Research On Key Technologies Of Business Process Scheduling Optimization Based On Individual Queue

With wide applications of Process Centered advanced manufacturing/management modes and Process Aware Information System, business process modeling and optimization has received much attention. Supported by the National Natural Science Foundation (No.50675210) and the Provincial Science and Technology Plan of Zhejiang province (No.2009C31036), this dissertation makes a systematical research on the key technologies of business process scheduling optimization. The main structure of the dissertation includes:Chapter 1 is introduction. The research background of this dissertation is elaborated firstly. Then the related concepts of business processes scheduling optimization are introduced and the research problem, object and assumptions of this dissertation are discussed. Thirdly, the related researches on business processes scheduling optimization are reviewed and the existing problems are summarized. Finally, research target, significance, content and framework of the dissertation are described.Business process modeling for scheduling execution is studied in chapter 2. Firstly, scheduling execution modes of business process are analyzed and the business process model is formalized which adds some necessary information for scheduling execution based on the process reference model defined by WfMC. Then the three basic rules for producing business process model with good logical structure are presented based on structured program design specification, the algorithm for verifying logical structure of business process model is developed accordingly. Finally, expected frequency is defined and its calculation method is proposed. It is necessary to build extended business process model with good logical structure for business process scheduling optimization.Business process task assignment optimization is researched in chapter 3. According to business process model with good logical structure built in chapter 2, based on OPT and M/Hr/1 queueing theory the methods for calculating business process maximum throughput and cycle time are proposed firstly using individual queue mode. Secondly, the mathematical task assignment models and their solutions for maximizing throughput and minimizing cycle time are presented. Thirdly, dynamic re-assignment policy of tasks is developed for further improving business process performance. Finally, a set of simulation experiments are conducted, which verifies the effectiveness and advantages of the proposed methods. In the simulation experiment, the principles and methods that can be used to automatically translate extended business process model into the GPSS simulation model are proposed, which reduces the workload on the developers of business process simulation models and improves model quality.The supportive software system is developed and a case study is conducted in chapter 4. Firstly, the requirements of supportive software system are analyzed and the database conceptual model is also designed. Then, a numerical algorithm is proposed for calculating similarity of business process model. Thirdly, the functional architecture of the supportive software system is established and the key functional module algorithms are designed accordingly. Finally, the proposed methods and their supportive software system for business process scheduling optimization are applied to manufacturing process of Coal-fired steam boiler furnace, which illustrates the practical applicability and efficiency of the proposed methods.In chapter 5, the whole researches in this dissertation are summarized and the innovations are pointed out. Lastly the dissertation gives some suggestions for future research works.