Joint Reserve-Location Of Slow Moving Spare Parts In Thermal Power Plant

Among the various types of power generation,the generation of thermal power still occupies an absolute advantage,and there is a long way to go to promote the structural reform of the supply side of the energy industry.For the thermal power industry,the inventory management of spare parts has always been a difficult problem.The inventory control is difficult,inventory redundancy leads to a large backlog of funds and overall utilization rate is very low,especially for slow moving spare parts that have a slow rate of low demand and high importance,so this paper puts forward the idea of joint optimization of entity joint reserve and site selection,and conducts more profound research.Firstly,this paper combs the basic theory of spare parts management,points out the definition and characteristics of slow moving spare parts,and introduces the current reserve model of spare parts and explains the basic principles of METRIC model.According to the previous literatures,the most important problems are the definition of slow-moving spare parts is not clear,the demand distribution is difficult to determine,and the quantitative research on the entity joint reserve.Therefore,this paper carries out research on the determination and reserve-site optimization modeling of slow moving spare parts.Secondly,according to the current status of slow moving spare parts management,there are three major problems that the standards are not uniform,the dimensions are not diversified,and the accuracy is not up to standard,and then the seven characteristics of slow moving spare parts are pointed out.Then AHP and BP neural network are used to determine the slow-moving spare parts,we establish an index system to determine the weighting and the quantitative rules of the indicators,and get the input and output values.It is found that the fitting degree is good and error between the value and the predicted value is very small when bought into BP model,so the spare parts are divided into slow moving,medium-speed moving and fast moving spare parts.Compared with ABC classification method,the results demonstrate that AHP-BP neural network model has a good performance in the determination of slow-moving spare parts.Thirdly,we analyze the applicability of the zero-inflated Poisson distribution,which points out that ZIP has great advantages in describing the demand of slow-moving spare parts.Then the VARI-METRIC model is used to establish the separate reserve inventory model with(T,S)strategy,which includes inventory holding cost and ordering cost.Then based on the disadvantages of the separate reserve inventory model,the entity joint reserve-location model is established,which adopts(S-1,S)strategy and considers joint reserve center and transshipment.At last,we design a particle swarm algorithm to solve the above two models.Finally,we take the actual data of spare parts of SH Energy Group as an example to make models solved,and the sensitivity analysis method is used to compare the total cost of the two modes.The research shows that the accuracy of the original classification method can be improved by using the AHP-BP method when considering the diversity of the characteristics of slow moving spare parts in thermal power plants.When considering the entities joint reserve between multiple thermal power plants under the group,it can effectively reduce inventory holdings cost with a 16.74%decline,meanwhile the case proves that slow moving spare parts are more suitable for entity joint reserve mode,but the joint reserve model is not suitable when the shortage cost,location cost are too high or inventory holding cost is too low.Thermal power plants should improve the ability of spare parts inventory management with standardization,synergy and digitization,and pay more attention to the management of slow moving spare parts,and achieve reduce costs and increase benefits.There are 18 figures,22 tables and 90 references in this dissertation.