Forecasting Models Used in Turkish Paint Industry

Only available on StudyMode
  • Download(s) : 177
  • Published : April 15, 2012
Open Document
Text Preview
International Journal of Computational Intelligence Systems, Vol.2, No. 3 (October, 2009), 277-287

A Hybrid Model for Forecasting Sales in Turkish Paint Industry Alp Ustundag*

Department of Industrial Engineering, Istanbul Technical University, Macka, Istanbul, 34367, Turkey

Received: 27/03/09 Accepted: 17/09/09
Abstract Sales forecasting is important for facilitating effective and efficient allocation of scarce resources. However, how to best model and forecast sales has been a long-standing issue. There is no best forecasting method that is applicable in all circumstances. Therefore, confidence in the accuracy of sales forecasts is achieved by corroborating the results using two or more methods. This paper proposes a hybrid forecasting model that uses an artificial intelligence method (AI) with multiple linear regression (MLR) to predict product sales for the largest Turkish paint producer. In the hybrid model, three different AI methods, fuzzy rule-based system (FRBS), artificial neural network (ANN) and adaptive neuro fuzzy network (ANFIS), are used and compared to each other. The results indicate that FRBS yields better forecasting accuracy in terms of root mean squared error (RMSE) and mean absolute percentage error (MAPE). Keywords: Sales forecast, Hybrid model, Fuzzy rule based system, Multiple linear regression and Paint industry.

1. Introduction The Turkish paint industry is the sixth-largest in Europe, with a capacity of 800000 metric tons annually. Paint is manufactured in the five Turkish cities with the highest population. The surge of new immigrants into these cities has increased demand for new housing and will be key to sustaining future growth in the paint market with demand estimated at about 600000 metric tons annually. Since the industry is dependent upon imported raw materials, companies should ensure that they are properly positioned to meet future Turkish demand. Therefore, it is important for paint companies to have an efficient and accurate forecasting model to predict future monthly sales. A major requirement of successful marketing is accurately forecasting sales. First, market opportunities are identified through marketing research. The size, growth and profitability of each market opportunity are then measured and/or forecasted. Sales forecasts are used by *



finance divisions to raise cash needed for investments and operations,  manufacturing divisions to establish capacity and output levels,  purchasing divisions to acquire the necessary supplies,  human resources divisions to hire the necessary number of workers. Thus, an accurate sales forecast facilitates effective planning. Over-estimates of demand can lead to several problems, such as occupancy of valuable shelf space and increased inventory carrying charges. On the other hand, under-estimates of demand can lead to stock depletion, lost sales, and expensive overtime production to compensate for costumer demand. Given the potentially negative impact of inaccurate forecasts, marketers use a variety of techniques to accurately forecast sales. This research focuses on monthly sales forecasting for the largest paint manufacturer in Turkey. A hybrid model that integrates AI and MLR methods is proposed

Istanbul Technical University, Department of Industrial Engineering, Macka 34367 Istanbul, Turkey, ustundaga@itu.edu.tr

Published by Atlantis Press Copyright: the authors 277

A. Ustundag

for predicting this company’s sales. The model is then applied to historical sales data for this specific Turkish paint producer. Fuzzy rule based system (FRBS), artificial neural network (ANN) and adaptive neuro fuzzy network (ANFIS) methods are used in the proposed hybrid model and their relative performances are compared to each other. The purpose of the study is to improve forecasting accuracy and thus to help managers improve their decision making. The remainder of the paper is organized as follows. Section 2 summarizes...
tracking img