Hierarchical Dynamic Decision Making

The success of manufacturing depends on timely and effective information. Planning in a manufacturing environment is influenced by the integrity of the available information. To generate such information often requires the use of forecasting. Forecasting, itself is a function of the efficacy of input data, which decision makers usually don’t have. Consequently, decision makers resort to guesses based on experience and intuition. The use of such unempirical data can be enhanced by a simulation approach. The problem addressed in this chapter involves the development of a hybrid analytic dynamic forecasting methodology that combines the techniques of analytic hierarchy process (AHP), factor analysis (FA), spanning tree (ST) technique, computer simulation, and the decision maker’s experiential intuition. The methodology presents a simulated scenario output to the decision maker, from which a more integrative forecast decision can be made. Military decision-making environment provides a suitable application platform for the DDM methodology. Forecast results can be used to make operations planning decisions in manufacturing. This chapter presents a probability extension and simulation implementation of AHP incorporating a ST approach and FA based on the previous work of Badiru et al. (1993) on the development of a simulation-based forecasting methodology for hierarchical dynamic decision making (DDM). The problem is to predict the level of change in a base forecast (generated by conventional techniques) due to some probabilistic qualitative factors that are not represented in the calculation of the base forecast. The ST creates a scenario set which requires fewer pair-wise comparison matrices from the decision maker. The approach generates a distribution of forecast outcomes rather than a single forecast value. Forecast scenarios are generated by rudimentary probability information specified by the decision maker. The DDM approach facilitates more interactive, data-efficient, and faster decision making under uncertainty.