Research On Time Workflow Model In Distributed Cooperative Work Environment

With the fast development of economic globalization, distributed cooperative work environment is the inevitable choice of operation environment of modern enterprises' business processes. Workflow management technology, which can model, optimize and instantiate business processes, offers a suitable solution to business processes reengineering. Building a workflow model which can describe business processes formally and be treated by computers is a key factor to make use of workflow management technology in enterprises. Because of the features of distribution, cooperation and dependence of modern business processes, workflow model must be able to express time constraints of multi-granularity, multi- time zone and cooperation. In the last decade, many workflow models such as activity net, IDEF, Petri net and Dialogue had been proposed to meet various application requirements. But they all focused on modeling asynchronous and structured processes, lacks an unified workflow model to address time constraints of distributed cooperative work environment.Two challenges exist for modeling time constraints in workflow: First, the semantic meaning of time is very complex, which has the characters of granularity, time zone, relativity and absoluteness, and we have to establish semantic affinity algorithm of static state and dynamic evolution; Second, the performance analysis of time is very complex, for the time delay of activity don't satisfy the exponential distribution which is required by queue theory and Markov chain, thus it can't hold the classic analysis methods. Although researchers had made promising progresses recently and the workflow model based on time Petri can achieve some good results, modeling time constraints still remains a challenging problem in workflow modeling. Our goal is to establish a formal and unified time workflow model for the processes in distributed cooperative work environment.In this dissertation, we present an integrative workflow model for processing time constraints in distributed cooperative work environment from semantic layer, presentation layer and performance analysis layer. This model starts with semantic layer, incorporating an integrative time constraints concepts based on the distributed and cooperative features of business processes, then figures processes by extended time Petri net, after that we provide the analysis algorithm of time consistency and performance.Specifically, we make the following contributions: By incorporating time space concept into semantic layer of workflow model, we providetime constraints concepts and time consistency checking approach of multi-granularity and multi-time zone, which can check the time consistency in building-time and running-time. â–  By incorporating collaborator concept into semantic layer of workflow model, we provide synchronization-cooperation concepts, and analyses the data and state dependences of synchronization-cooperation activities.By analyzing the deficiencies of time workflow net and incorporating time mapping function into time workflow net, we define the extended time Petri net, and analyses it's firing rules and running processes. By analyzing the structures of extended time workflow net, we propose time consistency algorithm which time complexity is linear. By incorporating transition probability finished firing functions into extended time Petri net, we define the performance analysis model of extended time Petri net, so we can get the performance index such as throughput.After introducing the concept of workflow model and the problems of modeling time constraints in workflow model, we discuss the concepts of multi-granularity, multi-time zone and synchronization-cooperation in workflow model in Chapter 2 and Chapter 3; Chapter 4 proposes the extended time Petri net and it's properties, constructing rules and firing rules; In Chapter 5 and Chapter 6, we discuss the time consistency checking and performance analyzing algorithms; Chapter 7 introduces the application in the workflow management system supported DFMC based on our time workflow model; Finally, we conclude this dissertation and discuss some directions for future work..