An investigation into inputs and outputs of collaborative business process (CBP) tasks

In this era of globalization, companies collaborate with one another to pool their resources as well as to remain in the competition. In the collaboration, some business processes of each of the organizations are linked together, these are identified as collaborative Business Processes (cBPs). So...

Full description

Saved in:
Bibliographic Details
Main Author: Pan, Justin Ke Pei
Other Authors: Lee Siang Guan, Stephen
Format: Final Year Project
Language:English
Published: 2010
Subjects:
Online Access:http://hdl.handle.net/10356/39847
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:In this era of globalization, companies collaborate with one another to pool their resources as well as to remain in the competition. In the collaboration, some business processes of each of the organizations are linked together, these are identified as collaborative Business Processes (cBPs). Some Business Process Management (BPM) and Business-to-Business (B2B) systems have been introduced but these systems are often costly to implement and rigid, leading to a lack of adoption. On top of that, no system has been able to dynamically formulate business processes from a given set of business goals and constraints. The Genesis methodology, created by the team in A*STAR Integrated Manufacturing Services and Systems (IMSS), is one that has been successful in formulating dynamic cBP by decomposing high level B2B business goals, criteria,and compound tasks into operational level, granular tasks ideal for direct Web service execution. This is done by dynamic sequencing of primitive tasks through Hierarchical Task Network (HTN) planning algorithm and a Web ontology (BOWL) that supports the hierarchical decompositions and ordering of cBP tasks. However, the Genesis + BOWL system’s lack of description of the tasks’ inputs and outputs makes it impossible to match and invoke the right Web Service to execute each task. In this report, the author presents TIBOD, short for Task Input-Behaviour-Output Description, a method which provides a rich description of each executable BOWL task in terms of inputs, outputs, and its behaviour. TIBOD is modelled as a Web ontology, a machine readable format that can be merged with BOWL, and be used to scope down to matching Web services to realize the dynamically generated task sequence.