Title: Evaluating the trustworthiness of BPEL processes based on data dependency and XBFG

Authors: Chunling Hu; Cuicui Liu; Bixin Li

Addresses: Department of Computer Science and Technology, Hefei University, No. 99 Jinxiu Avenue, Hefei, Anhui Province, China ' Software Engineering Institute, Southeast University, No. 2 Southeast University Road, Nanjing Jiangsu Province, China ' School of Computer Science and Engineering, Southeast University, No. 2 Southeast University Road, Nanjing Jiangsu Province, China

Abstract: Composite services implement value-added functionality by composing service components of various granularities. Trust is an important criterion to judge whether a composite service can behave as expected. There is a great need for a flexible trust evaluation method for composite services. In this paper, a data dependency-based trust evaluation approach for composite services is proposed. Firstly, we derive define-use pairs of variables to identify data dependency between service components in BPEL processes modelled by an extensible BPEL flow graph (XBFG); then, dependency links including both direct and indirect data dependencies are used to evaluate the trust of these service components; furthermore, on the basis of BPEL structure and XBFG, reduction rules are proposed to evaluate the global trust of BPEL processes. Experimental results demonstrate that the proposed approach is effective for the trust evaluation of BPEL composite services and stable with the growing number of service components in BPEL.

Keywords: composite service; trust evaluation; data dependency; dependency link; reduction rules; business process execution language; BPEL; extensible BPEL flow graph; XBFG; service component; value-added functionality; global trust.

DOI: 10.1504/IJCSE.2020.105221

International Journal of Computational Science and Engineering, 2020 Vol.21 No.1, pp.146 - 161

Received: 27 Feb 2018
Accepted: 21 May 2018

Published online: 22 Feb 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article