162, Production and Operations Management
12.6 APPLICATIONS OF FORECASTING
Forecasting is used to accurately estimate or predict future trends, such as demand for the next
‘month or year. Earlier in the chapter, we discussed some of the potential areas of management
where forecasting can play a significant role (see Section 12,3). Table 12.1 lists the areas and
the main application within each area
Table 12.1. Application Areas of Forecasting
area ‘Application
[Accounting Estimation of CostProft
Finance ‘Cash flew and funding
Human Resource | Hiringrectutngraining’seminar
Marketing Pricing (statiefsynamic), strategy
Mis. IT services
[Operations Production schedules, MRP, agaregale planning, capaciy planning
Productservice design | New products and services
Banking sectors Risk management
[Manufacturing sector | Estimating demand
12.7 QUALITATIVE TECHNIQUES OF FORECASTING
Qualitative forecasting techniques generally employ the opinion of experts to generate
forecasts. A key advantage of these methods is that they can be applied when historical data
are not available. Moreover, even when historical data are available, significant changes in
‘environmental conditions affecting the relevant time series may make the use of past data
inrelevant and questionable in forecasting future values of the time series. Qualitative forecasting,
‘methods offer a way to generate forecasts in such cases.
12.7.1 Important Quali
Some important qualitative forecasting methods are described in the following sections
Delphi technique
The Delphi technique uses a panel of experts to produce a forecast. Each expert is asked to
provide a forecast specific to the need at hand. After the initial forecasts are made, each expert
reads what every other expert wrote and is. of course, influenced by their views. A subsequent
forecast is then mad by cach expert. Each expert then reads again what every other expert
‘wrote and is again influenced by the perceptions of the others
Historical background of Delphitechnique: ‘The Delphimethod belongs tothe subjective
intuitive methods of foresight. Delphi was developed in the 1950s by the RAND Corporation,
Santa Monica, California, in operations research. Fred Woudenberg reported that the name
Delphi’ was coined by Kaplan, an associate professor of philosophy at the UCLA working for
the RAND Corporation in a research effort directed at improving the use of expert predictions,
in policy making, Kaplan referred to the ‘principle of the oracle” as a statement that does not
have the property of being “true” or “false”. Thus, Delphi. for the moder foresight method.
is more than a simple brand name, The name can be traced back to Delphi, a sacred place in
ancient Greece, which people used to visit to get answers to their queries by donating some
gold. These answers, uttered by priests or priestesses, were called oracles. In the modern DelphiForecasting 163
method also, expert's opinion are taken to estimate future data of any item/problem having
no past record, Therefore, the journey of Delphi method is traced back to 600 BC, Tablel2.2
presents the historical evolution of the modern Delphi method.
Table 12.2 Historical Journey of Delphi Method
Year Description
(600 BC (Oracle, Delphi Greece
1980 First study and implemented in Miltary research
1964 ‘Comprehensive study in Delph, USA.
1970 15 Japanese Delphi study, STA
1990-91 ‘5 Japanese Delphi study
1992-03 “1 Comprehensive German Delph eludy, Delphi 83
199405 “apanese-German mini Delph study
1996 8® Japanese Delph study
1997-08 2 Comprehensive German Delph study, Delphi 98
"2000-01 FUTURE: the Delphi stucy process is going on where ler counties are nveWved
Definition of Delphi: ‘The Delphi method is based on structural surveys and makes use of
the intuitive available information of the participants, who are mainly experts. Therefore, it
delivers qualitative as well as quantitative results. Delphi technique is a survey is steered by a
monitor group, comprises several rounds of a group of experts, who are unknown to each other
‘and the aim is to reach at a consensus, After each survey round, a standard feedback about the
statistical group judgement is calculated from median and quartile of single prognoses and
if possible, the arguments and counter-arguments of the extreme answers are fed back. The
‘essentials points of the Delphi technique are as follows’
i, Delphi is an expert survey conducted in two or more rounds.
ii, Starting from the second round, a feedback is given (about the results of previous rounds)
to the experts participating in the survey
iii, The same experts assess the same matters once more, being influenced by the opinions of
the other experts. One expert sees the opinion of another and if he/she is influenced by the
comments of the other, he/she is requested to review his/her comments in the light of other
comment. This is done to bring more accuracy in the judgement.
iv. Contents of Delphi studies are always about those issues on which information does not
exist. Else there are more efficient methods of forecasting
v, Delphi isa judgement-based process with uncertain aspects, where the judgement is based
‘on the expert's experience and subject knowledge.
vi. Experts are involved, who on the basis of their knowledge and experience, are able to
assess the future in a competent way. During the rounds, they have the opportunity to
gather new information.
vii, Delphi method, being a qualitative approach, the emphasis is more on the experience of
the expert than the results of mathematical models,
viii, Delphi tries to make use of self-fulfilling and self-destroying prophecies in the sense of
shaping or even creating the future
When to use Delphi: The Delphi method is mainly used when long-term issues have to be
assessed. As itis a procedure to identify statements (or topics) that are relevant for the future.164 Production and Operations Management
it reduces the complex knowledge to a single statement and makes it possible to judge upon.
‘Therefore, its use in combination with other methodologies like scenarios, technology list or
‘others can be interesting, On the other hand, in more complex issues, when the themes cannot
be reduced that much or when thinking and discussions in alternatives are the major target,
Delphi is not the method of choice. For example, consider the situation in which a company
‘ora ministry has to decide which of two research programmes to support, A or B. Programme
A is proposed from faculty A and industry A and the peers fiom discipline A have given their
reviews, Programme B in conjunction with industry B originates from faculty B and the peers
of discipline B have made up their minds, Everybody did her or his best, but how to decide
between them is somewhat not so straightforward,
Organizing a Delphi process: There ate different possibilities to organize a Delphi
process. However, before starting, you should answer the following questions:
+ What is my objective?
+ How many resources (manpower, money...) do I have?
+ Is Delphi the right choice?
+ How can I formulate the statements?
+ What are my questions?
As mentioned before, the usual way is a combination of methods as the topics have to be
formulated especially, a process that already needs methods like creativity procedures or
‘can even be combined with scenatios or future workshops. In the following, a more standard
procedure is described, It starts with the organization of the process as such. In Figure 12.1,
this is illustrated by the “real” example of the Delphi "98 (Hunter and Storz, 2006) in Germany
Organisation ofthe Delphi-Process
Steering Committe®
wos YN
A 5 c D E F ExportPanels
Yager
Sento
Send to experts
Results, Analysis, Discussion
Figure 12.1 Organization of a Delphi Survey
‘Source: Hunter and Stor 2008)
‘The first step is to find a steering committee (if you need one) and a management team
with sufficient capacities for the process. Then expert panels to prepare and formulate the
statements are helpful unless it is decided to let that be done by the management team. The
‘whole procedure has to be fixed in advance: Do you need panel meetings or do the teams workForecasting 165
virtually’ Is the questionnaire electronic or on paper? This means that logistics (from Internet
programming to typing the results from the paper versions) have to be organized. Will there
be follow-up workshops, interviews, presentations? If yes, these also have to be organized and.
pre-pared. Printing of brochures, leaflets, questionnaire, reports have also be considered. The
last organizational point is the interface with the financing organization if this is different from
the management team
Formulating topics: When the organization is roughly defined, the fields of the Delphi
should be decided on, In some cases, one thematic field is enough, in many cases it is wished
to get an overview so that more fields are decided on and handled in a flexible way. There is
always the possibility to add or disclose or re-name fields. To give some examples, the Delphi
"98 fields were:
+ Information and communication
+ Service and consumption
‘+ Management and production
+ Chemistry and materials
+ Health and life processes
+ Agriculture and nutrition
+ Environment and nature
+ Energy and resources
+ Construction and dwelling
+ Mobility and transport
‘Then, the topics have to be formulated. This is a time-consuming, process. It has to be clear,
‘where the topics stem from, The easiest way is desk research and to take topics from literature
and surveys that are available. However, the more creative way isto find working groups who
have the task to structure the field and formulate topics. One can start from seratch, but itis
very difficult to focus the themes, then. Therefore, the more efficient way is to find an already
‘existing material from research, Then brainstorming, brain writing or other ercative activities
‘can add themes. When there is the critical mass of topics, then you need a filter system. What
are the topics that are relevant for your specific Delphi with your specific objectives? Here
‘you can already make some formal or informal judgments (from discussion to giving points or
school notes, even computer semi-anonymous topic selection is possible). It is recommended.
not fo have more than 50 topics per questionnaire but it also depends on the question onc
intends to ask
Analysis of Delphi results: In most Delphi surveys, a lot of statistical data is gathered
which can be used in differemt ways. But comments are also often asked and can help to
interpret the statistics or be analysed in a qualitative way. Especially the combination of Delphi
and scenarios makes many qualitative presentations possible. A Delphi manager should think
about the way to analyse in advance because this has implications on the criteria and the whole
design of the questionnaire as described above
Nominal group technique
‘The nominal group is similar to the Delphi technique in that it utilizes a group of participants.
usually experts. Afler the participants respond to forecast-related questions, they rank their
responses in order of perceived relative importance, Then the rankings are collected and166 Production and Operations Management
aggregated, Eventually, the group should reach a consensus regarding the priorities of the
ranked issues,
Sales force's opinion
The sales staff is often a good source of information for future demand, They may be asked for
their input and aggregate their responses into a sales force composite forecast. Caution should
be exercised when using this technique as the members of the sales force may not be able to
distinguish between what customers say and what they actually do. Also, ifthe forecasts will
be used to establish sales quotas, the sales force may be tempted to provide lower estimates.
Executive opinions
‘Sometimes senior-level managers meet and develop forecasts based on the knowledge of their
areas of responsibility. This is sometimes referred to as a jury of executive opinion.
Market researchisurveys
In market research, consumer surveys are used to establish demand. Such marketing research.
involves a questionnaire that solicits personal, demographic, economic and marketing
information. Researchers collect such information in person at retail outlets and malls, where
the consumer can experience taste, fel, smell and sce a particular product. The rescarcher must
be careful that the sample of people surveyed is representative of the desired consumer target.
12.7.2 Merits and Demerits of Qualitative Techniques
The merits and demerits of the qualitative techniques of forecasting can be summed up as
follows (Table 12.3)
Table 12.3 Qualitative Forecasting Technique: Merits and Demerits
Warts Demerits
Abilty to predict changes in patterns |Ablty to forecast inaccurately f forecasters only consider recently
perceived information.
Allow decision makerste incorporate | nab to forecast if there is complex information,
Fich data sources consisting oftheir
Intution, experience and expert
Judgement
‘The ability to forecast accurately can be reduced due to forecasters
tendency to infer relationships in the absence ofa pattern
Abily to forecast accurately may be reduced when a forecaster tris to
justify 2 forecast that proves to be inaccurate.
‘Quaitatve forecasting techniques are expensive and time intensive.
Source: Hogar) ana Makdak 1981)
12.8 QUANTITATIVE TECHNIQUES OF FORECASTING
Itis another method of forecasting used to predict demand, costs, profit, prices, etc., by using
analysis of past data, Quantitative forecasting can be applied when the following conditions exist:
+ Information about the past is available.
+ This information can be quantificd in the form of numerical data.
+ Itcan be assumed that some aspects of the past pattern will continue into the future. It is
also known as the assumption of continuityForecasting 167
12.8.1 Important Quantitative Methods
‘Some important quantitative forecasting methods are described in the following sections.
Time series method
‘Time series analy sis predicts future demand from past data, A time series isa sct of time ordered.
‘observations on a variable during successive and equal time periods. For example, demand in
units for successive and equal time period of month, i., demand in January, February, March,
April, May, ete. There are some interactive components used in time series, such as levels,
trends, seasonal variations, cyclical variations and random variations as shown in Figure 12.2
+ Levels indicate the scale or magnitude of a time series.
‘+ Trend identified the rate of growth or decline ofa series over time.
Raw Data
Trend Component
‘Seasonal Component
Cycle Component
Random Component
° 4 2 a 4
Tine (years), ——=
Figure 12.2 Various Components of a Time Series
+ Seasonal variations consist of annually recurring movements above and below the trend line.
+ Demand fluctuates in a repetitive pattern from year to year.
+ Seasonal periodic peaks and valleys should occur at the same time every year
+ Seasonal variations should be of larger magnitude than the random variations,
+ Cyclic variations are long-term oscillations or swings about a trend line. The eycles may
vary with respect to the time of occurrence. the length of the phases and the amplitude of
the Muctuations.