Fly Glue Trap Demand Forecasting

Abstract

The aim of this research is to compare various forecasting techniques that provide the best choice for the collected data by minimizing mean square error. In order to forecast the demand of fly glue trap, we collect 3 types of such products which are yellow, blue and green ones. The methods which are applied to compare are simple moving average, weighted moving average, exponential smoothing, stationary with additive and multiplicative seasonal effect, double moving average, double exponential smoothing (Holt’s method) and Holt-winter’s method for additive and multiplicative seasonal effect. The results find that double moving average, double exponential smoothing (Holt’s method) is the best forecasting method that can minimize mean square error which are 551,161.21   for yellow type of fly glue trap while double exponential smoothing (Holt’s method) is the best choice that minimizes mean square error 14,447.7 and 77,545,737.4 for blue and green type of fly glue trap, respectively. Therefore, an entrepreneur can apply such recommended methods to forecast the future demand of the product.