Construction and selection of quantitative planning models — a general procedure illustrated with models for stock control

The paper investigates the process of quantitative model construction when for a particular realworld decision problem different quantitative models are available. It develops first some general features for constructing a model hierarchy comprising all quantitative models which should be taken into consideration. The main objective is an investigation into the selection process giving the best quantitative model for the particular decision problem at hand. This selection, however, can be performed in different ways possibly ending up with different models. A myopic selection principle is suggested which leads to a class of strategies. Different strategies within this class are discussed showing the complexity of the selection procedure. The investigations are illustrated using inventory models particularly stressing the role of cost parameters.