Research And Application Of Inventory Control Policy Under Fuzzy Environment

Inventory management models are used widely in practical world and they can help the decision-makers to obtain the optimal production or ordering lot sizing effectively. For the traditional inventory management models, researchers always assume the key parameters involved in inventory management system such as costs and demands are all constant values, then make decision to obtain the production or ordering lot sizing of single-item or multi-items. But in fact, many factors are nondeterministic. For example, the demand fluctuates as the market changing and cost is affected by the season change, which form the uncertainty environment of inventory management system. These uncertain factors should be considered in inventory planning control system.Firstly, two single-item inventory management models are considered. One is researched as the demand is continuous fuzzy number without considering Deterioration. A model is established to minimize the cost and solved by the method of integral sorting to obtain the optimal ordering lot sizing, then a numerical example is provided to illustrate the problem. The other is the inventory management model on deteriorating item that the demand is nondeterministic and a model is established to maximize the average net profit. The fuzzy parameters fluctuation range is estimated by the method of confidence intervals, then the model is defuzzied by the method of signed distance and optimized to receive the production quantity satisfying the constraints. Finally a numerical example is also provided to illustrate the problem.Secondly, the inventory management model of multi-items with fuzzy costs and inventory capacity restriction is considered. An expected value model is established based on fuzzy costs and fuzzy constraints. For seeking the solution of uncertain model, this dissertation discusses the theory of fuzzy programming and designs a hybrid intelligent algorithm based on fuzzy simulation, neural network, genetic algorithm, and the genetic operation, such as coding, mutation, selection and crossover etc. Lastly the feasibility and validity of this model and algorithm is illustrated by using simulation data and hybrid intelligent algorithm.Finally, the development of a inventory prototype system is introduced, in which the results of the research are applied. The feasibility and application value of fuzzy inventory model is displayed fully.