Sales forecasting is one of the most common phenomena observed in industry, as it assists other subsidiary department of the industry such as finance, human resources, marketing, supply chain etc. Although forecasted values are obtained through several qualitative and quantitative methods, each method has its own pros and cons. The selection of these models depends upon the knowledge, availability of data and context of forecasting. The purpose of this research paper is to forecast the packaged food product sales using mathematical programming. The scope of the subject is wide and the techniques chosen reflect particular interests and concerns. Methodology: In this study Linear Programming is used to estimate the parameters of time series forecast by minimising one error index (MAD, MAPE, MPE) and they are compared with the time series forecasting method. Findings: Linear Programming is used to estimate the parameters of times series forecast with optimisation objectives to minimise forecasting error and it is compared them with the traditional time series forecasting models. The linear programming approach improves the accuracy of forecast and outperforms all the other techniques. Practical Implications: The use of a mathematical programming provides a formal, logical way of thinking about this decision process. This should increase the understanding of this problem area and increase the quality of decisions.
Sales Forecasting, Time Series Analysis, Forecast accuracy, Mathematical Programming.
Saurabh Gupta; Nishant Kumar, Modelling and Forecasting Packaged Food Product Sales using Mathematical Programming, HCTL Open International Journal of Technology Innovations and Research (IJTIR), Volume 14, April 2015, eISSN: 2321-1814, ISBN (Print): 978-1-62951-946-3.