4PL环境下基于Multi-Agent的信任评价模型研究

发布时间：2018-01-13 10:04

本文关键词：4PL环境下基于Multi-Agent的信任评价模型研究　出处：《暨南大学》2015年硕士论文　论文类型：学位论文

【摘要】：伴随经济全球化的深入和电子商务的日益繁荣,物流在社会经济中的重要支撑作用不断地被突显,助推着全球经济技术的快速发展。为满足物流业自身的多国化、全球化的需求发展,以信息技术为主导一场物流革命悄然兴起,第四方物流(Fourth Party Logistics,4PL)概念迅速崛起,在国内,阿里巴巴、银泰等集团纷纷投入巨资建立中国智能物流骨干网项目(China smart Logistic Network,CSN)打造第四方物流平台。通过研究4PL信息平台,不仅可以整合现有的第三方物流(Third Party Logistics,3PL)资源,还通过大大地降低物流准入门槛,引入巨大的社会潜在资源,节省物流成本,充分提高国家乃至全社会的运行效率。4PL平台是运行在开放的复杂社会网络之中,众多参与者具有多元和显著的社会化特征。参与者所持目的未知性,使得参与平台中的个体可能因互相无法信任而不能在平台中进行正常的经营、金融等活动,干扰了4PL平台的正常运作。因此进行4PL平台的信任模型研究具有重要意义,也是目前复杂社会网络的研究热点之一。本文提出一种4PL环境下基于多代理信任评价模型与信任评估算法。在信任评价模型建立方面,针对4PL应用场景复杂的个体特性,提出采用多代理(Multi-Agent)的方法,利用代理的自治特性降低环境中信任计算的复杂性;在信任评估算法方面,着重考虑环境中个体对信任的认知差异所带来的评价差异对反馈可信度的影响,提出一种基于个体差异的信任评价算法(Trust Evaluation base on the Individual Difference),动态修正反馈信任,提高信任评估的准确率;基于前面提出的信任评估模型与信任评估算法,本文设计了一个基于多代理的4PL信任评价系统并对其进行仿真实验,仿真实验结果表明,对比现有的信任模型,本文所提出的信任评价模型具有较好的表现。
[Abstract]:With the deepening of economic globalization and the increasing prosperity of e-commerce, the important supporting role of logistics in the social economy is constantly highlighted. The rapid development of global economy and technology. In order to meet the needs of the logistics industry itself and the development of globalization, the information technology is leading a logistics revolution rising quietly. Fourth party logistics fourth Party logistic 4PL concept rose rapidly, in the domestic, Alibaba. Yintai and other groups have invested heavily in the establishment of China smart Logistic Network. By studying 4PL information platform, we can not only integrate the existing third Party logistics 3PL resources. By greatly lowering the threshold of logistics entry and introducing huge social potential resources, the cost of logistics can be saved. To fully improve the efficiency of the country and even the whole society. 4PL platform is run in the open complex social network, many participants have multiple and significant social characteristics. The purpose of the participants is unknown. The participants in the platform may be unable to trust each other and can not carry out normal operations, finance and other activities in the platform. It interferes with the normal operation of 4PL platform, so it is of great significance to study the trust model of 4PL platform. This paper presents a multi-agent trust evaluation model and a trust evaluation algorithm based on 4PL. In view of the complex individual characteristics of 4PL application scenario, a multi-agent multi-agent method is proposed to reduce the complexity of trust computing in the environment by using the autonomy of agent. In the aspect of trust assessment algorithm, we focus on the impact of the evaluation difference on the feedback credibility caused by the cognitive difference of individual trust in the environment. A trust Evaluation base on the Individual difference algorithm based on individual difference is proposed. Dynamic correction of feedback trust to improve the accuracy of trust assessment; Based on the trust assessment model and the trust assessment algorithm, a multi-agent based 4PL trust evaluation system is designed and simulated. The simulation results show that 4PL trust evaluation system is based on multi-agent. Compared with the existing trust models, the trust evaluation model proposed in this paper has a better performance.
【学位授予单位】：暨南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F259.2;TP309

【参考文献】