demand Management
Supply Chain
Management
- the creation across
the supply chain and
its markets of a
coordinated flow of Demand
Management
demand.

Marketing / Supply
Demand
Chain Relationship
Planning
Management

supply chain relationship management


= sharing rewards for behavioural reasons! Sales Forecasting
Redrawn from (Merter and Moon, 2005, p. 8) Management
forecasting
Forecasting is not easy!
"It will allow us to add more precision, more detail, more accuracy to our
forecasts on all 6me scales for tomorrow, for the next day, next week, next
month and even the next century," said Met Office chief execu6ve Rob Varley.
Demand Management
Supply Chain
Management
- the creation across
the supply chain and
its markets of a
coordinated flow of Demand
Management
demand.

Marketing / Supply
Demand
Chain Relationship
Planning
Management

supply chain relationship management


= sharing rewards for behavioural reasons! Sales Forecasting
Redrawn from (Merter and Moon, 2005, p. 8) Management
DEMAND
THE AIM IS TO ATTEMPT TO UNDERSTAND
CUSTOMER DEMAND IN THE ABSENCE OF
‘REAL’ INFORMATION

WHY?

NEED TO MAKE DECISIONS AND TO ACT
TODAY…..
…ABOUT MEETING FUTURE BUSINESS
DEMANDS

Production – Purchasing - Design
A good example of over-optimistic sales forecasting
… leading games developer Eidos which has had to admit that sales forecasts for its main
product have proved to be over-optimistic…
The product development process for games console titles is fraught with risk. All the
development costs come up front, with many titles taking 18-24 months to be designed and
tested before launch to the consumer market. Getting the sales forecast right for such
products is important. The games developer needs to be reasonably certain that a new title
will earn a satisfactory rate of return. For Eidos, it could have used previous sales volumes of
the earlier Tomb Raider titles as a guide. It would also have discussed market conditions with
its distributors in key consumer markets to get their views on likely sales volumes.
For Eidos, their main product is the Lara Croft Tomb Raider franchise. The latest installment -
Tomb Raider Underworld - was launched in November 2008. However, although the product
has sold substantial quantities, these have been less than the market, and Eidos
management, were forecasting. In fact the sales shortfall so far is around £20m.
Eidos has admitted it sold only 1.5m copies of Underworld between 18 November, when the
game was launched, and the end of 2008. Eido’s retail distributors have also had to offer
heavy price discounts in order to shift the product.
The announcement of the sales shortfall triggered a profits warning from Eidos and an
admission that the company may breach key loan agreements with its banks. Not a good
sign. The news marked a torrid 2008 in which 92 per cent has been wiped off the Eidos share
price.
SOURCE:
http://beta.tutor2u.net/business/blog/a-good-example-of-over-optimistic-sales-forecasting
DEMAND
BUT HOW TO KNOW ABOUT THE FUTURE?
Crystal Ball?
Intuition?

OR IS THERE A BETTER WAY?
Is there a pattern?
Can the pattern be modelled?
What assumptions need to be made?

PROBLEM = UNCERTAINTY
CHALLENGE = ACCURACY
DEMAND
What factors affect the customer demand?

-  Weather?
-  Fashion?
-  Economic conditions?
-  New technologies?
-  Political developments?
-  Demographic changes?
-  Competitors?
-  ‘events’?
-  ????


Can the past / present reveal future behaviour?
SYMPTOMS [leading indicators]? (present)
REGULARITIES [laws, theories]? (past)
Walmart Example

http://www.youtube.com/watch?v=SUe-tSabKag#t=3m25s

SAS Example

“This demonstration explained how you can use SAS to analyze and review
predicted forecast values until they align with your business goals.”

https://www.youtube.com/watch?v=it_QCC5AE1o
“There is no method of prediction that is best for all cases”
Professor Robert Fildes (‘Father of Forecasting’)

‘knowing which one is best gives much scope for confusion, using them
without any knowledge is very dangerous ..!’
It was embraced ardently by investors, banks and ratings agencies like
Moody’s and Standard and Poor’s. It was even enshrined in the
regulatory framework for Basel II to determine capital requirements for
“There is no method of prediction that is best for all cases”
banks with structured credit on their books.
Professor Robert Fildes (‘Father of Forecasting’)
It was as fertile as it was popular. It ballooned a family of credit
instruments like collaterized debt obligations and collaterized loan
obligations which, in turn, gave birth to really fancy derivatives like
‘knowing which one is best gives much scope for confusion, using them
collaterized debt obligations squared–or CDOs that invested in other
without any knowledge is very dangerous ..!’
CDOs.
So, eventually, several trillion dollars were invested in these things. And
why not? The risks were known and therefore under control.

Well … the risks weren’t known and were, in fact, totally out of control.
As soon as prices in the housing market started to swoon default
correlations shot for the moon. What were once low numbers became
fatally high, triggering a chain reaction in exploded paper.
-  For those interested:
•  ARIMAX = “Auto Regressive Integrated
Moving Average with Exogeneous Input”
•  UCM = Unobserved Component Models
•  ESM = Exponential Smoothing Models

- How do we make sense of these when
behavioural inputs are as important as method?

02 Sep 2014: International Valuation Standards Council (IVSC)


issues revised guidance on the valuation of equity derivatives:

“Removal of formulae in favour of relying on descriptive narratives
of concepts. This change reflects concern about the practical
problems with formulae, including accounting for relationships
that cannot be expressed through concise mathematical
expressions”

Source: http://www.iasplus.com/en-gb/news/2014/09/ivsc
Demand Patterns
Quantity

Quantity
Time Time

(a) Horizontal: Data cluster about a horizontal line (b) Trend: Data consistently increase or decrease

Year 1

Quantity
Quantity

Year 2

| | | | | | | | | | | | | | | | | |
J F M A M J J A S O N D 1 2 3 4 5 6
Months Years
(c) Seasonal: Data consistently show peaks and valleys (d) Cyclical: Data reveal gradual increases and
decreases over extended periods

ALSO (e) RANDOM DEMAND PATTERN


Krajewski, Ritzman, Malhootra (2010 ) Copyright © 2010 Pearson Education, Inc. Publishing as Prentice Hall.
DEMAND
APPROACHES
ACTUAL ORDERS
FORECASTS
-  Judgemental
-  Quantitative

BY
Product item, family,
Customer
Geographical location

Time horizon (short, medium, long)

Pessimistic, optimistic, most likely,
Georgoff, D.M. & Murdick, R.G. (1986)
Manager’s Guide to Forecasting. Harvard
FORECASTING Business Review, (Jan-Feb ) 110-120
Every forecasting situation is limited by constraints like time,
funds, competencies, or data:

Present 20 techniques, evaluated on 16 dimensions


DIMENSIONS QUESTIONS
Time Span Is the forecast period a: present need or short-, medium-, or long-term projection?
Urgency Is the forecast needed immediately?
Frequency Are frequent forecast updates needed?
Resource Mathematical sophistication Are quantitative skills required?
requirements
Computer Are computer capabilities limited?
Financial Are only limited financial resources available?
Input Antecedent Are only limited past data available?
Variability Does the primary series fluctuate substantially?
Internal consistency Are significant changes in management decisions expected?
External consistency Are significant environmental changes expected?
External stability Are significant shifts expected among variable relationships?
Output Detail Are component forecasts required?
Accuracy Is a high level of accuracy critical?
Capability for reflecting direction changes Should turning points be reflected promptly?
Capability for detecting direction changes Should turning points be reflected early?
Form Is an interval or probabilistic forecast critical?
Judgement methods Naive extrapolation Simple subjective extension of current events into future

Sales force composite Compilation of sales force forecasts of expected sales in each of their areas

Jury of execution opinion Consensus of a jury of experts from different functions from within the company

QUALITATIVE Scenario methods Construction of snap-shot narratives about possible futures .

Delphi technique Iterative approach to seek views from experts on an issue and give feedback about each others
responses to establish whether views change

Historical analogy Using similar historical / past event to predict

Counting methods Market testing Consumer market Establishes attitudinal & purchase intentions from consumers
survey

Market survey Industrial market Establishes attitudinal & purchase intentions from more knowledgeable respondents
survey

Time series methods Moving averages Averaging recent values to predict future outcomes

Exponential smoothing Combines recent values and estimated forecast using weights to determine future outcome

Adaptive filtering Used a weighted combination of actual and estimates but ‘systematically altered to reflect data
pattern changes’

Time series extrapolation Uses least squares function applied to data series with time as independent variable

Time series decomposition Predicts ‘ expected outcomes from trend, seasonal, cyclical and random components, which are
isolated from a data series.

Box-Jenkins Complex ‘iterative procedure that produces an autoregressive, integrated moving average
model, adjusts for seasonal and trend factors, estimates appropriate weighting parameters, tests
the model, and repeats the cycle as appropriate

Association (Causal Correlation methods Uses historical patterns of co-variation between variables
methods) Uses a ‘predictive equation derived by minimising the residual variance of one or more
Regression models
predictor (independent) variables’

Leading indicators Uses ‘one or more preceding variables that is systematically related to the variable to be
predicted’

Econometric models Uses ‘an integrated system of simultaneous equations that represent relationships among
elements of the national economy derived from combining history and economic theory’

Input-output models Uses ‘a matrix model that indicates how demand changes in one industry can directly and
cumulatively affect other industries

Georgoff, D.M. & Murdick, R.G. (1986) Manager’s Guide to Forecasting. Harvard Business Review, (Jan-Feb ) 110-120
Resources

*  MENTZER, J. T. & MOON, M. A. 2005. Sales forecasting


management: a demand management approach. 2nd ed.
Thousand Oaks, CA, Sage Publications.
*  Georgoff, D.M. & Murdick, R.G. 1986. Manager’s Guide to
Forecasting. Harvard Business Review, (Jan-Feb ) 110-120

*  Chambers, John C.; Mullick, Satinder K.; Smith, Donald D.1971.


How to choose the right forecasting technique. Harvard
Business Review, (Jul-Aug) 49(4), 45-70

You might also like