Research Gateway

Abstract

Demand Forecasting is very crucial for any business organization. Many times people involved in business activities have to take decisions on which business prospects depends. The decision-making should be accurate and precise as company’s revenue depends on this. For this forecasting models are developed which aids in making decisions. The objective of this work is to study the basics of Support Vector Machine (SVM) and its application in supply chain management and develop an SVM model, which will predict the future demand with high accuracy as compared to the conventional forecasting methods. To demonstrate the effectiveness of the present study, demand forecasting issue was investigated in a piston manufacturing industry as a real world case study. In this proposed research, a SVM model is developed using radial basis kernel function and sigmoid function. Various factors that affect the product demand such as produced units, inventory, sales cost, and number of competitors have been taken into consideration in the development of model. A comparative analysis of SVM model and various traditional forecasting methods like exponential smoothing, moving average and autoregressive model has been done based on the results obtained from forecasting models.