Enhancement of Intermittent Demands in Forecasting for Spare Parts Industry

The standard method to forecast intermittent demand is Croston’s Method. This method is available in ERP type solution such as SAP and specialized forecasting software packages (e.g., Forecast Pro), and often applied in practice6. In this paper, two forecasting methods, Croston’s Method and a New Method (Bisection of Croston’s Method), are compared. Two kinds of spare parts were chosen for the analysis. By using a modified Croston’s Method, we forecast the average of last two demands over a fixed lead time. The mean square errors are shown to meet the theoretical and practical requirements of intermittent demand. Based on these measures, the best statistical summary can be obtained. The out-of-sample comparison results indicate superior performance of the New Method. In addition, the results show that the mean square error is a well-behaved accuracy measures for intermittent demand. Methods/Statistical Analysis: Croston’s Method was implemented for forecasting irregular demands. Bisection Numerical method was compared with existing Croston’s Method. The results are based on the Mean Square Error Values (MSE) values. Results/Findings: Bisection Method showed MSE Values of Proposed Method is lesser than the MSE values of Existing Method. Conclusion/Application: 'Bisection Method ' can be a better method to predict the Production of Spare Parts Industries.