Determination of combined forecasting weights based on multiple effect evaluation criteria optimization

The determination approach of combined forecasting weights based on multiple effect evaluation criteria is proposed. This approach is different from the traditional combined forecasting approach by improving single criterion. In this approach, for each effect evaluation criteria, programming model is constructed, and the weights of combined forecasting are obtained. Then by selection some criteria which the effect of combined forecasting on the criteria is superior to the effects of the combined approach, and minimizing the deviation of weights on combined forecasting with weights of combined forecasting on each effect evaluation criteria, the programming model is constructed. The weights on combined forecasting are gained by using genetic algorithm to solve the programming model. Finally an example is given to show the feasibility and availability of this method.