Authors:
Liping YANG
;
Ying LIU
and
Xin ZHOU
Affiliation:
IBM China Research Lab, China
Keyword(s):
Computational Model, Balance Equation System, Bottleneck, Business Process Model
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Business Process Management
;
e-Business
;
Enterprise Engineering
;
Enterprise Information Systems
;
Information Systems Analysis and Specification
;
Knowledge Management and Information Sharing
;
Knowledge-Based Systems
;
Modeling Formalisms, Languages and Notations
;
Requirements Analysis And Management
;
Symbolic Systems
Abstract:
During designing a business process model, to predict its performance is very important. The performance of business operational process is heavily influenced by its bottlenecks. In order to improve the performance, finding the bottlenecks is critical. This paper proposes a theoretical analysis method for bottleneck detection. An abstract computational model is designed to capture the main elements of a business operational process model. Based on the computational model, a balance equation system is set up. The bottlenecks can be detected by solving the balance equation system. Compared with traditional bottleneck detection methods, this theoretical analysis method has two obvious advantages: the cost of detecting bottlenecks is very low because they can be predicted in design time with no need for system simulation; and it can not only correctly predict the bottlenecks but also give the solutions for improving the bottleneck by solving the balance equation system.