Optimization of Forecasting Moving Average Error in Probabilistic Demand Using Genetic Algorithm Based Fuzzy Logic

Demand forecasting is one of the most critical factors in production planning. The uncertain demand, which is the basic idea of planning the production level, nowadays is one of serious problems. The inaccurate demand forecasting could affect to excess production or shortage stocks which led to lost sales. Usually, the forecasted result is hard to represent real condition. Some studies already conducted related to fuzzy time series, each of them has its own advantages and disadvantages compared to other approaches. This research presents the combination of simple moving average forecasting and fuzzy logic model to demand forecast. Then, genetic algorithm (GA) is applied to optimize the mean square error (MSE) inside the fuzzy system. The MSE before and after GA optimization is 0,2192 and 0,1821, respectively.