【作者】 陈静；

【导师】 孙林夫；

【作者基本信息】 西南交通大学 ， 计算机应用技术， 2011， 博士

【摘要】 随着经济全球化和社会化分工的发展,企业为了能更好的生存和发展,企业间业务协作越来越紧密,业务协作呈现大规模、多目标、多角色和多层次特征,企业之间的竞争已发展为产业链之间的竞争,产业链协作模式也逐渐由以龙头企业为核心的单链式模式拓展到业务关联的多条产业链间的相互协作、共同发展。同时,企业为了顺应市场的快速变化和市场竞争的新要求.需要不断更新自身业务流程和模式,而业务流程的更新和业务模式的变化,必须有配套的信息系统做支撑,才能使企业具有快速的市场响应能力和更大的竞争力。建设适应新业务及新流程等应用的信息系统,从技术上要求新的应用信息系统必须能快速搭建并实施,能够做到“随需应变”。支持企业间柔性业务协作的公共服务平台无疑是中小企业信息化的重要手段之一。因此,针对新的产业链协作模式——业务关联的多产业链协作,研究业务关联的多产业链协作网络模型、特征及其支持业务关联的多产业链的多个业务协作的公共服务平台具有重要的意义。本文结合国家高技术研究发展计划(863计划)“业务关联的中小企业群信息化服务平台开发与应用”(项目编号：2009AA043501)、国家高技术研究发展计划(863计划)“集团企业业务协同服务支持集成平台开发与应用”(项目编号：2007AA04Z188),针对业务关联的多产业链协作网络模型、网络模型演化算法及务关联的多产业链柔性业务协作的公共服务平台模型、架构、盟员优选、平台认证和平台数据安全等相关技术进行了理论和应用研究。具体包括以下几个方面：(1)针对业务关联的多产业链协作模式,抽象出了业务关联的多产业链协作网络,并建立了其数学模型,模型既考虑了节点的近邻数(即节点的度)和节点与近邻之间的边权重,又考虑了节点本身所具备的能力,较好地体现了实际的网络；分析了业务关联的多产业链协作网络特征,基于复杂网络理论,综合考虑了节点的群落特性、节点间双向择优选择机制及节点间协作关系消亡等因素,设计了业务关联的多产业链协作网络模型演化算法；应用业务关联的多产业链协作网络模型和演化算法,对汽车行业业务关联的多产业链协作网络节点强度分布和重要度进行了分析。(2)为构建柔性的业务关联的多产业链协作公共服务平台,提出了面向柔性的业务关联的多产业链协作公共服务平台模型：即基于业务分解、业务建模、构件抽象、构件集成及组合的平台的构建模型和基于构件的平台集成模型；基于业务关联的多产业链协作公共服务平台模型,设计了基于软总线和软构件的公共服务平台架构。在此架构下,平台被设计为公共构件、基础业务构件、专用业务构件和软总线的关系集合,公共构件和业务构件通过在平台的软总线上注册、发布和授权,实现了构件与平台之间的即插即用,从而实现了产业链协作业务信息系统与平台之间的即插即用、平台的可扩展性和对企业柔性业务的支持。本文研究成果已应用到汽车及发动机产业链协作服务平台的开发和应用中,应用结果表明,本文设计的平台模型和架构能较好地支持企业业务变化对信息化的要求,是切实可行的。(3)对基于物元分析的产业链联盟盟员优选与评价方法进行了理论和应用研究,建立了基于物元分析的业务关联的产业链协作联盟盟员优选与评价模型,该模型将评价指标的不可共度性及矛盾性转换为相容问题,利用关联函数和贴近度函数进行综合评价。基于可拓理论,将产业链协同联盟盟员评价结果分为正域、可拓域及负域,该方法不仅能评价待评方案之间的细微差别,具有更高的精确度,更能从变化的角度评价变化的事物。最后,给出了基于物元分析的二阶段产业链协同联盟盟员企业优选与评价算法。另外,本文的研究纠正了以往研究中关于物元分析方法在综合评价应用中存在的一些错误,并开展了相关的理论研究。(4)针对业务关联的多产业链协作公共服务平台特点和基于Web的应用程序应用范围的开放性特点,以及对认证口令在输入及网络传输过程中的保密需求,建立了基于双随机输入单元的软输入模型,基于双随机软输入模型,设计并实现了业务关联的多产业链协作公共服务平台的一次性口令认证方法。该方法的核心思想是,当用户需要输入认证口令时,认证服务器动态生成双随机输入软键盘,即每次生成的软键盘的界面接口布局是随机的,且其接口对应的输入字符也是随机的,随机生成的软键盘是加盖时间戳的,是一次性的。研究结果表明,本文提出的基于双随机软输入模型的一次性口令认证方法,能有效抵御输入截获攻击、内存截获攻击及重放攻击,具有较好的安全性,为基于Web应用程序的身份认证提供了一个新方法。(5)针对由第三方开发和运营的业务关联的多产业链协作公共服务平台数据安全需求,基于数据加密技术、HASH技术、数字签名技术和Web Service技术,设计了“数据银行”模式的业务关联的多产业链协作公共服务平台的数据安全解决方案BMCCP DSS,并设计和实现了与公共服务平台数据安全相关的算法,即业务协作单据提交和签名算法、业务协作单据签名验证算法、加密敏感字段单据解密算法(即,单据在平台客户端显示算法)及业务协作单据修改算法。研究结果表明,本文提出的公共服务平台数据安全解决方案,解决了数据在第三方存储的机密性、传输保密性、数据完整性、发送不可抵赖性及防重放攻击等数据安全问题,且方案实现简单、安全性高。(6)将研究成果应用到汽车及发动机产业链协作公共服务平台的开发中,验证了研究结果的有效性和实用性。更多还原

【Abstract】 With development of economic globalization and social division, the business collaboration among enterprises which presents the large-scale, multi-objective, multi-role and multi-level feature, is more increasingly close. Competition among enterprises has developed into competition among industrial chains and industrial chain collaboration mode has gradually expanded to inter-collaboration and common development among business-related multi-industrial chains from single-chain mode among leading enterprises. Meanwhile, the enterprises, which in order to adapt to the rapid market changes and new demands of market competition, need to update their business processes and business mode. The business processes updating and business mode changing are supported by corresponding information systems, which enable enterprises to further enhance rapid response capability and competitiveness. The new information system constructed to adapt to new business and new process must be able to quickly build and implement on demand. Public service platform supporting flexible business collaboration among enterprises is undoubtedly an important means of small and medium enterprises information. Therefore, the study on model and feature of business-related multi-industrial chains collaborative network and public service platform supporting flexible business collaboration among business-related multi-industrial chains have great significance aiming at new industrial collaboration mode:business-related multi-industrial chains collaboration.The theory, methods and key techniques are researched, including model and model evolutionary algorithm of business-related multi-industrial chains collaborative network, model and architecture of public service platform supporting flexible business collaboration among business-related multi-industrial chains, evaluation and selective preference model of industrial chain collaboration alliance members, authentication method and data security of public service platform. Research works in this dissertation are partial supported by the National High-Tech Research and Development Plan of China:the development and application of the information platform for business-related small-medium enterprise group (No.2009AA043501) and the development and application of collaboration services integration platform for group enterprise (No.2007AA04Z188), which mainly concentrates on the follwing several aspects, i.e.(1) Aiming at collaboration mode for business-related multi-industrial chains, the business-related multi-industrial chains collaborative network is proposed and its mathematical model is established, which not only takes into account the number of neighboring nodes (ie. node degree) and the edge weighted between the nodes and its neighbors, but also capability of the node itself, and better reflects the actual network. Characteristics of business-related multi-industrial chains collaborative network are analyzed. Considering following factors including node community characteristics, two-way preference selection mechanism among modes and the demise of the relationship among nodes, the model evolutionary algorithm for business-related multi-industrial chains collaborative network is put forward based on complex networks theory. The nodes strength distribution and the nodes important degree of business-related multi-industrial chains collaborative network in automotive are analyzed using networks model and model evolutionary algorithm of business-related multi-industrial chains collaborative network.(2) To build flexible public service platform for business-related multi-industrial chains collaboration, a flexible public service platform model is proposed, which include platform construction model base on business decomposing, business modeling, component modeling, component integrating and platform integration model base on component. Based on the platform model, the public service platform architecture based on software integration bus and software component for business-related multi-industrial chain collaborating is designed, which is made up of common components, basic business components, special business components and software bus. The plug and play relation between the component and the platform is realized by component registered, component released and component authorized on platform software bus. So the plug and play relation between the information systems of industrial chain collaboration and the platform is achieved, and the platform scalability and business collaboration flexibility are realized. The research results have been applied to automobile & engine industrial chain collaborative public service platform development and implementation. The research results showed that the model and architecture proposed in the dissertation, which support information requirements for business changing, is practicable.(3) Theory and application research of the evaluation and selective preference method of industrial chain collaboration alliance members are studied based on matter-element. The evaluation and selective preference model is constructed, which converted the contradictions into the compatibility problems. Comprehensive evaluation is conducted by using correlation function and closeness degree function. Based on extension theory, the evaluation result of industrial chain collaboration alliance members is divided into three categories:positive domain, negative domain and extension domain. The method proposed in the dissertation not only can evaluate the nuances difference among programs and has higher accuracy, but also can evaluate things from perspective of changing. Finally, the two-stage evaluation and selective preference algorithm of industrial chain collaboration alliance members based on matter-analysis is designed. In addition, the study corrects some errors of the previous studies on matter-element analysis application in the comprehensive evaluation field, and the relevant theoretical studies are carried out.(4) Aiming at public service platform features of business-related multi-industrial chains collaboration, open feature of web-based application and authentication password confidentiality requirements in process of input and network transmission, soft input model based on double random input unit is put forward. One time password authentication method of public service platform of business-related multi-industrial chains collaboration is designed and implemented based on double random soft input model. The core of this method is:when the user need to input authentication password, the authentication server dynamically generates double random input soft keyboard, the interface layout of soft keyboard generated with time stamp affixed by the authentication server each time is random, and its interface to the corresponding input character is also random. The study results showed that based on this method, the password entered in client and password transmitted over the network are mapped to several groups of different random string each time without any computing in client, effectively solves some password authentication problems including the capture/replay attack, memory and input intercepted assault, and provides a new approach for web-based application authentication(5) Aiming at data security requirements of public service platform of business-related multi-industrial chains which is developed and operated by third-party, the data security solutions of public service platform of business-related multi-industrial chains(BMCCP_DSS) are designed based on data encryption technologies, hash technologies, digital signature technologies and web service technologies, which has feature of data bank. The related algorithms of data security of public service platform are designed and implemented, which include submission and signature algorithm for business collaboration documents, signature verification algorithm for business collaboration documents, decryption algorithm for encrypting sensitive fields in business collaboration documents (ie. documents display algorithm in the platform client) and update algorithm for business collaboration documents. The research results showed that the solutions proposed in this dissertation implement data stored confidentiality in third-party, transmission confidentiality, data integrity, non-repudiation of sending, anti-replay attacks and other data security issues, and that the solutions have the following feature, implementing simply, high security and practicable.(6) Taking the automobile & engine industrial chain collaborative public service platform for sample, the validity of the researching result in this dissertation is verified.更多还原