Policy capturing of the aggregate production planning process by means of transform analysis

Journal Title
Journal ISSN
Volume Title

The Aggregate Production Planning (APP) decision-making process is studied at the industry level by means of empirical research and transform analysis. The result is a model which is used for a better understanding of the dynamic properties of current APP practices in various industries and therefore for an improvement in production planning. As a first step, the critical APP variables at the industry level are identified and linear decision rules are hypothesized using the APP theory at the firm level - mainly the Linear Decision Rule and the Management Coefficient Model - as well as empirical and economic studies. Two decision rules for workforce and three decision rules for production are thus formulated as a function of workforce, forecasted sales, and inventory levels. The empirical research consists of validating these decision rules with the use of policy capturing of national economic data for three types of industries: total manufacturing, primary metals, and fabricated metal products. The validation by means of stepwise multiple regression identifies the "best" decision rules for each industry and leads to general findings concerning their predicting performance. One of these findings demonstrates that managers do not in general use a fixed set of rules even though the basic variables (workforce, production, inventory, and sales) seem to be considered. Furthermore, the critical variables and their relative importance appear to vary from one industry to another and from one period of economic activity to an other. The effect of the length of the planning horizon on the performance of the decision rules is also studied. Using the geometric z-transform and the power of a signal flow graph reduction technique, a model for each industry grouping the validated decision rules is built and transfer functions relating APP outputs and inputs are derived. The stability of each model is considered and the responses to various forcing functions (impulse, step, ramp, and sinusoid) are observed. The advantage of a sensitivity analysis of current APP decision-making in the various industries is demonstrated and a number of conclusions are drawn. At the same time, the economic implications of the model are discussed. Finally, the conclusions, limitations, contributions and applications of this research are presented along with some suggestions for future work.

Decision making, Industrial management