【作者】 吴钟；

【导师】 聂规划；

【作者基本信息】 武汉理工大学 ， 产业经济学， 2013， 博士

【摘要】 Web服务组合是现代服务业与信息产业融合的产物,它由众多相对简单的Web元服务按照一定的业务流程逻辑组合而成。随着网络客户对服务质量(QoS)要求的提高,网络服务集成商(WSI)开始更加关注网络客户的个性化需求。在Web服务组合的市场决策与优化管理过程中,也逐渐出现了元服务供应商(ISV)、其他合作者甚至是客户的参与。目前,对Web服务组合技术的相关研究工作已经日趋成熟,而在同样重要的Web服务组合管理领域,却存在着众多有待被研究的问题,如：如何提高其灵活性和互动性来满足不同客户的个性化需求,如何完善Web服务组合服务质量指标体系来实现服务功能和服务质量的有效扩展,如何优化配置WSI和ISV的网上网下服务资源,如何实现WSI、ISV、其他合作者与客户的多边协同来优化管理Web服务组合等。本文针对上述问题,利用网络消费者行为理论、价值链理论、协同管理理论、模糊数学、灰色局势决策理论等相关知识和方法来寻求解决途径。首先研究面向协同的Web服务组合QoS指标体系,并在此基础上,对基于客户反馈的Web服务组合QoS关键指标进行提取；然后基于关键QoS指标,研究Web服务组合市场决策模型；最后提出面向多边协同的Web服务组合优化管理方法。主要研究内容和创新如下：(1)建立完善的Web服务组合综合QoS指标体系。从客户偏好出发,在分析网络客户行为的基础上,深度挖掘面向供应商的SQoS指标和面向客户的CQoS指标,提出CQoS与SQoS指标映射机制,建立面向协同的Web服务组合综合QoS指标体系,研究QoS指标体系的评价模型,提出QoS指标权重的确定方法。该综合QoS指标体系将SQoS指标和CQoS指标有效统一起来,达到了QoS指标体系设计的完整性和互动性目的。(2)提取基于客户反馈的Web服务组合QoS关键指标。收集客户反馈信息,利用统计工具对其进行数据化处理,在Web服务组合综合QoS指标体系的基础上提取Web服务组合的关键QoS指标——Web服务组合价格、服务综合满意度、服务描述一致性、服务费用合理度、响应时间、成功率和可用性,构建基于客户反馈的Web服务组合QoS关键指标模型。该Web服务组合的关键QoS指标为面向多边协同的Web服务组合市场决策与优化管理提供了重要依据。(3)构建基于关键QoS指标的Web服务组合市场决策模型。针对客户对服务功能的需求与偏好,提出WSI投资组合优化、双寡头WSI定价策略和ISV选择等Web服务组合市场决策模型,为WSI的市场决策与选择提供了理论依据。通过建立WSI投资组合优化模型,提升了Web服务组合的优越性,实现了客户的个性化服务组合供应和WSI利润的最大化；通过建立双寡头WSI定价策略模型,为对称性垄断和非对称性垄断市场下的双寡头垄断WSI定价策略的制定提供参考；通过建立ISV选择模型,解决了WSI合理选择能提供相同或不同功能的ISV以提高自身经济效益的问题。(4)提出面向多边协同的Web服务组合优化管理方法。将价值网理论引入到Web服务组合协同管理中,研究面向多边协同的Web服务组合与元服务的协同管理、线上服务与线下服务的协同管理和基于价值网的Web服务组合协同管理方法。通过该方法,以投资组合优化模型、定价策略模型、ISV选择模型为导向,WSI与外部的ISV、合作者以及客户之间通过不断的循环反馈而相互作用,使Web服务组合与Web元服务之间、线上服务与线下服务之间逐渐达到一种动态平衡的状态,推动了Web服务组合动态地演化与优化,实现了Web服务组合链的价值最大化,促进了现代服务业与信息产业更好的融合。更多还原

【Abstract】 Web service composition is an integrated product of the information industry and modern service industry. It is formed by lots of relatively simple web services allocated by certain business process logic. With the increasing quality of service (QoS) requirement by network customers, web service integrators (WSIs) begin paying more attention to their individual needs. Independent service vendors (ISVs) and other collaborators or even customers gradually get involved in the process of web service composition marketing decision and web service composition optimization management.Currently, the study of web service composition technology is fledged, while many problems still need to be studied in web service composition management, such as how to improve the flexibility and interactivity of web service composition to meet the individual needs of different customers, how to improve the QoS indicator system of web service composition to achieve the effective expansion for the service functions and QoS, how to optimize and configure the online and offline service resources of the WSIs and ISVs and how to achieve a multilateral collaboration oriented web service composition optimization management by WSIs, ISVs, other collaborators and customers.To solve those problems above, this dissertation seeks to use online consumer behavior theory, value chain theory, collaborative management theory, fuzzy mathematics, grey situation decision-making theory and other related knowledge and methods. First, it studies a collaboration oriented web service composition QoS indicator system. On that basis, it extracts key QoS indicators of web service composition from customers’feedback. Then, it studies the web service composition marketing decision models based on the key QoS indicators. At last, it puts forward a multilateral collaboration oriented web service composition optimization management method. The main contributions of this thesis are as follows:(1) It builds a qualified web service composition QoS indicator system. Starting from the customer preference, it mines deeply the SQoS indicators from the web service composition providers’view and the CQoS indicators from customers’ view, puts forward a CQoS and SQoS indicators mapping mechanism, builds a collaboration oriented QoS indicator system, studies the evaluation model of the QoS indicator system and proposes the method for determining the QoS indicators’ weights based on the analysis of network customer behaviors. The QoS indicator system effectively unifies the SQoS indicators and CQoS indicators, realizes the integrity and interaction of the indicator system designing.(2) It extracts key QoS indicators of web service composition from customers’ feedback. It collects customers’feedback information and uses statistics tools to turn those into data, then extracts the key QoS indicators (the price of web service composition, the total service satisfactory, the service description consistency, the service expense valuation, the response time, success rate and service avalability) of web service composition to build a key QoS indicator of web service composition based on customers’feedback. The key QoS indicator of web service composition underlies the multilateral collaboration oriented web service composition marketing decision and optimization management.(3) It builds the web service composition marketing decision models based on the key QoS indicators. In light of the customers’ needs and preferences for services’ functions, it puts forward several web service composition marketing decision models such as the optimized investment composition model for the WSI, the WSI pricing strategy model in duopoly market and the ISVs selection model. It also provides the theoretical basis for the WSI marketing decision and selection. By building the optimized investment composition model for the WSI, the web service composition is enhanced, the customers’individual web services are provided collaboratively and the WSI profit is maximized. By building the pricing strategy model in duopoly market, it provides references for establishing the duopoly WSI pricing strategy model in symmetry and asymmetry monopoly market. By building the ISVs selection model, it solves the problem of reasonably selecting the ISV of similar or different functions to gain more profits.(4) It puts forward a multilateral collaboration oriented web service composition optimization management method. It applies the Value Net Theory in the web service composition collaboration management, and studies the multilateral collaboration oriented management between web service composition and web services, online services and offline services, and web service composition in the value net. Applying this method, WSI which is oriented by the optimized investment composition model, the pricing strategy model in duopoly market and the ISVs selection model is interacted among the external ISV, collaborators and customers by the ongoing feedback. And it gradually forms a dynamic balance between the web service composition and web services, and between the online services and offline services. It makes web service composition dynamically evolved and optimized, maximizes the value of the web service composition chain and integrates the modern service industry and information industry.更多还原