A Trust-Based Approach to Selection of Business Services
Sufen Li,Yushun Fan,Xitong Li
DOI: https://doi.org/10.1080/0951192x.2011.574235
IF: 4.1
2011-01-01
International Journal of Computer Integrated Manufacturing
Abstract:With the increasing popularity of the service-oriented thinking in the business area, an overwhelming number of business services have arisen. These business services collaborate with each other and form a business network or a service-oriented business ecosystem (SOBE). It is a key issue to select appropriate business services to share and integrate them in the SOBE. Trust degrees play an important role in the selection of business services. However, existing researches mainly focus on quality of service (QoS)-based selection methods and few of them take trust degrees into account. In order to select services more accurately, a trust-based approach for selection of business services is proposed based on the formal definition of the SOBE. First, the method for acquisition and calculation of trust values is developed. The method also considers other parameters including QoS attributes, the physical distance between users and business services and the waiting time. Second, one node in a business flow may need to select two or more business services to achieve the user's requirements. Thus, the multi-service selection method for one node is presented. Third, a fuzzy chance-constrained programming model is proposed for the selection of business services by considering four kinds of factors: QoS attributes, trust relationship, physical distance and waiting time. Using the characteristic of fuzzy constraints, a function is developed to convert the fuzzy chance constraints into its crisp equivalents and then the equivalent crisp model is solved by the generic algorithm (GA)-based algorithm. For the case in which no satisfying solution can be found by the GA-based algorithm, a negotiation method is proposed to reach at a satisfying solution. Finally, a case study is conducted to demonstrate the feasibility and effectiveness of the proposed approach.