*Intended learning outcomes: Explain the trend extrapolation forecast and the Delphi method. Describe scenario forecasts.*

During the various phases of the product life cycle, different forecasting techniques are used.

Life-cycle analysisis based on applying to a new product (in a quantitative manner) past demand patterns covering introduction, growth, maturity, saturation, and decline of similar products ([APIC16]).

For the phases of introduction and decline, in particular, future-oriented forecasting techniques are used, both quantitative and qualitative. A technique representative of each class will be presented in the following.

## 10.4.1 Trend Extrapolation Forecast

*A trend extrapolation forecast* attempts to estimate a variable in the future based on the same variable as known at a specific point in time.

In materials management, it may happen that the demand known at a particular point in time t encompasses only a portion of the demand needed for the coming period. Figure 10.4.1.1 provides an example.

**Fig. 10.4.1.1** Demand B_{0} for period t known at time 0.

Extrapolation calculates the total anticipated demand from the demand already known for a product or product family. It compares the *base demand* B_{t}(t+k), 1 ≤ k ≤ ∞, known at time t, to the demand N_{t+k} observed after the closing of a delivery period t+k. This is shown in Figure 10.4.1.2. The variables for the calculation are chosen either as defined in Figure 10.1.3.4 or in a similar fashion. k stands for the *forecast distance*.

**Fig. 10.4.1.2** Actual demand N_{t+k}, divided by base demand B_{t}(t + k).

This quotient, λ_{t}(k) = N_{t+k} / B_{t}(t+k), 1 ≤ k ≤ t is called the extrapolation constant. The dilemma of this definition? Not until the end of period t+k can we determine the actual value of the extrapolation constant, namely λ_{t+k}(0) = N_{t} / B_{t-k}(t), that we had to estimate k periods ago, that is, as λ_{t}(k). Hence, the idea is to smooth the quotients over several periods using exponential smoothing. From now on, let λ_{t}(k), 1 ≤ k ≤ t, be the mean after period t for forecast distance k. The previous mean is used to calculate the new mean using exponential smoothing with smoothing constant α according to the formula in Figure 10.4.1.3.

**Fig. 10.4.1.3** Smoothing of quotient means for extrapolation.

The extrapolation constant is defined for every forecast distance and can be used to extrapolate total demand, at the moment not completely known, from the base demand. Figure 10.4.1.4 gives the forecast value P_{t}(t+k) for the forecast distance k at the end of period t.

**Fig. 10.4.1.4** Extrapolated forecast values for forecast distance k.

The technique described here assumes that the customers’ basic order behavior does not change on the time axis or that it does so very slowly. This means that from a change in customer orders on hand, we can infer a proportional change in total demand. Since this assumption is often invalid in the average case, the technique will yield useful results only when used in combination with other forecasting techniques, such as intuitive ones.

The planner can use this same technique to forecast seasonal components. In the grocery industry, for example, the retailer must give orders to the producers early enough to ensure that shipments arrive on time. Assuming that the retailers’ order behavior does not change significantly from year to year, the producer can derive standardized quotients from sales over multiple years; the probable total demand for the season in a future year can be extrapolated from the demand already known at a specific point in time.

## 10.4.2 Intuitive Forecasting Techniques

In theDelphi method forecast(the name refers to the oracle at Delphi in antiquity), “expert opinion” is gathered through several structured anonymous rounds of written interviews.

The method generally proceeds in various iterations. Figure 10.4.2.1 shows the desired progression during the successive rounds of questioning.

**Fig. 10.4.2.1** Delphi forecasting method: increasing consensus.

The mean of the answers shifts in a specific direction. At the same time, when the dispersion of the answers narrows, there is an increase in the consensus about the direction taken. To arrive at this result, a single iteration should include the following steps:

- The questionnaire is meaningfully constructed or altered. The questionnaires are distributed and completed once again.
- The answers are statistically evaluated by determining mean and dispersion. The results of the evaluation are sent to the experts.
- All the experts are asked to defend their views against extreme arguments. Those who change their opinion as a result of this procedure must provide justifications. The “extreme” respondents must either support their theses with arguments or abandon them.

As for all surveys, the problem for the Deplphi method also lies in formulating the right questions, quantifying the answers, and identifying extreme responses.

The experts are chosen from various areas of an organization, including the sales and marketing units. They are selected for their competence in the field and their broad vision, not for their hierarchical position within the company. The composition of the group should remain anonymous so that the experts cannot identify and be influenced by the responses of other individuals.

## 10.4.3 Scenario Forecasts

Scenario forecasts(orScenario-based forecasts) are plans for how an organization will respond to anticipated future situations ([APIC16]).

Scenario planningis a planning process that identifies critical events before they occur and uses this knowledge to determine effective alternatives ([APIC16]).

Ascenario driveris a key factor or key parameter in determining how the future environment that the organization works in will look.

Scenario
planning and the scenario forecasts that the planning leads to can be used as
tools for use when dealing with situations where the *long-term*
mechanisms of action are either unknown or not fully understood. This applies
in particular where various influencing factors could play a role in a
company's surrounding systems — especially ones that might not even be
considered at first glance. In this respect, scenario planning consciously
assumes that the the past is not necessarily an accurate predictor for the
future. Scenario planning forms part of analyzing the macro environment, which
means it is part of the first stage of the strategic process of designing the
supply chain in Figure 2.1.0.1.

Several alternative scenarios explore the way that social, technical, (macro) economic, environmental or political trends may develop over time, and identify their drivers. There may be several drivers in each scenario, and each driver can also influence other drivers. For scientific discussion about scenario planning, see for example [Scho93]. Figure 10.4.3.1 shows the principle of scenario planning along a time axis.

**Fig.** **10.4.3.1** Scenario planning along a time axis

S_{i}
represents the scenario i, i ≥ 1. The scenario drivers originate from a
specific context in one or more of the company's surrounding systems.
Individual drivers may also feature in several different scenarios. Scenarios
may overlap for this reason, and also for other reasons. A funnel-shaped
representation for each scenario shows its postulated development along the
time axis. S_{i}’, S_{i}",
… represent a revision of scenario S_{i} at time point t’, t", ….
This revision is based on a reassessment of the effective development of the
scenario between the previous assessment and the assessment at the time of the
revision. New scenarios can be added at any time (e.g. S_{4} at time
point t’, S_{5} at time point t"). At any stage, it may be decided
that scenarios are no longer applicable (e.g. S_{1} at time point
t", indicated by Ω).

One well-known example of a company that uses scenario planning is Royal Dutch/Shell. See also for example [Corn05]. In the late 1960s they developed various scenarios that could potentially affect their two most important planning variables, namely the demand for energy and the price of crude oil. These two initial and linearly independent planning variables, which also influence each other, largely determine the other planning variables used by Shell. In a world that had until then been characterized by solid economic growth, one scenario considered a disruptive increase in the price of oil. In 1974, this became a reality when Arab oil-producing countries imposed an oil embargo following the Yom Kippur war.

The core components and processes used in scenario planning have a lot in common with systems engineering (see chapter 19.1). Figure 10.4.3.2 shows a possible approach.

**Fig.**** ****10.4.3.2** Procedure for scenario planning

Only after the completion or (possibly rolling) revision of the scenario planning should possible scenario forecasts for the microeconomic variables of interest be developed within the framework of the corporate strategy, e.g. long-term planning of demand for products or the long-term progression of procurement costs for raw materials. For scenario forecasts, intuitive procedures such as those described in 10.4.1 and 10.4.2 can be used.

The described procedure is complex. Scenario forecasting is therefore more suitable where there are fewer planning variables, and where there are also greater financial consequences if the forecasts are not accurate. This was the case, for example, with their use by Shell as described above. Developing the scenarios, on the other hand, provides extensive knowledge, which can be held “in stock”. That in turn offers time savings if sudden and radically changed peripheral systems impact on the planning variables.

## Course sections and their intended learning outcomes

##### Course 10 – Demand Planning and Demand Forecasting

Intended learning outcomes: Produce an overview of forecasting techniques. Explain history-oriented techniques for constant demand in detail. Identify history-oriented techniques with trend-shaped behavior. Describe three future-oriented techniques. Disclose how to use forecasts in planning.

##### 10.1 Overview of Demand Planning and Forecasting Techniques

Intended learning outcomes: Produce an overview on the problem of demand planning. Present the subdivision of forecasting techniques. Disclose principles of forecasting techniques with extrapolation of time series and the definition of variables.

##### 10.2 Historically Oriented Techniques for Constant Demand

Intended learning outcomes: Describe the moving average forecast. Explain the first-order exponential smoothing forecast. Differentiate between the moving average forecast and the first-order exponential smoothing forecast.

##### 10.3 Historically Oriented Techniques with Trend-Shaped Behavior

Intended learning outcomes: Explain the regression analysis forecast and the second-order exponential smoothing forecast. Describe the Trigg and Leach adaptive smoothing technique. Produce an overview on seasonality.

##### 10.4 Future-Oriented Techniques

Intended learning outcomes: Explain the trend extrapolation forecast and the Delphi method. Describe scenario forecasts.

##### 10.5 Using Forecasts in Planning

Intended learning outcomes: Produce an overview on the choice of suitable forecasting technique. Describe consumption distributions and their limits, continuous and discontinuous demand. Explain demand forecasting of variants of a product family. Present safety demand calculation for various planning periods. Disclose the translation of forecast into quasi-deterministic demand.

##### 10.6 Summary

.

##### 10.7 Keywords

.

##### 10.8 Scenarios and Exercises

Intended learning outcomes: Choose an appropriate forecasting technique. Calculate an example for the moving average forecasting technique and for the first-order exponential smoothing technique. Differentiate between the moving average forecast and the first-order exponential smoothing forecast.

##### 10.9 References

.

##### Case [Course 10]

.