摘要
随着市场竞争加剧和客户需求的多样化、定制化,建设基于产品的服务系统成为大型设备制造企业提升竞争力的重要手段。产品服务系统在增强了企业对产品服务能力的同时,也为设备备件的库存管理带来了许多难题。
本文以XXX企业的大型设备产品的备品备件(文中也称为备件)库存优化管理的需求为背景,从一个典型的设备制造企业的实际出发,通过实际调研与总结、分析与抽象、研究与验证,提出了在此类制造企业里,设备售后服务中备件库存管理的解决方案,其中包括,备件需求的预测、库存控制策略的优化、库存管理流程的改进。
首先,目前备件需求预测的研究在历史数据的选取和预测方法上存在诸多不合理,如缺少数据预处理及与忽视数据与设备的特性之间的关系,需要给予解决。在考虑不同备件之间需求相关性进行预处理的基础上,以大型空气压缩机的备件需求为例,利用BP神经网络方法,根据其备件历史需求数量的时间序列数据建立预测模型。最后将预处理后的数据输入到神经网络预测模型之中,并将模型的预测结果与未考虑备件之间需求相关性的预测结果进行比较,可以有效解决神经网络的“欠训练”问题,平均偏差率显著降低。
其次,针对目前的多级库存管理中存在的问题,分析设备制造企业售后服务系统中备件需求特点,建立多级库存控制模型。在考虑售备件供应水平和缺货损失的基础上,对模型中的库存订购策略、安全库存,库存成本等问题进行研究。最后以总收益最高为最优化的目标,求解在这一目标下的库存补货策略参数,为企业解决售后服务系统中的多级库存中的库存控制问题提供定量的解决方案。试验证明文中提出的方法能够有效的控制售后服务系统中备件的库存成本,提高收益水平。
最后,用流程分析和仿真的手段对企业当前的备件库存管理的流程进行了改进。通过对备品备件服务流程的现状分析,对备件库存管理的三个主要方面,即组织机构、业务流程和资源配置,利用流程建模的方法进行了方案设计。从系统中的基本事件出发,建立备件管理系统的仿真模型,完成对备件库存管理中的基本事件建模。随后,使用Arena仿真软件对备件供应进行动态仿真,并通过仿真结果来评定所采用的库存管理流程的正确性及资源配置的有效性。
综上,本文针对服务化的设备制造企业,提出了相应的备品备件库存管理方法,对类似制造企业的售后服务中的备件库存管理有一定的实际参考价值。
With increasingly market competition and various customized demands, the product-service system has been playing more and more important parts in the large-scale equipment suppliers. The product-service system can strengthen the service ability, but it also has brought many problems in the spare parts inventory management.
With the background of the optimization of spare part inventory management for complex machine project, a solution is developed through efforts of investigation and analysis, problem abstracting, theoretical research and application, which can be used in the similar company. This research includes the forecast of spare parts demand, optimal of inventory control strategy and improvement in inventory management system.
Firstly, there are some mistakes in choice, pretreatment and forecasting the demand of spare parts demand in some researches, such as improper data set, using raw datum indiscriminately and ignoring the relationship between the datum and the equipments’characters, which need to be improved. Considering the characters of industrial equipments, which are spare parts in large-scale air compressors here, spare parts historical demand data series were pretreated. Based on this a forecast model of time series demand of spare parts was presented with BP neural networks. In the end, the processed demand time series datum were input into neural networks forecasting model. The forecasting results between raw datum and processed datum, which were using neural networks, were compared .The phenomena of“lack-training”vanished, and the average deviation rate remarkably reduced.
Secondly, a model on multi-echelon inventory system of spare parts in the after-sale service market was proposed based on the analysis to the characteristic of spare parts demand in the product-service system. Considering the spare parts supply quality in the after-sale service, the factor, such as ordering quantity of spare parts,safety storage, inventory cost, were researched on the basis of the model. Finally the inventory control parameters were exploited, which aims at the maximum profit in the spare part inventory. The method can provided quantity solution in the multi-echelon inventory management. Experiments show the method that proposed in this paper is valid.
Finally, modify the inventory management system through analyzing and simulating the inventory management structure. The research focuses on the three main parts of the spare parts inventory, such as organize structure, working process, and collocation of inventory resource, on the basis of analyzing the present product-service system in the enterprise. This paper analyzed the basic process in the spare parts management system, established the optimized simulation model and arithmetic. Then, the spare parts supply chain in the product-service system was carried out in the visualized Arena simulation situation. The result of the simulation can be used in validating the inventory control strategy and provide the advice on the related source collocation.
In conclusion, the system method raised in this paper provides solution in the large-scale equipment suppliers’spare parts inventory management, especially when the supplier is transfer its after-sale service from a traditional system to a product-service system.
引文
[1] Tysseland BE. Spare Parts Optimization Process and Results: OPUS10 cases in the Norwegian Defence [J], International Journal of Physical Distribution & Logistics Management,2009,39(1):8-27.
[2]严婷婷,李宏余.基于AHP的库存产品分类模型研究[J],物流技术,2005(11):63-65.
[3]何立明,赵罡,程邦勤.气体动力学[M],北京:国防工业出版社,2009.
[4]何哲,孙林岩,高洁等.服务型制造在大型制造企业的应用实践[J],科技进步与对策,2009,26(6):106-108.
[5]范如国,韩民春.博弈论[M],湖北:武汉大学出版社,2006.
[6] Tongdan Jin, Haitao Liao. Spare parts inventory control considering stochastic growth of an installed base[J]. Computers & Industrial Engineering, 2009,56(1):452-460.
[7]李冬.麦芽生产企业物流优化设计技术研究[硕士论文].上海:上海交通大学,2009.
[8]宋高歌,黄培清.产品服务系统中的契约结构选择[J].统计与决策,2007(24): 45—47.
[9]王琦,基于排队论和业务流程重组理论的门诊流程优化研究[博士论文].武汉:华中科技大学,2008.
[10] Michael Hammer. Reengineering work: don’t automate but obliterate [J]. Harvard Business Review, 1990 (4):104-112.
[11] Askin RG, Krishnan S. Defining inventory control points in multi-product stochastic pull system [J]. International Journal of Production Economics, 2009,120(2):418-429
[12] Ghobbar AA. Forecasting intermittent demand for aircraft spare parts: A Comparative evaluation of methods [J]. Journal of Aircraft, 2004,41(3):665-673.
[13] Gupta UC, T Rao. On the M/G/1 machine interference model with Spare Parts[J]. European Journal of Operational Research, 1996, 89(1):164-171.
[14]史耀媛,史忠科.基于非单点模糊正则网络的时间序列预测模型[J].西北大学学报,2006,36(6):887-890.
[15] Croston JD. Forecasting and stock control for intermittent demands[J]. Operational Research Quarterly, 1972, 23(3):289-303.
[16] Willemain TR, Smart CN, Schwarz H F. A new approach to forecasting intermittent demand for service parts inventories [J]. International Journal of Forecasting, 2004, 20(3):375-387.
[17] Syntetos AA, Boylan JE. The accuracy of intermittent demand estimates[J]. International Journal of Forecasting, 2005, 21(2):303-314.
[18]曲立,张群.备件管理综述[J].实验室研究与探讨,2006,25(7):875-880.
[19] Bagchi U. Modeling lead-time demand for lumpy demand and variable lead time. Naval Research Logistics, 1987, 34(5):687–704.
[20] Rudi N, Groenevelt H. End-of-period vs. Continuous Accounting of Inventory-RelationCosts[J].Operational Research,2009,57(6):1360-1366.
[21] Baker R. Safety Stocks and Component Commonality[J]. Joumal of Operations Managment.1985, 6(1):13-22.
[22] Baker Retal. The Effect of Commanlity on Safety Stock in a Simple Inventory Model[J].Management Science.1986,32(8):982-988.
[23] Inderfurth K, Minner S. Safety Stocks in a Multi-stage Inventory Systems Under Different Service Measures[J]. European Journal of Operational Research.1998(106):57-73.
[24] Miroslav Z. The Analytic Approach vs. The Simulation Approach to Determining Safety. Stock[J]. Problems and Perspetive in Management.2005,2(3):119-127.
[25] Matsuyama K.Inventory Policy with Time Dependent Setup Cost[J].International Journal of Production Economics.1995(42):149-160.
[26]姚洪义,高云静.价格竞争下基于报童问题的需求模型分析[J].重庆大学学报.2007,30(11):144-147.
[27]周艳菊,邱莞华,王宗润.不同约束下多产品报童问题解的比较研究[J].系统工程与电子技术. 2008,30(1):97-103.
[28] Moinzadah K, Lee H L. Batch size and stocking levels in multi-echelon repairable systems[J]. Management Science, 1986, 32(12): 1567-1581.
[29] Shtub A, Simon M. Determination of reorder points for spare parts in a two-echelon inventory system: The case of non-identical maintenance facilities [J]. European Journal of Operational Research (S0377-2217), 1994, 73(3): 458-464.
[30] Sherbrooke C C. Multi-echelon inventory systems with lateral supply[J]. Naval Research Logistics (S0894-069X), 1992, 39(1): 29-40.
[31]周泓等.仿真使用Arena仿真[M].北京:机械工业出版社,2007.
[32] Flexsim仿真软件介绍[EB].http://wenku.baidu.com/view/a6c8fa4733687e21af45a982.html,2009.
[33]孟祥辉,徐宗昌,刘文宝.需求预测研究综述[J],装甲兵工程学院学报,2009,23(4):21-25.
[34]孙焰.现代物流管理技术-建模理论及算法技术[M].上海:同济大学出版社,2004.
[35]黄意.企业设备维护外包服务商选择评价研究[硕士论文].湖南:湖南大学,2009.
[36] Agrawal N, Smith S.A.. Estimating negative Binomial demand for retail inventory management with unobservable lost sales[J].Naval research logistics,1996, 43(6): 839-862.
[37] Foundations of Finance[EB].http://gsbwww.uchicago.edu/fac/eugene.fama/research/,2009.
[38] [美]哈根等.神经网络设计[M].北京:机械工业出版社,2005.
[39] Tyworth J.E., Guo Y. Inventory control under gamma demand and random lead time[J].Journal of business logistics,1996, 17(1): 291-304.
[40]杨方燕,郭钢.基于ABC分类的安全库存优化算法研究[J].企业管理与信息化,2004,33(l):83-86.
[41]胡文卿,基于SCOR的供应链建模与仿真[硕士论文].上海:同济大学,2006.
[42] ZuoJun Shen. A Joint Location-Inventory Model [J].Transportation Science.2003, 37(1):40-55.
[43] Hau Lee, Mitigating supply chain risk through improved confidence[J].International Journal of Physical Distribution & Logistics Management,2004,34(5):388-396.
[44]王正元,宋建社,何志德,曹继平.一种备件多级库存系统的仿真优化模型[J].系统仿真学报, 2007,19(5):1003-1010.
[45]杨宇航,赵建民,李志忠.备件管理系统仿真研究[J].系统仿真学报.2004,16(5): 981-991.
[46] van Kooten, J. P. J., Tan T. The final order problem for repairable spare parts undercondemnation[J].Journal Of The Operational Research Society,2009,60(10):1449-1461.
[47] Haley C.W, Higgins R C. Inventory Policy and Trade Credit Financing[J].ManagementScience, 1973, 20(1):464-471.
[48] Zheng Y.S. On properties of stochastic inventory systems[J].Management Science, 1992,38(1): 87-103.
[49] Closs, DJ, Nyaga, GN, Voss, MD. The differential impact of product complexity, inventory level, and configuration capacity on unit and order fill rate performance[J]. Journal of Operations Management.2010, 28(1):47-57.