An improved sales forecasting approach by the integration of genetic fuzzy systems and data clustering: Case study of printed circuit board

https://doi.org/10.1016/j.eswa.2011.01.132Get rights and content

Abstract

Success in forecasting and analyzing sales for given goods or services can mean the difference between profit and loss for an accounting period and, ultimately, the success or failure of the business itself. Therefore, reliable prediction of sales becomes a very important task. This article presents a novel sales forecasting approach by the integration of genetic fuzzy systems (GFS) and data clustering to construct a sales forecasting expert system. At first, all records of data are categorized into k clusters by using the K-means model. Then, all clusters will be fed into independent GFS models with the ability of rule base extraction and data base tuning. In order to evaluate our K-means genetic fuzzy system (KGFS) we apply it on a printed circuit board (PCB) sales forecasting problem which has been used as the case in different studies. We compare the performance of an extracted expert system with previous sales forecasting methods using mean absolute percentage error (MAPE) and root mean square error (RMSE). Experimental results show that the proposed approach outperforms the other previous approaches.

Research highlights

► We present a novel sales forecasting approach by integration of Genetic Fuzzy Systems (GFS) and Data Clustering. ► We use the K-means model to construct the K-means genetic fuzzy system (KGFS). ► We evaluate the performance of the developed expert system by Printed Circuit Board (PCB) sales forecasting problem. ► We compare the performance of presented KGFS against the previous sales forecasting methods and show the results.

Section snippets

Introduction and literature review

Sales is the most direct measure of the outcome of marketing efforts and so sales forecasting models play a significant role in marketing planning. In the very competitive and dynamic environment that most businesses face, forecasting is a beneficial tool and an indispensable strategy for business survival. Reliable prediction of sales can improve the outcome of business, because it allows functional areas, such as production, sales and marketing, and finance, to effectively develop programs to

K-means genetic fuzzy system

As mentioned before, this article presents a hybrid artificial intelligence method called K-means genetic fuzzy system (KGFS) to construct an expert system (ES) for sales forecasting problems. We apply a combination of K-means data clustering, GA and fuzzy logic approach to construct the ES. At first we apply K-means clustering technique to cluster our raw data. Then data in different clusters divided by K-means will be fed into independent genetic fuzzy systems.

The framework of KGFS is shown

Experimental results

In this section we develop the proposed KGFS using monthly PCB sales data from 1999/1 to 2003/12. Actually, we use the first 4 years (48 data) for training the model and the last year (12 data) for testing the model. We use the same variables used in mentioned PCB sales forecasting studies. The influential variables are as follows:X1:Preprocessed historical data that addresses the effects oftrend, seasonality and random noise;X2:Consumer price index;X3:Liquid crystal element demand;X4:Total

Conclusion

This paper presents a novel approach based on genetic fuzzy systems and K-means clustering (KGFS) in building a sales forecasting expert system. The KGFS approach has the following novel features:

  • •

    It reduces effects of noisy data by means of clustering data set into k different clusters.

  • •

    GAs have been demonstrated to be a powerful tool for automating the definition of the fuzzy rule based systems. KGFS uses a genetic algorithm for extracting rule base of the fuzzy expert system.

  • •

    For the purpose of

References (33)

  • R. Kuo et al.

    Fuzzy neural networks with application to sales forecasting

    Fuzzy Sets and Systems

    (1999)
  • C. Kwong et al.

    A methodology of generating customer satisfaction models for new product development using a neuro-fuzzy approach

    Expert Systems with Applications

    (2009)
  • S. Li

    The development of a hybrid intelligent system for developing marketing strategy

    Decision Support Systems

    (2000)
  • S. Thomassey et al.

    A neural clustering and classification system for sales forecasting of new apparel items

    Applied Soft Computing

    (2007)
  • S. Thomassey et al.

    A global forecasting support system adapted to textile distribution

    International Journal of Production Economics

    (2005)
  • M. Aiken et al.

    Forecasting market trends with neural networks

    Information Systems Management

    (1999)

    • Cited by (0)

    View full text
    Copyright © 2011 Elsevier Ltd. All rights reserved.