Integral Logistics Management — Operations Management and Supply Chain Management Within and Across Companies

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.


10.1.1 The Problem of Demand Planning

Demand forecasting is the process of estimating the future demand.

A forecast error is the difference between actual demand and demand forecast. It can be stated as an absolute value or as a percentage.

A forecasting technique is a systematic procedure for forecasting demand according to a particular model.

A certain degree of uncertainty and therefore forecast errors characterize every forecast, regardless of whether people or IT-supported techniques do the forecasting. The latter are a complement to human intui­tion and crea­tivity. When planning the demand, we should make appropriate use of both according to the situation.

If there are only a few items and only a limited amount of information that can be stated explicitly, human forecasting tends to be more precise. This is because human intelligence can process fragmentary information as well as knowledge derived by analogy, thus taking many further factors nec­essary for forecasting into account. This can be important, e.g., in rough-cut planning, where we need only forecast few demands for item families.

On the other hand, when there are many items, or when we can use infor­ma­tion on demand that is expressed explicitly, an IT-supported fore­cas­ting technique generally provides more precise forecasting. This is due to the capa­city of computers to process large quantities of data accurately.

  • Tendencies or trends, such as seasonality, can be calculated from consumption statistics. The length of the time frame to be observed makes this a difficult task for human beings.
  • People tend to weigh unusual events too heavily. In this case, an IT-supported forecasting technique is more neutral in its “reactions.”
  • People tend to focus overly on the recent past. If a forecast proves too high for the current period, they tend to forecast a demand that is too low for the next period, even though this is not justified from the medium-term perspective.

Demand planning is always based upon certain funda­mental assumptions and constraints. Parameters are used to keep their selection as general or flexible as possible. If the demand situation changes, demand planning should reexamine the choice of both parameters and technique and change them if necessary. Figure 10.1.1.1 shows a possible procedure for choosing the forecasting technique and its Parameters.

Fig. 10.1.1.1      A possible demand planning procedure.

  • Choose a demand forecasting technique based on existing consumption or on partially known demand figures.
  • Produce a forecast for future demand by applying the technique.
  • When possible, make a visual check of the forecast and, if neces­sary, correct forecast values that vary too widely from intuitive assumptions. This check allows input of human knowledge of the behavior of the market into automated forecasting techniques.
  • Break the demand forecast down into the needed resources — goods and capacity — according to temporal range and level of detail. This allows planners to estimate the consequences of implementing a forecast and to work out better variations if necessary.
  • Adopt the optimal variant of the forecast as the production plan or procurement plan. These plans represent the independent demand; they are subsequently provided either to the next most detailed or shorter-term planning or to execution.
  • At certain intervals in time, perform an analysis to see whether the course of demand or consumption agrees with the forecast. If the deviation analysis reveals too great a difference, repeat the cycle.

10.1.2 Subdivision of Forecasting Techniques

Figure 10.1.2.1 shows one possible subdivision of forecasting techniques.

8121

Fig. 10.1.2.1      Breakdown of forecasting techniques.

  • Historically oriented forecasting techniques predict future demand based on historical data, for example, on consumption statistics. If a forecast can be made only for an item family or a rough-cut item, then the predicted quantity must subsequently be applied to the detailed items with the use of an allocation key. Historically oriented forecasting techniques can be further subdivided into:
    • Mathematical forecasting techniques, predominant among which is the extra­polation of a time series. Future demand is calculated by extrapolating a series of demands in the past. Such procedures are used widely.
    • Graphical forecasting techniques, where a time series is represented graphically; a mean course and width of deviation are judged by “eyeballing” and are projected into the future based on past experience.
  • Future-oriented forecasting techniques take information already at hand about futu­re demand into account, such as bids, firm orders, orders in the concluding phases, or surveys of consumer behavior. Such techniques are further subdivided into:
    • Mathematical forecasting techniques; for example, extrapola­tion. Beginning with confirmed orders, future order volume is calculated empirically.
    • Intuitive forecasting techniques attempt to estimate the future behavior of target customers in an intuitive way. These techniques are particularly useful when new or significantly enhanced products are introduced to the market, or if surrounding systems significantly change. Surveys, juries of executive opinion, or estimations are simple intuitive techniques. Relevant information can be provided by the sales department, the sellers, or market research institutes that use surveys to assess customer behavior, or by customers themselves (direct contact). Rather technical methods are expert systems, neural networks, decision support systems (DSS), or other statistics and operations research techniques . As typically intuitive techniques, the Delphi method and the Scenario-based forecasts will be presented .

combination of these techniques is also thinkable. For example, forecasts produced using a mathematical technique may be “eyeballed” for accuracy using a graphical representation.

Another possible subdivision of forecasting techniques is the following (see [APIC16]):

  • Qualitative forecasting techniques based on intuitive expert opinion and judgment (e.g., manual forecast, Delphi method)
  • Quantitative forecasting techniques using historical demand data to project future demand; these techniques are further subdivided as follows:
    • Intrinsic forecasting techniques are based on internal factors, such as an average of past sales, and are useful for individual product sales.
    • Extrinsic forecasting techniques are based on a correlated leading indicator (a business activity index that indicates future trends), such as estimating sales of disposable diapers based on birth rates or estimating furniture sales based on housing starts ([APIC16]). Extrinsic forecasts tend to be more useful for large aggregations, such as total company sales.

10.1.3 Principles of Forecasting Techniques with Extrapolation of Time Series and the Definition of Variables

Particularly for forecasting based on historical data, statistical techniques are used that are based on a series of observations along the time axis (here see [BoJe15], [IBM73], or [WhMa97]). The following values are fundamental to the determination of stochastic requirements:

A time series is the result of measurement of particular quantifiable variables at set observation intervals equal in length.

The statistical period or observation interval is a time unit, namely, the period of time between two measurements of the time series (e.g., 1 week, 1 month, 1 quarter).

The forecast interval is the time unit for which a forecast is prepared ([APIC16]). This time unit best corresponds to the statistical period.

The forecast horizon is the period of time into the future for which a forecast is prepared ([APIC16]). It is generally a whole number multiple of the statistical period.

As an example, Figure 10.1.3.1 shows the frequency distribution[note 1001] of the observed variable “customer order receipts” during the most recent statistical period as a histogram. [note 1002]

Fig. 10.1.3.1      Example of a time series.

A demand model attempts to represent demand by drawing the curve that shows the least scattering of the measured values.

Curve fitting is the process performed to obtain that curve, by means of a straight line, polynomial, or another curve.

We assume that the scattering (dispersion) of values is random and, most often, distributed normally. This presupposes that while demand values do indeed have a fluctuating pattern, it is possible to make fairly good appro­xi­­mations. Figure 10.1.3.2 presents some common cases of demand models.

Fig. 10.1.3.2      Possible and common demand models.

Matching a particular demand model to a particular time series leads to the choice of a forecasting technique. The forecasting technique is thus based on a concept or a model of the course of demand. This concept forms the basis for the perception of regularity or a regular demandand the model is

  • Either an econometric model, mostly defined by a set of equations, formulating the interrelation of collected data and variables of the model of the course of the demand as a mathematical regularity,
  • Or an intuitive model as an expression of the perception of an intuitive regularity.

It is quite possible that for a single time series several models will overlap.

(Statistical) decomposition or time series analysis is a breakdown of time series data into various components by analysis; for example, into:
- (Long-term) trend component
- Seasonal component
- Nonseasonal, but (medium-term) cyclical component
- Marketing component (advertising, price changes, etc.)
- Random component (nonquantifiable phenomena), e.g., due to noise, that is random variation or a random difference between the observed data and the “real” event. 

Mathematical statistics offers various methods for determining the mean, deviation, expected value, and dispersion (scattering)[note 1003] of measured values for a time series. Its ability to reproduce the demand for a demand model accurately depends upon the situation. Figure 10.1.3.3 shows a morphology of possible statistical features and the statistical methods that they characterize.

Fig. 10.1.3.3      Statistical methods to determine mean and dispersion.

  1. Calculation of dispersion. Two basic methods are used:
    • Extrapolation, or estimation by calculation of deviations of individual values in the previous statistical periods from the mean, postulated by the demand model.
    • Direct, that is, retrospective determination of the forecast error as the difference between actual demand and projected demand according to the demand model.
  2. Measure of dispersion. There are two standards here:
    • Mean square deviation σ: (sigma, i.e., standard deviation)
    • Mean absolute deviation (MAD)
  3. Weighting of values. Most commonly encountered are:
    • Equal weighting of all measured value
    • Exponential weighting of measured values in the direction of the past

In most cases, we only measure satisfied demand for all models. This equates consumption with demand. The basic problem with this measure­ment is that real demand is not taken into account. The customer order receipts mentioned in Figure 10.1.3.1 may have been higher, for example, if a better demand model had resulted in better availability. Strictly speaking, the customer orders that could not be filled should have been measured as well. The problem with this, however, is that the customer orders may be filled at a later time period. At that time, there may be other orders that will then be unfilled, etc. Determining the exact amount of demand in the past by employing a “what would have happened if” method rapidly proves itself redundant; later demand on the time axis is most likely dependent on satisfied demand in the preceding periods on the time axis.

The following sections use the variables defined in Figure 10.1.3.4. The nomenclature was chosen in such a way that the index always shows the point at the end of the statistical period in which a value is calculated. The period to which the value refers is shown in parentheses.

Fig. 10.1.3.4      Definitions of variables, each calculated at the end of a statistical period.



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]

    .


Print Top Down Previous Next