基于Web服务的业务流程优化研究及应用
|
摘要
业务流程优化问题是港口码头作业流程中的重要问题。通过提高集装箱码头作业流程的重组效率来提高自身竞争力,已成为集装箱港口最为关心的问题之一。由于港口码头作业的业务种类繁多,流程复杂且经常会随着业务操作细节的更新而改变,传统的集中式集装箱管理系统在实际应用中存在着柔性不足、适应性和扩展能力有限、互操作性不强等缺点,不能满足现代港务物流复杂应用的高扩展性和松耦合性的需求,而基于面向服务体系结构的Web服务技术为业务流程优化提供了技术支持。因此,研究基于Web服务的业务流程优化具有积极的理论意义和现实的应用价值。
论文首先深入研究了业务流程优化技术的国内外研究现状,分析了当前比较流行的几种解决业务流程优化方法以及各自存在的问题。研究证明业务流程优化问题属于NP难问题。作者在现有研究的基础上,针对基于Web服务的业务流程优化方法重组效率较低以及综合服务质量较差等问题,提出了基于有向图和改进遗传算法的业务流程优化方法,以提高业务流程优化的综合性能。其中,由于图的构建过程复杂度较高,本文通过反向遍历的方法找到图的构建过程中的部分有效路径,缩减了构建全图的时间开销;为了提高算法的全局搜索能力,引入了适用于求解组合优化问题的遗传算法,给出了有向图中满足业务流程特点的编码方式、适应度函数、相关运行参数以及改进的遗传迭代操作。其中在编码方式中,提出了一种基于矩阵形式的变长染色体编码方式,使得业务流程优化模型中的解空间与遗传算法中的搜索空间能正确对应起来。最后,作者进行了仿真实验,并将本文方法、传统图规划方法以及文献[33]中基于有向层次图方法进行了对比和分析,验证了本文所提出方法的优越性;同时结合重庆港物集装箱综合管理系统,从实际系统业务流程的角度出发,证明了本文方法的可行性和实用性。
Optimization of business process is one of most important issues in port terminal business processes. That to improve container harbor’s competition power, though improving the combination efficiency of container harbor’s business process, has become one of container harbor’s greatest concerns. As the business process is too complex and will change according to the operation, because of the too many and different businesses as well as that there are some shortcomings in traditional centralized container management system in practical applications, such as lacking of flexibility, limited adaptability and scalability, poor interoperability and so on, it cannot meet the modern port logistics demand of complex applications which needs high scalability and loose coupling. For Web services technology based on service-oriented architecture provides technical support for optimizing business processes, it can be seen that it is an active theoretical significance as well as a practical value to release some research on business processes optimization based on Web Service.
In this paper, we first reviewed the current studies on business processes optimization both domestic and abroad, analyzing the merits and defaults of several main-stream research methods which solve the problem of business processes optimization. After a thorough investigation, the issue of business processes optimization is NP-hard problem has been proved. Based on the current studies, to tackle the poor quality in combination efficiency and integral service in optimization of business processes based on directed graph, this paper proposed an algorithm based on the directed graph and an improved genetic algorithm, to improve the comprehensive performance of the Web Service dynamic combination. Due to the high complexity of graph building, the paper found out the partial effective paths in the procedure of graph building by means of back traversal to avoid the wasting time on building the whole graph. To enhance the ability of global search, a genetic algorithm fitting for solving the problem of combination optimization has been brought in to the present coding method, adaptation function, relevant running parameters and improved genetic iteration operation, which satisfy the characteristics of service combination in the directed graph. In the coding method, a matrix form based length varying chromosome coding method has been proposed to match the solution spaces in the combination model and the search spaces in the genetic algorithm correspondingly. Finally, though a simulation experiment which proved the validity of the current method, a contrastive analysis comparing used in this paper and the traditional one as well as the directed level graph method in literature [33] has been carried out to verify the superiority of the method proposed in this paper. In the meantime, considering the Chongqing Container Port Transportation Management System and the perspective of process in system businesses, this method was proved to be both feasible and practicable.
论文首先深入研究了业务流程优化技术的国内外研究现状,分析了当前比较流行的几种解决业务流程优化方法以及各自存在的问题。研究证明业务流程优化问题属于NP难问题。作者在现有研究的基础上,针对基于Web服务的业务流程优化方法重组效率较低以及综合服务质量较差等问题,提出了基于有向图和改进遗传算法的业务流程优化方法,以提高业务流程优化的综合性能。其中,由于图的构建过程复杂度较高,本文通过反向遍历的方法找到图的构建过程中的部分有效路径,缩减了构建全图的时间开销;为了提高算法的全局搜索能力,引入了适用于求解组合优化问题的遗传算法,给出了有向图中满足业务流程特点的编码方式、适应度函数、相关运行参数以及改进的遗传迭代操作。其中在编码方式中,提出了一种基于矩阵形式的变长染色体编码方式,使得业务流程优化模型中的解空间与遗传算法中的搜索空间能正确对应起来。最后,作者进行了仿真实验,并将本文方法、传统图规划方法以及文献[33]中基于有向层次图方法进行了对比和分析,验证了本文所提出方法的优越性;同时结合重庆港物集装箱综合管理系统,从实际系统业务流程的角度出发,证明了本文方法的可行性和实用性。
Optimization of business process is one of most important issues in port terminal business processes. That to improve container harbor’s competition power, though improving the combination efficiency of container harbor’s business process, has become one of container harbor’s greatest concerns. As the business process is too complex and will change according to the operation, because of the too many and different businesses as well as that there are some shortcomings in traditional centralized container management system in practical applications, such as lacking of flexibility, limited adaptability and scalability, poor interoperability and so on, it cannot meet the modern port logistics demand of complex applications which needs high scalability and loose coupling. For Web services technology based on service-oriented architecture provides technical support for optimizing business processes, it can be seen that it is an active theoretical significance as well as a practical value to release some research on business processes optimization based on Web Service.
In this paper, we first reviewed the current studies on business processes optimization both domestic and abroad, analyzing the merits and defaults of several main-stream research methods which solve the problem of business processes optimization. After a thorough investigation, the issue of business processes optimization is NP-hard problem has been proved. Based on the current studies, to tackle the poor quality in combination efficiency and integral service in optimization of business processes based on directed graph, this paper proposed an algorithm based on the directed graph and an improved genetic algorithm, to improve the comprehensive performance of the Web Service dynamic combination. Due to the high complexity of graph building, the paper found out the partial effective paths in the procedure of graph building by means of back traversal to avoid the wasting time on building the whole graph. To enhance the ability of global search, a genetic algorithm fitting for solving the problem of combination optimization has been brought in to the present coding method, adaptation function, relevant running parameters and improved genetic iteration operation, which satisfy the characteristics of service combination in the directed graph. In the coding method, a matrix form based length varying chromosome coding method has been proposed to match the solution spaces in the combination model and the search spaces in the genetic algorithm correspondingly. Finally, though a simulation experiment which proved the validity of the current method, a contrastive analysis comparing used in this paper and the traditional one as well as the directed level graph method in literature [33] has been carried out to verify the superiority of the method proposed in this paper. In the meantime, considering the Chongqing Container Port Transportation Management System and the perspective of process in system businesses, this method was proved to be both feasible and practicable.
引文
[1]范玉顺,工作流管理技术基础[M].北京:清华大学出版社,2001.
[2] W3C,WDSLVersion2.0[EB/OL].www.w3.org/TR/2007/REC-wsdl20-primer-2007626,2007.
[3] W3C,SOAPVersion1.2[EB/OL].www.w3.org/TR/2007/REC-soap12-part0-2007427,2007.
[4] UDDI Version3.0 Published Specification[EB/OL].http://uddi.org/pubs/uddi_v3.htm.
[5]肖军,邓海鸿等,基于Petri网的工作流建模及时间性能分析[J].吉林大学学报,2009,Vol 27(01):104-111.
[6]单修慧,基于高级Petri网的柔性工作流模型映射[D].东营:中国石油大学硕士论文,2011:7-14.
[7]范玉顺,李慧芳.基于时间Petri网的工作流模型分析[J].软件学报,2004,15(1):20-29.
[8] Ghallab,M, Howe,A, Knobloek,C, McDermott,D,PDDL-the planning domain definition language[A],In AIPS-98 Planning Committee 1998[C].1998:438-449.
[9] Keller G,Meinhardt S.SAP R/3 analyzer[J].Business process reengineering based on the R/3 reference model,1994,43(2):530-546.
[10]潘敏,石广田等,基于EPC的工作流建模及实现[J].微电子学与计算机,2008:25(7):44-47.
[11]张劲松,薛春芳等.基于过程链的产品开发过程建模研究[J].华中科技大学学报,2005:33(5):64-66.
[12]张晓东,柴跃廷等.基于业务规则的事件驱动建模方法[J].清华大学学报,1999:39(7):25-28.
[13]陈冬梅,王忠群.基于UML活动图的动态工作流过程模型的研究[J].安徽工程技术学院学报,2009,24(2):61-63.
[14]申利民,隋峰等.基于扩展UML活动图的工作流模型研究[J].计算机应用研究,2009,26(2):587-590.
[15]文必龙等.基于UML2.0活动图的工作流模型研究[J].齐齐哈尔大学学报,2006,22(5):46～48.
[16]王聪,王智学.UML活动图的操作语义[J].计算机研究与发展,2007, 44 (10) :1801-1807.
[17]钟明洋,符云清,蒋盼.基于层次泳道的业务流程最优化建模[J].计算机科学,2011,38(4):192-195.
[18]胡锦敏等.基于ECA规则和活动分解的工作流模型[J].软件学报,2002,13(4):761-767.
[19]李忠星.基于扩展ECA的分布式工作流研究及其应用[D].青岛:中国海洋大学硕士论文,2008:19-38.
[20] Sharma Chakravarthy,Roger Le,Rajesh Dasari,ECA Rule Processing in Distributed and Heterogeneous Environments Proceeding of the International Symposium on Distributed Objects and Applications[J],1999.
[21] Robert M.Event-Oriented Dynamic Adaptation of Workflow Model[J],Architecture and Implementation, University of Leipzig,2003.
[22] MariaCláudia Cavalcantia, Rafael Targinoa, Fernanda Bai?o. Managing structural genomic workflows using Web services[J].Data & Knowledge Engineering ,2005,53(1):45-74.
[23] Peng Yuea, Liping Dia,Wenli Yanga.Semantics-based automatic composition of geospatial Web service chains[J].Computers & Geosciences,2007,33(5):649-665.
[24] Maria Cláudia Cavalcanti.Structural Genomic Workflows Supported by Web Services[C]. Proceedings of the 14th International Workshop on Database and Expert Systems Applications.2003:3-7.
[25] Xutao Du,Chunxiao Xing.Nested Web Service Interface Control Flow Automata[C].4th International Conference on Next Generation Web Services Practices.2008:129-136.
[26]唐长红.基于Web Services技术的科技管理工作流系统设计[J].制造业自动化.2010,33 (1) :64-66.
[27]张驰.基于Web服务的业务流程管理研究与实现[D].武汉:华中科技大学硕士论文,2008:8-24.
[28] HASHEMIAN SV,MAVADDAT F.A graph-based approach to Web services composition[C] //Proceedings of the Symposium on Applications and the Internet.Washington,DC:IEEE Computer Society,2005:183-189.
[29]曹利培,刘静,缪淮扣.基于图的Web服务组合优化的研究[J].计算机科学,2007,34(2):95-99
[30]刘家茂,顾宁,施伯乐.基于Mediator的Web Services无回溯反向链动态合成[J].计算研究与发展,2005,42(7):1153-1158.
[31]刘峰,谭庆平,杨艳萍.基于深度优先搜索的Web服务合成算法[J].计算机工程与科学,2006,28(12):80-82.
[32]邓水光,吴健等.基于回溯树的Web服务自动组合[J].软件学报,2007,18(8):1896-1910.
[33]冯兴杰,王辉,许亚娟.基于有向层次图的Web服务自动组合方法[J].计算机应用, 2011.2,31(2):568-571.
[34] P.Lawrence(Ed.).Workflow Handbook1997[M].Workflow Management Coalition.American: NewYork:1997.
[35] Layna Fischer.Workflow Handbook 2002[M]. USA:Workflow Management Coalition,2002.
[36]黄永刚.Web服务原理及实现[J].甘肃科技纵横,2004(6) :40-43.
[37] OASIS.OASIS Web Service Security (WSS) TC[EB/OL]. Organization for the Advancement of Structured Information Standards, November 2006b.URL: www.oasis-open.org/committee /tc home.php?wg abbrev=wss.
[38]柴晓路,粱宇奇.Web Service技术/架构和应用[M].北京:电子工业出版社,2003:34-89.
[39]杨晓慧.浅析图论的发展、应用与未来[A].论文选登, 2004, 12.
[40] Aho A V Hoperoft J E, Ullman J D. Data Structure and Algorithms[M]. Addison-Wesley Publishing Company, Inc, 1983.
[41]王小平,曹立明.遗传算法-理论、应用与软件实现[M].西安:西安交通大学出版社, 2002.
[42] Jomikow C Z, Michalcwicz Z. An Experimental Comparison of Binary and Floating Point Representations in Genetic Algorithm.In: Proc.Of 4th Inl.Conf. on Genetic Algorithms,Morgan Kaufmann,1991:31-36.
[43]代任,杨雷,张斌,高岩.支持服务组合选取的QoS模型及优化求解[J].计算机学报,2006,29 (7):1167-1178.
[44]邹金安,杨艳萍.图在自动Web服务组合中的应用研究[J].小型微型计算机系统,2008,29 (11):2020-2026.
[45] Ervin Laszlo,阂家撒等.进化—广义综合理论[M].北京:社会科学文献出版社,1988.
[46] LIU Y. Research on Fitness Function in Genetic Algorithm [J]. Journal of Lanzhou Polyechn-ic College,2006,13(3):4-8.
[47]曹丽培,李爱玲,刘静.基于QoS的两阶段Web服务选择方法[J].计算机工程与设计, 2009,30(3):747-751.
[48]周明,孙树栋.遗传算法原理及应用[M].北京:国防工业出版社,1992.
[2] W3C,WDSLVersion2.0[EB/OL].www.w3.org/TR/2007/REC-wsdl20-primer-2007626,2007.
[3] W3C,SOAPVersion1.2[EB/OL].www.w3.org/TR/2007/REC-soap12-part0-2007427,2007.
[4] UDDI Version3.0 Published Specification[EB/OL].http://uddi.org/pubs/uddi_v3.htm.
[5]肖军,邓海鸿等,基于Petri网的工作流建模及时间性能分析[J].吉林大学学报,2009,Vol 27(01):104-111.
[6]单修慧,基于高级Petri网的柔性工作流模型映射[D].东营:中国石油大学硕士论文,2011:7-14.
[7]范玉顺,李慧芳.基于时间Petri网的工作流模型分析[J].软件学报,2004,15(1):20-29.
[8] Ghallab,M, Howe,A, Knobloek,C, McDermott,D,PDDL-the planning domain definition language[A],In AIPS-98 Planning Committee 1998[C].1998:438-449.
[9] Keller G,Meinhardt S.SAP R/3 analyzer[J].Business process reengineering based on the R/3 reference model,1994,43(2):530-546.
[10]潘敏,石广田等,基于EPC的工作流建模及实现[J].微电子学与计算机,2008:25(7):44-47.
[11]张劲松,薛春芳等.基于过程链的产品开发过程建模研究[J].华中科技大学学报,2005:33(5):64-66.
[12]张晓东,柴跃廷等.基于业务规则的事件驱动建模方法[J].清华大学学报,1999:39(7):25-28.
[13]陈冬梅,王忠群.基于UML活动图的动态工作流过程模型的研究[J].安徽工程技术学院学报,2009,24(2):61-63.
[14]申利民,隋峰等.基于扩展UML活动图的工作流模型研究[J].计算机应用研究,2009,26(2):587-590.
[15]文必龙等.基于UML2.0活动图的工作流模型研究[J].齐齐哈尔大学学报,2006,22(5):46～48.
[16]王聪,王智学.UML活动图的操作语义[J].计算机研究与发展,2007, 44 (10) :1801-1807.
[17]钟明洋,符云清,蒋盼.基于层次泳道的业务流程最优化建模[J].计算机科学,2011,38(4):192-195.
[18]胡锦敏等.基于ECA规则和活动分解的工作流模型[J].软件学报,2002,13(4):761-767.
[19]李忠星.基于扩展ECA的分布式工作流研究及其应用[D].青岛:中国海洋大学硕士论文,2008:19-38.
[20] Sharma Chakravarthy,Roger Le,Rajesh Dasari,ECA Rule Processing in Distributed and Heterogeneous Environments Proceeding of the International Symposium on Distributed Objects and Applications[J],1999.
[21] Robert M.Event-Oriented Dynamic Adaptation of Workflow Model[J],Architecture and Implementation, University of Leipzig,2003.
[22] MariaCláudia Cavalcantia, Rafael Targinoa, Fernanda Bai?o. Managing structural genomic workflows using Web services[J].Data & Knowledge Engineering ,2005,53(1):45-74.
[23] Peng Yuea, Liping Dia,Wenli Yanga.Semantics-based automatic composition of geospatial Web service chains[J].Computers & Geosciences,2007,33(5):649-665.
[24] Maria Cláudia Cavalcanti.Structural Genomic Workflows Supported by Web Services[C]. Proceedings of the 14th International Workshop on Database and Expert Systems Applications.2003:3-7.
[25] Xutao Du,Chunxiao Xing.Nested Web Service Interface Control Flow Automata[C].4th International Conference on Next Generation Web Services Practices.2008:129-136.
[26]唐长红.基于Web Services技术的科技管理工作流系统设计[J].制造业自动化.2010,33 (1) :64-66.
[27]张驰.基于Web服务的业务流程管理研究与实现[D].武汉:华中科技大学硕士论文,2008:8-24.
[28] HASHEMIAN SV,MAVADDAT F.A graph-based approach to Web services composition[C] //Proceedings of the Symposium on Applications and the Internet.Washington,DC:IEEE Computer Society,2005:183-189.
[29]曹利培,刘静,缪淮扣.基于图的Web服务组合优化的研究[J].计算机科学,2007,34(2):95-99
[30]刘家茂,顾宁,施伯乐.基于Mediator的Web Services无回溯反向链动态合成[J].计算研究与发展,2005,42(7):1153-1158.
[31]刘峰,谭庆平,杨艳萍.基于深度优先搜索的Web服务合成算法[J].计算机工程与科学,2006,28(12):80-82.
[32]邓水光,吴健等.基于回溯树的Web服务自动组合[J].软件学报,2007,18(8):1896-1910.
[33]冯兴杰,王辉,许亚娟.基于有向层次图的Web服务自动组合方法[J].计算机应用, 2011.2,31(2):568-571.
[34] P.Lawrence(Ed.).Workflow Handbook1997[M].Workflow Management Coalition.American: NewYork:1997.
[35] Layna Fischer.Workflow Handbook 2002[M]. USA:Workflow Management Coalition,2002.
[36]黄永刚.Web服务原理及实现[J].甘肃科技纵横,2004(6) :40-43.
[37] OASIS.OASIS Web Service Security (WSS) TC[EB/OL]. Organization for the Advancement of Structured Information Standards, November 2006b.URL: www.oasis-open.org/committee /tc home.php?wg abbrev=wss.
[38]柴晓路,粱宇奇.Web Service技术/架构和应用[M].北京:电子工业出版社,2003:34-89.
[39]杨晓慧.浅析图论的发展、应用与未来[A].论文选登, 2004, 12.
[40] Aho A V Hoperoft J E, Ullman J D. Data Structure and Algorithms[M]. Addison-Wesley Publishing Company, Inc, 1983.
[41]王小平,曹立明.遗传算法-理论、应用与软件实现[M].西安:西安交通大学出版社, 2002.
[42] Jomikow C Z, Michalcwicz Z. An Experimental Comparison of Binary and Floating Point Representations in Genetic Algorithm.In: Proc.Of 4th Inl.Conf. on Genetic Algorithms,Morgan Kaufmann,1991:31-36.
[43]代任,杨雷,张斌,高岩.支持服务组合选取的QoS模型及优化求解[J].计算机学报,2006,29 (7):1167-1178.
[44]邹金安,杨艳萍.图在自动Web服务组合中的应用研究[J].小型微型计算机系统,2008,29 (11):2020-2026.
[45] Ervin Laszlo,阂家撒等.进化—广义综合理论[M].北京:社会科学文献出版社,1988.
[46] LIU Y. Research on Fitness Function in Genetic Algorithm [J]. Journal of Lanzhou Polyechn-ic College,2006,13(3):4-8.
[47]曹丽培,李爱玲,刘静.基于QoS的两阶段Web服务选择方法[J].计算机工程与设计, 2009,30(3):747-751.
[48]周明,孙树栋.遗传算法原理及应用[M].北京:国防工业出版社,1992.