Study On Stocking Decision System Of Office Commodities

Stocking decision is an important business of enterprise operation and management, which focuses on making the best stocking plan of determining the reasonable stocking quantity of merchandises to produce the maximum profits for the enterprises.Traditional patterns of stocking decision generally rely on personal estimation, which disjoints the stocking, the sale and profits sometimes, and cannot effectively deal with and exploit the history data of stock, sale and storage, lack the capabilities of making comprehensive analysis and decision and presenting scientific prediction about the future trends.Consequently, this thesis considers a domestic sale enterprise of office commodities, office Mate corporation, and studies its decision support system (DSS) of stocking, i.e. the design and implement of the stocking DSS on the basis of building the mathematical model of stocking decision, which consists of two steps:First, the sale prediction on short-period sale of office Mate corporation is made using artificial neural networks. The key of sale prediction is to improve the prediction accuracy, where the main difficulties lie in two aspects, the selection of algorithms and elimination of limitations of the algorithms and techniques used. On the basis of the analysis and study of prediction techniques, neural network principles and BP algorithm, a momentum factor is introduced and a batch treatment method of sample set is adopted to improve the generalization and accelerate the convergence rate of BP network. Its application shows that the BP network with momentum factor has faster convergent rate and can achieve much more accurate results in prediction.Second, based on the sale prediction, the mathematical model of stocking decision of office commodities and algorithm is presented by the use of simplex method of linear programming, and the visual design of stocking DSS is given in Dephi7. The application of the system for a period, verified its effectiveness and reliability, and created considerable economic profits for the enterprise.