CPM revisited - An architecture comparison

1Citations
Citations of this article
5Readers
Mendeley users who have this article in their library.
Get full text

Abstract

In this report we discuss the architecture of the Collaborative Process Managers (CPMs) used to support inter-enterprise collaboration. A CPM must provide three logical functions, the conversation management for handling inter-enterprise document flows; the process management for controlling local workflows of document manipulation and other related tasks; and the action management for invoking local services that actually implement these tasks. Conversation models and workflow models have some similarity as well as considerable differences. The provisioning, interaction and integration of these three functions are very practical challenges faced by many organizations. Particularly, extending existing workflow engines for supporting inter-enterprise business collaboration, has become the common interest of the e-business industry. We shall first compare five different CPM architectures based on our own prototypes, and then propose the architecture characterized by interfacing a conversation manager and a process manager through asynchronous task activation, and by using the conversation manager as the conversation model driven task activator for local process management. Our experience reveals the advantages of this architecture over the others, as it allows the maximal usability of existing workflow system components, supports both conversation flow and local work flow, and provides a dynamic and simple interface between conversation management and process management. Further, by providing conversation managers under different conversation models, a CPM can support multiple inter-enterprise interaction standards. © Springer-Verlag Berlin Heidelberg 2002.

Cite

CITATION STYLE

APA

Chen, Q., & Hsu, M. (2002). CPM revisited - An architecture comparison. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2519, 72–90. https://doi.org/10.1007/3-540-36124-3_5

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free