Article Preview
TopIntroduction
The Internet has recently become a global common platform on which organizations and individuals communicate among each other to carry out various commercial activities and to provide value-added services. The term e-service generally refers to service provided over the Internet. Organizations that offer such services are known as e-service providers. The adoption of e-services in business-to-business (B2B) environment, however, arouses the need for a more in depth study on process collaboration across organizations. Most existing research and practice are still focusing on regular e-service process enactment, which represents only the basic knowledge (Chiu et al., 2003b). Effective collaboration requires robustness and quality. More robust B2B process collaboration requires the capturing and dissemination to business partners of the knowledge for correct and effective exception detection and handling, i.e., enforcement. Exception detection in particular has not been adequately addressed in the literature. We distinguish exception detection from exception handling (Chiu et al. 1999; 2001). The former concerns the knowledge of “what” has been deviated from an agreed collaboration process while the latter concerns “how” deviations can be controlled or compensated. Besides these mandatory actions, quality collaboration involves also optional actions that relate to business relationship management. Yet, most work on relationship management focuses on customer relationship management (Tiwana 2001) in the B2C context, instead of B2B. Thus, the problem of process collaboration is challenging because a generally accepted infrastructure for controlling or monitoring the business processes of an organization’s counter-parties is not available.
The study in this paper is motivated by our previous work on the feasibility of modeling e-Contracts based on cross-organization workflows with workflow views (Chiu et al., 2002). We have also studied the engineering of e-Contracts for its enactment (Cheung et al. 2002) and enforcement (Chiu et al., 2003b). Based on these foundation studies together with our recent work on collaborative workflow (Chiu et al., 2009; Wong & Chiu, 2007), we identified the difficulties in modeling e-Contracts (Krishna et al., 2004) and the limitations towards quality collaboration. In particular, requirements solely based on e-Contracts are inadequate because of their incompleteness and ambiguities (Chiu et al., 2003b). Although research on e-service has been steadily progressing, requirement engineering for e-service collaboration beyond basic enactment is almost unexplored.
In this paper, we propose a methodology to structure B2B process collaboration in multiple layers, namely, collaboration requirements layer, business rule layer, and system implementation layer, as well as from multiple perspectives of process enactment, enforcement, and relationship management. Based on our experience on event-driven process execution and exception handling from the ADvanced Object Modeling Environment (ADOME) Workflow Management System (WFMS) (Chiu et al., 1999; 2001), we further utilize event-condition-action (ECA) rules from the active database paradigm (Dayal, 1988) as a unified foundation artifact for eliciting various requirements for cross-organizational e-service process collaboration. In this paper, we do not require the use of a WFMS so that light weight collaborating systems as well as large distributed and scalable systems can be developed. Conceptual models of the three layers can be expressed in the Unified Modeling Language (UML), a widely accepted notation in object-oriented modeling (Lunn, 2002). We believe that B2B process collaboration should be subject to a life cycle similar to that of a software system, i.e., definition, analysis, and realization (Chiu et al., 2003b). This approach facilitates a more thorough understanding of B2B process collaboration from its fundamentals to system implementation, which has been demonstrated in our earlier work on basic cross-organizational process enactment (Cheung et al., 2002) and related applications (Chiu et al., 2005; 2009).