Comparative analysis of forecasting methods applicable to Electronics product

Abstract

Forecasting is a method of estimating the future demand or statistics or metrics of a particular need based on past data or historical data. For any business organization, forecasting makes informed estimates or predictions of specific business metrics such as sales growth, or customer demand, or economy-wide forecasts for the coming year. This helps the business organization make sound business decisions, effective planning, and optimum utilization of resources. In this globalization era, electronics industry needs to trade-off among cost, quality, and availability of the product. Otherwise, competitors would take the position in the market. In order to remain in the driving market of the electronics goods production, every business organization should pay attention to demand forecasting methods to reduce the variation between actual and forecasted demand. Today cell phone is the most demandable and widely used electronic product in the world. In this regard, various forecasting methods related to cell phones had been studied in this research work. The existing demand forecasting techniques, namely qualitative forecasting methods were practiced by the retail stores. Thereafter various time series smoothing forecasting strategies had been utilized and on the basis of lowest Mean Absolute Deviation (MAD) value, the top forecasting method was selected. After analyzing the data, it has been identified that for customary cell-phone models, almost 50% are fitted to double exponential smoothing, 33% are fitted to single exponential smoothing, and 17% are fitted to naive forecasting. Similarly, it has been revealed that for conventional cell-phone models, nearly 33% are appropriate to double exponential smoothing as well as to naive forecasting, 17% are appropriate to single exponential smoothing, and the remaining are suited for three months moving average method. Regarding cumulative forecasting, it has been obtained that double exponential smoothing method is best fitted for both customary and conventional cell-phones. In the end, few recommendations for the current companies have been suggested in order to develop the forecasting strategies of the organizations.