Research On Spare Parts Classification And Inventory Control Based On Demand Characteristics

With the continuous improvement of industrial automation production level,the dependence of enterprises on equipment is increasing year by year,which means that reasonable management of equipment-related spare parts will directly affect the operation performance and economic benefits of enterprises.Therefore,most enterprises reserve a certain amount of spare parts to maintain the normal operation of the equipment and ensure the continuous production,so as to avoid the adverse effects such as equipment shutdown caused by the shortage of spare parts.However,excessive storage of spare parts will lead to an increase in the cost of enterprise inventory,occupy enterprise funds,and affect enterprise operation.Therefore,under the premise of meeting equipment demand,reducing spare parts inventory has become an important problem that enterprises need to solve urgently.Based on this,this paper takes the spare parts of enterprises as the research object,and on the basis of analyzing the laws of spare parts demand,studies the problems of spare parts classification,demand distribution and spare parts inventory management,in order to provide scientific classification and reasonable management for spare parts management of continuous production enterprises.The main research contents are as follows:First of all,this paper analyzes and summarizes the theory related to spare parts,and expounds the commonly used classification methods and inventory management modes of spare parts.Then,it studies and analyzes relevant literature on spare parts management.The distribution is not applicable,the general analysis of the existing inventory model is poor,and there are few researches on the joint inventory of spare parts.Therefore,this paper conducts analysis and research from three aspects: spare parts classification,demand distribution and inventory modeling.Secondly,the law of spare parts demand is analyzed.On this basis,this paper proposes a classification method combining AHP-BP neural network,and uses this method to classify spare parts into three types: fast turnover spare parts,ordinary turnover spare parts and slow turnover spare parts,and then verified the AHP-BP neural network method has high accuracy and timeliness in spare parts classification through examples.After that,through literature research,a selection criterion for spare parts demand distribution was proposed,and an improved K-S goodness-of-fit test method was provided to select the demand distribution of each spare part;The variety of spare parts is the research object,and the(t,S)and(S-1,S)inventory strategies are used to establish a single-variety inventory control model with random demand and lead time for fastturning spare parts and slow-turning spare parts;On this basis,considering the correlation of spare parts types,taking multi-variety spare parts as the research object,combining spare parts classification,demand distribution selection results and inventory modeling,using(t,S)and(s,S-1,S))inventory strategy establishes a multi-variety joint replenishment inventory control model based on compound distribution for fast-turnover spare parts and slow-turnover spare parts.Finally,the spare parts classification method,demand distribution model and inventory control model are respectively studied by taking the spare parts of SH Group as a case.Firstly,it is verified that AHP-BP neural network classification method can improve the accuracy and timeliness of spare parts classification.Secondly,the fitting effect of the improved K-S test on the demand distribution of slow moving spare parts is compared with the traditional K-S fitting test,which verifies that the composite Poisson-geometric distribution and composite Poisson-Poisson distribution have good fitting effect on the slow moving spare parts,and the improved K-S method is more suitable for fitting the demand of slow moving spare parts.Finally,the single-variety inventory model and multi-variety combined replenishment inventory model are solved,and the total cost of the two inventory models is compared and analyzed.The results show that(T,S)inventory strategy can reduce the cost of the three kinds of spare parts by 7.5%,18.75% and 1.1%,respectively,compared with the traditional inventory strategy.In the category of slow turnover spare parts,compared with the traditional inventory strategy,the(S-1,S)inventory strategy can help reduce the inventory cost of the three kinds of spare parts by 44%,14% and 26% respectively.Compared with the traditional assumed demand distribution,the composite Poisson distribution can help reduce the inventory cost of the three kinds of spare parts by 37%,41% and 9.8% respectively.In the solution of multi-variety combined inventory model,it is found that for three kinds of fast turnover spare parts,combined replenishment reduces the total cost by 1.3% compared with decentralized replenishment,and for three kinds of slow turnover spare parts,combined replenishment reduces the total cost by 35% compared with decentralized replenishment.This article contains 19 figures,36 tables,and 109 references.