{"title":"Consumer Product Demand Forecasting based on Artificial Neural Network and Support Vector Machine","authors":"Karin Kandananond","volume":63,"journal":"International Journal of Economics and Management Engineering","pagesStart":313,"pagesEnd":317,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/7938","abstract":"The nature of consumer products causes the difficulty\nin forecasting the future demands and the accuracy of the forecasts\nsignificantly affects the overall performance of the supply chain\nsystem. In this study, two data mining methods, artificial neural\nnetwork (ANN) and support vector machine (SVM), were utilized to\npredict the demand of consumer products. The training data used was\nthe actual demand of six different products from a consumer product\ncompany in Thailand. The results indicated that SVM had a better\nforecast quality (in term of MAPE) than ANN in every category of\nproducts. Moreover, another important finding was the margin\ndifference of MAPE from these two methods was significantly high\nwhen the data was highly correlated.","references":"[1] K. Bansal, S. Vadhavkar, and A. Gupta, \"Neural Networks Based\nForecasting Techniques for Inventory Control Applications,\" Data Min\nKnowl Disc, vol. 2, no. 1, pp. 97-102, 1998.\n[2] F.E.H. Tay and L. Cao, \"Application of Support Vector Machines in\nFinancial Time Series Forecasting,\" Omega, vol. 21, pp. 309-317, 2001.\n[3] K. Kim, \"Financial Time Series Forecasting using Support Vector\nMachines,\" Neurocomputing, vol. 55, no. 1-2, pp. 307-319, 2003.\n[4] W. Huang, Y. Nakamoria and S.-Y. Wang, \"Forecasting Stock Market\nMovement Direction with Support Vector Machine,\" Comput Oper Res ,\nvol. 32, no.10, pp. 319-326, 2005.\n[5] Z. Hua and B. Zhang, \"A Hybrid Support Vector Machines and Logistic\nRegression Approach for Forecasting Intermittent Demand of Spare\nParts,\" Appl Math Comput, vol. 181, pp. 1035-1048, 2006.\n[6] R.S. Gutierrez, A.O. Solis and S. Mukhopadhyay, \"Lumpy Demand\nForecasting using Neural Networks,\" Int J Product Econ, vol. 111, pp.\n409-425, 2008.\n[7] K. Kandananond, \"\"Forecasting Electricity Demand in Thailand with an\nArtificial Neural Network Approach,\" Energies, vol. 4, pp. 1246-1257,\n2011.\n[8] StatSoft, Inc. (2011). Electronic Statistics Textbook. Tulsa, OK:\nStatSoft. WEB: http:\/\/www.statsoft.com\/textbook\/.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 63, 2012"}