A decision support system for improving forecast using genetic algorithm and tabu search

Abstract

The intrinsic uncertainties associated with demand forecasting become more acute when it is required to provide invaluable dimensions for the decision-making process. The concept of decision support system (DSS) is very broad and it can take many different forms. In general, we can say that a DSS is a computerized system for assisting decision making. Forecasting models has been recognized as one of the tools used in DSS. The need and relevance of forecasting tools has become a much-discussed issue and this has led to the development of various new tools and methods for forecasting in the last two decades. One traditional tool for forecasting time series data is the Winter’s method with three parameters that determine the accuracy of the model. The search for the best parameter value of ?, ? and ? and their combinations using trial and error method is time consuming. Hence, a good optimization technique is required to select the best parameter value to minimize the fitness function. We employ the unique search of Genetic Algorithm (GA) to generate and search for the best value and due to the nature of GA that is based on random search; the near optimum solution could be improved by the introduction of a more systematic search known as Tabu Search (TS). Our study shows that combining both GA and TS search methods generate a more accurate forecast