基于生产负荷率预测的多供应商订单分配模型
|
摘要
产业集群的成功依赖于集群整体运作效率,而制造资源配置方式是决定生产运作效率的重要因素。现有的许多资源配置模式,比如区域性虚拟企业合作方式,都是围绕既定的生产任务去实施跨企业的资源重组,而基于负荷率的订单分配策略是在既定的制造资源条件下通过生产订单的调配实现资源的合理利用。基于负荷率的订单分配策略以多个供应商之间的生产负荷率均衡作为优化准则,即订单分配的结果将使得多个供应商之间的生产负荷率在订单完成期内趋于均衡。在此学术思想下,本文研究了供应商生产负荷率预测问题,并基于预测建立了考虑负荷率均衡的订单分配模型。论文的主要研究内容和成果如下:
(1)通过对产业集群生存现状的文献调研和实地考察,指出了出于实惠的“利己”目的导致横向合作来协调负荷是低效率的,进一步阐明了由采购商通过订单分配去调节产业集群内供应商生产负荷的思想。
(2)综述了国内外围绕生产负荷的文献报道,包括企业内负荷计划、企业间负荷协调以及负荷预测等相关研究。从追求目标、基本思想、侧重因素等方面,比较分析了基于负荷率的订单分配策略和区域性虚拟企业合作方式的异同点。
(3)针对订单分配下的供应商生产负荷率预测问题,提出了基于招投标的生产信息获取方法;从生产负荷率的组成分类入手,给出了生产能力与确定负荷的计算表达式,构建了基于灰色预测理论和最大信息熵原理且融合同期及环期信息的不确定负荷预测模型,并利用算例验证了所提出的生产负荷率预测理论与方法的可行性。
(4)引入了相对熵作为生产负荷率均衡度量函数,建立了基于采购成本最小化和多供应商之间生产负荷率均衡化的多目标优化模型。针对提出的模型,进行了算例分析,验证优化模型的有效性。
Whether an industrial cluster could be successful or not is determined by the efficiency of overall operation,the mode of manufacturing resource allocation is important to improve the overall operation. There are two modes in manufacturing resource allocation,one of them is regional virtual enterprise,in which cross-enterprise resources are organized based on the predetermined productive tasks,and the other one is an order allocation strategy based on load rate,in which the orders are allocated base on the conditions of manufacturing resources,these two modes have the same purpose. The order allocation strategy basing on load rate takes the equilibrium among these loading rates of different suppliers as the optimization criterion,which makes these load factors of different suppliers toward equilibrium during the order due date after these orders are allocated. Under this academic thought,this dissertation attaches importance on the problem of production load rate forecasting, order allocation model oriented to the optimized utilization of suppliers, manufacturing resource is constructed based on load rate forecasting.
The main work of this paper can be summarized as follows:
(1) Through literature research and field survey on current existence status of industry clusters, pointed out that the horizontal cooperation that stems from“self-interestedness”purpose is inefficient to coordinate load rate equilibrium, and further clarified the ideas that the purchaser regulates the load among suppliers by allocating orders.
(2) This paper have made a general overview of the present domestic and international research on production load, including internal load plan and control, load coordinate among enterprises and load forecasting methods. Meanwhile, by the comparison of order allocation based on load rate with regional virtual organization, profoundly analyze the same and different points from five respects of pursuing goal, basic thought, major factors, coordination mechanism and implementation situation.
(3) For the problem of production load rate forecasting, a bidding mechanism is adopted to obtain production information from the supplier .A method to calculate production capacity and determined load is brought forth, and forecasting model approach to uncertain load based grey prediction theory and maximum information entropy theory is put forward, comprehensive considering the history information about doxi and link relative. Empirically, the model effectively solves problems
(4) Relative entropy is introduced as inequilibrium measure function. The muti-objective optimization modal is set up based on the optimization crieteria of cost minimization and production load rate among supplier balancing. The simulation studies are implemented foe the above model, and the effectiveness of model are validated.
(1)通过对产业集群生存现状的文献调研和实地考察,指出了出于实惠的“利己”目的导致横向合作来协调负荷是低效率的,进一步阐明了由采购商通过订单分配去调节产业集群内供应商生产负荷的思想。
(2)综述了国内外围绕生产负荷的文献报道,包括企业内负荷计划、企业间负荷协调以及负荷预测等相关研究。从追求目标、基本思想、侧重因素等方面,比较分析了基于负荷率的订单分配策略和区域性虚拟企业合作方式的异同点。
(3)针对订单分配下的供应商生产负荷率预测问题,提出了基于招投标的生产信息获取方法;从生产负荷率的组成分类入手,给出了生产能力与确定负荷的计算表达式,构建了基于灰色预测理论和最大信息熵原理且融合同期及环期信息的不确定负荷预测模型,并利用算例验证了所提出的生产负荷率预测理论与方法的可行性。
(4)引入了相对熵作为生产负荷率均衡度量函数,建立了基于采购成本最小化和多供应商之间生产负荷率均衡化的多目标优化模型。针对提出的模型,进行了算例分析,验证优化模型的有效性。
Whether an industrial cluster could be successful or not is determined by the efficiency of overall operation,the mode of manufacturing resource allocation is important to improve the overall operation. There are two modes in manufacturing resource allocation,one of them is regional virtual enterprise,in which cross-enterprise resources are organized based on the predetermined productive tasks,and the other one is an order allocation strategy based on load rate,in which the orders are allocated base on the conditions of manufacturing resources,these two modes have the same purpose. The order allocation strategy basing on load rate takes the equilibrium among these loading rates of different suppliers as the optimization criterion,which makes these load factors of different suppliers toward equilibrium during the order due date after these orders are allocated. Under this academic thought,this dissertation attaches importance on the problem of production load rate forecasting, order allocation model oriented to the optimized utilization of suppliers, manufacturing resource is constructed based on load rate forecasting.
The main work of this paper can be summarized as follows:
(1) Through literature research and field survey on current existence status of industry clusters, pointed out that the horizontal cooperation that stems from“self-interestedness”purpose is inefficient to coordinate load rate equilibrium, and further clarified the ideas that the purchaser regulates the load among suppliers by allocating orders.
(2) This paper have made a general overview of the present domestic and international research on production load, including internal load plan and control, load coordinate among enterprises and load forecasting methods. Meanwhile, by the comparison of order allocation based on load rate with regional virtual organization, profoundly analyze the same and different points from five respects of pursuing goal, basic thought, major factors, coordination mechanism and implementation situation.
(3) For the problem of production load rate forecasting, a bidding mechanism is adopted to obtain production information from the supplier .A method to calculate production capacity and determined load is brought forth, and forecasting model approach to uncertain load based grey prediction theory and maximum information entropy theory is put forward, comprehensive considering the history information about doxi and link relative. Empirically, the model effectively solves problems
(4) Relative entropy is introduced as inequilibrium measure function. The muti-objective optimization modal is set up based on the optimization crieteria of cost minimization and production load rate among supplier balancing. The simulation studies are implemented foe the above model, and the effectiveness of model are validated.
引文
[1]郎咸平.产业链的阴谋:一场没有硝烟的战争.上海:东方出版社,2008.
[2]蒋正武等.上海混凝土小型空心砌块行业发展现状调查.墙材革新与建筑节能, 2006, (4): 37-39.
[3] STEVENSON M, HENDRY I. Aggregate load-oriented workload control: a review and a reclassification of a key approach[J]. International Journal of Production Economics, 2006, 104( 2):676-693.
[4] Arivind B, Harold J S. Intergration of job shop control systems: A state-of-the–artre view.Journal of Manufacturing Systems, 1984, 1(3): 71-79.
[5] SADAO SHIMO YASHIRO, KENICHI TSODA, HIROSHIAVANE. Input Scheduling and Load Balance Control for a Job Shop[J] . International Journal of Production Research, 1984 ,22 (4) : 597 - 605.
[6] BECHTE W. Theory and practice of load-oriented manufacturing control. International Journal of Production Research, 1988, 26(3): 375-395.
[7]田新时.一种以负荷为中心的生产管理系统.管理工程学报, 1994, 8(2): 97~110.
[8] Potts C N, Whitehead J D. Workload balancing and loop layout in the design of a flexible manufacturing system. European Journal of Operational Research, 2001, 129: 326-336.
[9]刘亚丽,张益华,杨宏波.考虑车间资源负荷比率的制造单元设计方法.成组技术与生产现代, 2004, 21(4);15-19.
[10]刘建国,朱恒民,王宁生.混合流水车间负荷平衡调度的免疫算法.西安电子科技大学学报,2006, 33(2): 655-657.
[11] A. Aubry, A. Rossi, M.-L. Espinouse, M. Jacomino. Minimizing setup costs for parallel multi-purpose machines under load-balancing constraint. European Journal of Operational Research,2008, 187:1115–1125.
[12]曹永宁,焦爱胜,张敏.柔性制造系统负荷分配混合遗传算法研究.机械研究与应用,2009,(1): 28-31.
[13]白俊杰,龚毅光,王宁生,唐敦兵.面向订单制造的可重构制造系统中虚拟制造单元构建技术.计算机集成制造系统,2009,15(2): 313-320.
[14]程方启,王洪飞,叶飞帆.横向型虚拟企业订单分配模型研究.机电工程,2009,29(4): 50-53.
[15]严浩云,李宏余.基于面向负荷的生产控制的交货期设置方法.计算机集成制造系统, 2008, 14(6): 1201-1205.
[16]刘建军,陈庆新,毛宁.基于能力验证的模具交货期可行性分析方法.计算机集成制造系统, 2009, 15(3): 618-624.
[17] Allan S .Carrie. From intergrated enterprises to regional clusters:the changing basis of completion. computers in industry, 2000,(42),:289-298.
[18]杨洵.同质竞争下产业链横向整合的经济绩效分析.科技管理研究,2008, (9): 99-101.
[19]李国富,叶飞帆,方志梅,Ashok Kochhar.面向中小企业的区域性虚拟组织实施策略与方法研究,中国机械工程,2005,16(21):1916~1920.
[20]程方启,叶飞帆,杨建国.水平型制造协作联盟订单分配多目标优化模型研究.计算机集成制造系统, 2008, 14(12): 2317-2321.
[21] Li L.Information sharing in a supply chain with horizontal competition [J]. Management Science, 2002, 48(9): 1196-1212.
[22]龙跃,易树平.基于Lotka - Volterra的汽车零部件横向企业协调决策分析.科研管理, 2009, 30(3): 67-73.
[23]刘晓,李海越,王成恩,储诚斌.供应商选择模型与方法综述.中国管理科学, 2004, 12(1): 139-148.
[24] Kelle P. Miller P A.Stockout Risk and Order Splitting[J]. 1nternational Journal of Production Economics, 2005 , 71 (1): 407-4l5.
[25] A. C.Pan. Allocation of order quantity among suppliers[J]. Journal of Purchasing and Materials Man agement, 1989, 2 5( 3 ): 3 6-3 9.
[26]谢经明,娄平,陈幼平,赫惠清,周祖德.供应链中基于多Agent协商的任务分配机制的研究.中国机械工程,2005, 16(21):1911-1915.
[27] Kawtummachai R, Hop N V. Order allocation in a multiple-supplier environment. International Journal of Production Economics, 2005, 93-94: 231-238.
[28] Arunkumar, N,Karunamoorthy, I,Anand, S,and Ramesh, B T. Linear approa ch for solving a piecewise linear vendor selection problem of quantity discounts usinglexicographic method[J]. International Journal of Advanced Manufacturing Technology, 2006, 28 (11):1254-1260.
[29] Najla Aissaoui, Mohamed Haouari, Elkafi Hassini. Supplier selection and order lot sizing modeling: A review. Computers & Operations Research ,2007,34: 3516– 3540.
[30]姬小莉.供应链订单任务分配模型及其混合遗传算法.西南交通大学学报,2005,40(6): 811-815.
[31]杨沅钊,易树平,高庆萱,王婷.汽车零部件供应商选择双层规划模型及求解.重庆大学学报(自然科学版),2007,30(7): 10-13
[32]董景峰,王刚,吕民,高国安.基于改进蚁群算法的多供应商选择问题求解.计算机集成制造系统, 2007,13(8); 1639-1640.
[33]张亦弛,刘鹏,晏湘涛等.风险规避的武器装备供应商订货任务分配模型.工业工程, 2008, 11(3):96-100.
[34]方小斌,陈治亚,张英贵.烟叶采购混合订单分配优化模型与算法.系统工程,2008, 26(10): 7-12.
[35]潘伟,汪寿阳,华国伟,张金隆.基于模糊权重的多目标订单分配模型.中国管理科学, 2009, 17(2): 80-85.
[36]刘延康.基于熵的预测与决策[硕士论文].广州:暨南大学,2005.
[37]陈琦,马向阳,贺敏.备件需求量的灰色理论预测法.组合机床与自动化加工技术, 2009, (6): 102-104.
[38] Wen-Yau Liang , Chun-Che Huang. Agent-based demand forecast in multi-echelon supply chain. Decision Support Systems , 2006, (42): 390– 407.
[39]傅学芳,刘序,孙肖高.基于Markov链的随机需求预测模型.潍坊学院学报, 2002, 2(2): 17-19.
[40]李勇,吴宝亮,杨秀苔,但斌.基于乘积ARIMA模型的产品不确定性需求预测.系统工程与电子技术, 2005, 27(1): 60-62.
[41] D.SHI, C.QUEK, R.TILANI1 and J.FU. Product Demand Forecasting with a Novel Fuzzy CMAC. Neural Processing Letters(2007) 25: 63–78.DOI 10.1007/s11063-006-9031-8.
[42]张宴,杨东,姚晓栋.基于时间序列法的半导体芯片生产需求预测.兵工自动化2007,26,1:33~34.
[43]华晓晖,闫秀霞.基于神经网络的订单预测研究.华东经济管理, 2007, 21(2): 108~110.
[44]韩小妹,韩景倜,张亮.一种特殊的冲击型负荷模型及其供应链可靠度研究.数学的实践与认识, 2007, 37(20): 1-6.
[45] Ferbar,L.,et al. Demand forecasting methods in a supply chain: Smoothing and denoising. International Journal of Production Economics (2008), doi: 10.1016/j.ijpe.2008.08.042.
[46]封云,马军海.供应链需求预测的非线性方法研究.北京理工大学学报(社会科学版),2008,10(5):82-86.
[47]峻峻,谭崇利,刘悦,尹亚锋.一类产品需求预测的支持向量机方法.上海大学学报(自然科学版), 2009, 15(1): 71-75.
[48]李自强.库存管理中需求预测系统的研究与开发[硕士论文].长沙:东南大学,2005.
[49]杜江,曹岩,白璃.制造企业的产能因素分析及度量.西安工业大学学报, 2007, 10(5): 447-450.
[50]朱成骐,孙宏斌,张伯明.基于最大信息熵原理的短期负荷预测综合模型.中国电机工程学报, 2005,25(19):1-6.
[51]高翔,郑建祥.基于最大熵概念的复杂随机变量统计模型.农业机械学报, 2008, 39(2):43-46
[52]孙棣华,刘卫宁,宋伟.基于相对熵的决策属性均衡性评价模型,系统工程理论与实践, 2001,(6): 83-85.
[53]雷德明,严新平.多目标智能优化算法及其应用.北京:科学出版社,2009.
[54]黎文瑞.随机生产环境下的模具制造单元资源配置方法研究[硕士论文].广州:广东工业大学, 2007.
[55]刘希宋,张涛,成勇.供应商联盟TCO理论的成本结构分析.哈尔滨工程大学学报, 2008, 29(7): 767-772.
[2]蒋正武等.上海混凝土小型空心砌块行业发展现状调查.墙材革新与建筑节能, 2006, (4): 37-39.
[3] STEVENSON M, HENDRY I. Aggregate load-oriented workload control: a review and a reclassification of a key approach[J]. International Journal of Production Economics, 2006, 104( 2):676-693.
[4] Arivind B, Harold J S. Intergration of job shop control systems: A state-of-the–artre view.Journal of Manufacturing Systems, 1984, 1(3): 71-79.
[5] SADAO SHIMO YASHIRO, KENICHI TSODA, HIROSHIAVANE. Input Scheduling and Load Balance Control for a Job Shop[J] . International Journal of Production Research, 1984 ,22 (4) : 597 - 605.
[6] BECHTE W. Theory and practice of load-oriented manufacturing control. International Journal of Production Research, 1988, 26(3): 375-395.
[7]田新时.一种以负荷为中心的生产管理系统.管理工程学报, 1994, 8(2): 97~110.
[8] Potts C N, Whitehead J D. Workload balancing and loop layout in the design of a flexible manufacturing system. European Journal of Operational Research, 2001, 129: 326-336.
[9]刘亚丽,张益华,杨宏波.考虑车间资源负荷比率的制造单元设计方法.成组技术与生产现代, 2004, 21(4);15-19.
[10]刘建国,朱恒民,王宁生.混合流水车间负荷平衡调度的免疫算法.西安电子科技大学学报,2006, 33(2): 655-657.
[11] A. Aubry, A. Rossi, M.-L. Espinouse, M. Jacomino. Minimizing setup costs for parallel multi-purpose machines under load-balancing constraint. European Journal of Operational Research,2008, 187:1115–1125.
[12]曹永宁,焦爱胜,张敏.柔性制造系统负荷分配混合遗传算法研究.机械研究与应用,2009,(1): 28-31.
[13]白俊杰,龚毅光,王宁生,唐敦兵.面向订单制造的可重构制造系统中虚拟制造单元构建技术.计算机集成制造系统,2009,15(2): 313-320.
[14]程方启,王洪飞,叶飞帆.横向型虚拟企业订单分配模型研究.机电工程,2009,29(4): 50-53.
[15]严浩云,李宏余.基于面向负荷的生产控制的交货期设置方法.计算机集成制造系统, 2008, 14(6): 1201-1205.
[16]刘建军,陈庆新,毛宁.基于能力验证的模具交货期可行性分析方法.计算机集成制造系统, 2009, 15(3): 618-624.
[17] Allan S .Carrie. From intergrated enterprises to regional clusters:the changing basis of completion. computers in industry, 2000,(42),:289-298.
[18]杨洵.同质竞争下产业链横向整合的经济绩效分析.科技管理研究,2008, (9): 99-101.
[19]李国富,叶飞帆,方志梅,Ashok Kochhar.面向中小企业的区域性虚拟组织实施策略与方法研究,中国机械工程,2005,16(21):1916~1920.
[20]程方启,叶飞帆,杨建国.水平型制造协作联盟订单分配多目标优化模型研究.计算机集成制造系统, 2008, 14(12): 2317-2321.
[21] Li L.Information sharing in a supply chain with horizontal competition [J]. Management Science, 2002, 48(9): 1196-1212.
[22]龙跃,易树平.基于Lotka - Volterra的汽车零部件横向企业协调决策分析.科研管理, 2009, 30(3): 67-73.
[23]刘晓,李海越,王成恩,储诚斌.供应商选择模型与方法综述.中国管理科学, 2004, 12(1): 139-148.
[24] Kelle P. Miller P A.Stockout Risk and Order Splitting[J]. 1nternational Journal of Production Economics, 2005 , 71 (1): 407-4l5.
[25] A. C.Pan. Allocation of order quantity among suppliers[J]. Journal of Purchasing and Materials Man agement, 1989, 2 5( 3 ): 3 6-3 9.
[26]谢经明,娄平,陈幼平,赫惠清,周祖德.供应链中基于多Agent协商的任务分配机制的研究.中国机械工程,2005, 16(21):1911-1915.
[27] Kawtummachai R, Hop N V. Order allocation in a multiple-supplier environment. International Journal of Production Economics, 2005, 93-94: 231-238.
[28] Arunkumar, N,Karunamoorthy, I,Anand, S,and Ramesh, B T. Linear approa ch for solving a piecewise linear vendor selection problem of quantity discounts usinglexicographic method[J]. International Journal of Advanced Manufacturing Technology, 2006, 28 (11):1254-1260.
[29] Najla Aissaoui, Mohamed Haouari, Elkafi Hassini. Supplier selection and order lot sizing modeling: A review. Computers & Operations Research ,2007,34: 3516– 3540.
[30]姬小莉.供应链订单任务分配模型及其混合遗传算法.西南交通大学学报,2005,40(6): 811-815.
[31]杨沅钊,易树平,高庆萱,王婷.汽车零部件供应商选择双层规划模型及求解.重庆大学学报(自然科学版),2007,30(7): 10-13
[32]董景峰,王刚,吕民,高国安.基于改进蚁群算法的多供应商选择问题求解.计算机集成制造系统, 2007,13(8); 1639-1640.
[33]张亦弛,刘鹏,晏湘涛等.风险规避的武器装备供应商订货任务分配模型.工业工程, 2008, 11(3):96-100.
[34]方小斌,陈治亚,张英贵.烟叶采购混合订单分配优化模型与算法.系统工程,2008, 26(10): 7-12.
[35]潘伟,汪寿阳,华国伟,张金隆.基于模糊权重的多目标订单分配模型.中国管理科学, 2009, 17(2): 80-85.
[36]刘延康.基于熵的预测与决策[硕士论文].广州:暨南大学,2005.
[37]陈琦,马向阳,贺敏.备件需求量的灰色理论预测法.组合机床与自动化加工技术, 2009, (6): 102-104.
[38] Wen-Yau Liang , Chun-Che Huang. Agent-based demand forecast in multi-echelon supply chain. Decision Support Systems , 2006, (42): 390– 407.
[39]傅学芳,刘序,孙肖高.基于Markov链的随机需求预测模型.潍坊学院学报, 2002, 2(2): 17-19.
[40]李勇,吴宝亮,杨秀苔,但斌.基于乘积ARIMA模型的产品不确定性需求预测.系统工程与电子技术, 2005, 27(1): 60-62.
[41] D.SHI, C.QUEK, R.TILANI1 and J.FU. Product Demand Forecasting with a Novel Fuzzy CMAC. Neural Processing Letters(2007) 25: 63–78.DOI 10.1007/s11063-006-9031-8.
[42]张宴,杨东,姚晓栋.基于时间序列法的半导体芯片生产需求预测.兵工自动化2007,26,1:33~34.
[43]华晓晖,闫秀霞.基于神经网络的订单预测研究.华东经济管理, 2007, 21(2): 108~110.
[44]韩小妹,韩景倜,张亮.一种特殊的冲击型负荷模型及其供应链可靠度研究.数学的实践与认识, 2007, 37(20): 1-6.
[45] Ferbar,L.,et al. Demand forecasting methods in a supply chain: Smoothing and denoising. International Journal of Production Economics (2008), doi: 10.1016/j.ijpe.2008.08.042.
[46]封云,马军海.供应链需求预测的非线性方法研究.北京理工大学学报(社会科学版),2008,10(5):82-86.
[47]峻峻,谭崇利,刘悦,尹亚锋.一类产品需求预测的支持向量机方法.上海大学学报(自然科学版), 2009, 15(1): 71-75.
[48]李自强.库存管理中需求预测系统的研究与开发[硕士论文].长沙:东南大学,2005.
[49]杜江,曹岩,白璃.制造企业的产能因素分析及度量.西安工业大学学报, 2007, 10(5): 447-450.
[50]朱成骐,孙宏斌,张伯明.基于最大信息熵原理的短期负荷预测综合模型.中国电机工程学报, 2005,25(19):1-6.
[51]高翔,郑建祥.基于最大熵概念的复杂随机变量统计模型.农业机械学报, 2008, 39(2):43-46
[52]孙棣华,刘卫宁,宋伟.基于相对熵的决策属性均衡性评价模型,系统工程理论与实践, 2001,(6): 83-85.
[53]雷德明,严新平.多目标智能优化算法及其应用.北京:科学出版社,2009.
[54]黎文瑞.随机生产环境下的模具制造单元资源配置方法研究[硕士论文].广州:广东工业大学, 2007.
[55]刘希宋,张涛,成勇.供应商联盟TCO理论的成本结构分析.哈尔滨工程大学学报, 2008, 29(7): 767-772.