A two-stage stochastic programming approach for production planning system with seasonal demand

Abstract

Seasonality is a function of a time series in which the data experiences regular and predictable changes that repeat each calendar year. Two-stage stochastic programming model for real industrial systems at the case of a seasonal demand is presented. Sampling average approximation (SAA) method was applied to solve a stochastic model which gave a productive structure for distinguishing and statistically testing a different production plan. Lingo tool is developed to obtain the optimal solution for the proposed model which is validated by Math works Matlab. The actual data of the industrial system; from the General Manufacturing Company, was applied to examine the proposed model. Seasonal future demand is then estimated using the multiplicative seasonal method, the effect of seasonality was presented and discussed. One might say that the proposed model is viewed as a moderately accurate tool for industrial systems in case of seasonal demand. The current research may be considered a significant tool in case of seasonal demand. To illustrate the applicability of the proposed model a numerical example is solved using the proposed technique. ANOVA analysis is applied using MINITAB 17 statistical software to validate the obtained results.