Long term Demand forecasting

We use judgmental methods like Delphi, Executive Opinions and Expert Opinions. We may also use extension of past history methods like moving averages, exponential smoothing, or we can causal forecasting models like Regression analysts and econometric models.

Regression analysis is a forecasting model that relates the dependent variable (sales, for example) to one or more independent variables (GNP disposable income, housing construction, road building, hospital services etc) For example, the demand refrigerators is a function of such independent variables as the new houses constructed in the previous years, new marriages solemnized during the previous year, disposable income available, and a seasonal and trend effect. We are here essentially developing a forecasting function called regression equation. It is a time consuming process and also very costly process to assess and identify the various independent variables for developing a regression equation. For long term forecasting, a forecast for each causal variable is needed.

Econometric forecasting is a further improvement on regression analysis. It consists of a system of simultaneous regression equations. For example, we can say demand for soaps is a function of its price and advertising and the personal disposable income of the consumer. However, price itself is a function of the total cost of production, selling and distribution and the mark up desired.

Production cost itself is a function of the manufacturing level and materials level. Similarly selling costs are a function of salesmen’s salaries and commissions, cost of servicing the territories etc. We thus have to estimate several relationships simultaneously. These methods are expensive and time consuming.

Judgmental methods are used when good data are not readily available. With the use of this method, we try to- convert the qualitative data in the form of a subjective opinion into quantitative data that we can use for forecasting. Outside experts may be consulted, we can convene a panel of experts to make or we can refer the case to futurist organizations. In each case, reliance is being placed on human judgment to interpret past data and make projections about the future.

Marketing research on consumer behavior, product, price distribution and ad research proves extremely useful in forecasting the long term demand of product.

Historical analogy compares the life cycle of a products to that of another similar product to make predictions, for example, demand for housemaids is related to demand for washing machines or demand for B&W TV sets is compared with demand for color TVs.

While dealing with multiple outputs, the demand for each output is estimated independently and then summed up. If the outputs are heterogeneous, it may mean separate capacity planning for each sub-process. For multiple outputs, capacity gap or slack is computed for each sub-process.

In case of new outputs, the uncertainty of demand is higher. Hence the probability of demand distribution in the term of an optimistic forecast and pessimistic forecast are arrived at, over and above expected projected demand.

“What? Gaming in the workplace? No way!” This is something that we hear from Corporate
Closely tied to the question of how much capacity should be provided to meet forecasted
The notion of focus naturally, almost inevitably from the concept of fit. Just as a
At its heart a capacity strategy suggests how the amount and timing of capacity changes
However, as with most strategic decisions, the issue is more complex than it first appears.