A Hybrid Method for Forecasting Utilizing Genetic Algorithm With An Application to the Average Daily Number of Patients

摘要

High accuracy demand forecasting is essentialin Supply Chain Management. In industries, how to improve forecasting accuracysuch as sales, shipping is an important issue. There are many researches madeon this. In this paper, a hybrid method is introduced and plural methods arecompared. Focusing that the equation of exponential smoothing method(ESM) isequivalent to (1,1) order ARMA model equation, new method of estimation ofsmoothing constant in exponential smoothing method is proposed before by uswhich satisfies minimum variance of forecasting error. Generally, smoothingconstant is selected arbitrarily. But in this paper, we utilize above statedtheoretical solution. Firstly, we make estimation of ARMA model parameter andthen estimate smoothing constants. Thus theoretical solution is derived in asimple way and it may be utilized in various fields. Furthermore, combining thetrend removing method with this method, we aim to improve forecasting accuracy.An approach to this method is executed in the following method. Trend removingby the combination of linear and 2nd order non-linear function and 3rd order non-linear function isexecuted to the data of the average daily number of patients for two cases (Thetotal number of patients in hospital, Outpatients number). The weights forthese functions are set 0.5 for two patterns at first and then varied by 0.01increment for three patterns and optimal weights are searched. Genetic Algorithmis utilized to search the optimal weight for the weighting parameters of linearand non-linear function. For the comparison, monthly trend is removed afterthat. Theoretical solution of smoothing constant of ESM is calculated for bothof the monthly trend removing data and the non monthly trend removing data.Then forecasting is executed on these data. The new method shows that it isuseful for the time series that has various trend characteristics and hasrather strong seasonal trend. The effectiveness of this method should beexamined in various cases.