UNCORRECTED PROOF DTD 5 Improved supply chain management based on hybrid demand forecasts § Luis Aburto a , Richard Weber b, * a Penta Analytics, Santiago, Chile b Department of Industrial Engineering, Faculty of Physical and Mathematical Sciences, University of Chile, Santiago, Chile Received 11 May 2004; received in revised form 27 May 2005; accepted 12 June 2005 Abstract Demand forecasts play a crucial role for supply chain management. The future demand for a certain product is the basis for the respective replenishment systems. Several forecasting techniques have been developed, each one with its particular advantages and disadvantages compared to other approaches. This motivates the development of hybrid systems combining different techniques and their respective strengths. In this paper, we present a hybrid intelligent system combining Auto- regressive Integrated Moving Average (ARIMA) models and neural networks for demand forecasting. We show improvements in forecasting accuracy and propose a replenishment system for a Chilean supermarket, which leads simultaneously to fewer sales failures and lower inventory levels than the previous solution. # 2005 Published by Elsevier B.V. Keywords: Supply chain management; Neural networks; Hybrid intelligent systems; Demand forecasting 1. Supply chain management in the retail industry The retail industry has undergone major changes during the past two decades. One of the main developments has been the introduction of supply chain management systems, which affects all agents of such a chain. Especially for supermarkets it has become necessary to apply advanced technologies in order to stay competitive in a highly dynamic environment [21]. In this paper, we describe the development of a hybrid intelligent system for demand forecast, which helped to improve supply chain management in the Chilean supermarket Economax. This supermarket, as well as any retail company, offers a broad range of products (about 5000 different stock keeping units, SKUs) purchased from a large number of manufacturers and distributors. In order to successfully provide such a variety of items to its customers at competitive prices, the supermarket and its providers have to manage efficiently the respective www.elsevier.com/locate/asoc Applied Soft Computing xxx (2005) xxx–xxx 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 § This research has been performed while the second author was with The Research Center for Advanced Science and Technology, RCAST, The University of Tokyo, Japan. * Corresponding author. Tel.: +56 26784056; fax: +56 26897895. E-mail address: rweber@dii.uchile.cl (R. Weber). 1568-4946/$ – see front matter # 2005 Published by Elsevier B.V. doi:10.1016/j.asoc.2005.06.001 ASOC 180 1–9