There are different methods for estimating future demands. These methods differ in costs. Flexibility and the necessary skill and sophistication. Which method is to be regarded as the best in a particular situation ?
According to Savage and Small a good forecasting method should possess the following five characteristics :
1. Plausibility : Management must be able to understand and have confidence in the techniques used. Understanding is essential for a proper interpretation of the result.
2. Simplicity and Case of Comprehension : Elaborate mathematical and econometric procedures may be judged less desirable if management does not really understand what the forecaster is doing and fail to understand the procedure.
3. Economy : The cost of forecasting has to be weighted against the importance of the forecast to the operations of the firm. There is no point in pursuing very high levels of accuracy at great expense if the forecast is of no importance to the firm. . The technique employed should be able to produce.
4 Availability : The technique employed should be able to produce  meaningful results quickly. If the technique takes a long time it may not be of much help to the management.
5. Accuracy: It is necessary to check the accuracy of past forecasts against Present performance and of present forecasts against future performance, Some comparison of the model with what actually happens and of the assumptions with what is borne out in practice are most desirable.
More detail is given in another article i.e criteria-good-demand-forecasting-method/
