A context-sensitive service composition framework for dependable service provision in cyber-physical systems. (2017)

• Record Type: Journal Article
• Authors: Wang, Tao; Niu, Chongli; Cheng, Lianglun; Wang, Jianhua
• Is Part Of: International journal of ad hoc and ubiquitous computing. Volume 24:Number 4(2017)
• Page Start: 237
• Page End: 251
• Publication Date: 2017
• Subjects: CPS -- cyber-physical systems -- service composition -- context-sensitive services -- service spanning tree -- service workflow spanning graph -- service combination optimisation -- service reliability -- service provision -- service-oriented architecture -- SOA -- web services -- service discovery -- workflow execution; Ubiquitous computing -- Periodicals; Embedded computer systems -- Periodicals; Electronic data processing -- Distributed processing -- Periodicals; Wireless communication systems -- Periodicals; Computer architecture -- Periodicals; 004.2
• Languages: English
• ISSNs: 1743-8225

Abstract

Most of the service-oriented architecture-based mechanisms for web services are insufficient to support-dependable service composition with the heterogeneous, dynamic physical entities in a cyber-physical system (CPS). Taking into account the contexts of physical entities, this paper presents a context-sensitive service composition framework in a CPS with dependability requirements. First, an ontology model for context-sensitive service specification of physical entities is proposed. Then, a service workflow spanning graph SWSG is used to map the workflow based abstract process model of a given task to multiple sets of atomic service instances. Finally, we present a two-phase context-sensitive service composition optimisation mechanism, which is able to select out the optimal service providers to implement the task effectively and dependably. The experiment results show that the precision and efficiency of service discovery, the success ratio of workflow execution and the performance of service combination optimisation have been greatly improved with our proposed method.