论文信息 - Demand Forecast in a Supermarket Using a Hybrid Intelligent System

Demand Forecast in a Supermarket Using a Hybrid Intelligent System

Demand forecasts play a crucial role in advanced systems for supply chain management. Determining the future demand for a certain product is the basis the respective 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 advantages. In this paper we propose a hybrid forecasting system combining ARIMA models and neural networks. We show improvements in forecasting accuracy and develop a replenishment system based on the respective forecasts for a Chilean supermarket chain.

Richard Weber | Luis Aburto

[1] Jiawei Han,et al. Data Mining: Concepts and Techniques , 2000 .

[2] Amar Gupta,et al. Brief Application Description. Neural Networks Based Forecasting Techniques for Inventory Control Applications , 1998, Data Mining and Knowledge Discovery.

[3] C C Reyes-Aldasoro,et al. A hybrid model based on dynamic programming, neural networks, and surrogate value for inventory optimisation applications , 1999, J. Oper. Res. Soc..

[4] Jatinder N. D. Gupta,et al. Neural networks in business: techniques and applications for the operations researcher , 2000, Comput. Oper. Res..

[5] Christoffer Brax,et al. Recurrent neural networks for time-series prediction. , 2000 .

[6] Chin Kuo,et al. Neural Networks vs. Conventional Methods of Forecasting , 1996 .

[7] William Remus,et al. Neural Network Models for Time Series Forecasts , 1996 .

[8] Peter C. McCluskey,et al. Feedforward and Recurrent Neural Networks and Genetic Programs for Stock and Time Series Forecasting , 1993 .

[9] Oliver Vornberger,et al. Forecasting Sales Using Neural Networks , 1997, Fuzzy Days.