Research On Distributed Flexible Simulation Technique For Complex Enterprise Process

Enterprise process engineering is a subject for supvervising the development and enactment of enterprise process with the methodology of system engineering.It includes four phrases:process definition, process simulation,process optimization and process enactment. Enterprise process simulation is the chiefly method to analyze enterprise business process dynamicly,and it is also the important step before deployment of enterprise process.Simulation Results of the enterprise process provide the decision foundation for Business Process Reengineering(BPR),therefore,enterprise process simulation has been a research focus nowdays.Ontology model which provides shared knowledge representation for certain domain is a conceptual specification system.Enterprise ontology model accurately describes all the concepts in the enterpris and the relationship between these concepts,so using it we can get a more general and accurate comprehension of enterprise.This thesis uses OWL (Ontology Web Language)to describe the enterprise ontology model, and builds the process simulation model instanced from enterprise process model for process simulation on a shared knowledge representation.Enterprise process model is used for describing all the business contents of a complex enterprise.The model is composed of subprocesses in diferent abstract layers.Each sub-process represents a business process which has an integrated function in the enterprise. Analyzing the performance of all the sub-process in different layers,we can get more deeply comprehension for enterprise process.This thesis researches the deep-first notology-oriented enterprise process simulation technique,and designs a flexible simulation algorithm for the process ontology individual.The proposed algorithm can support effectively the abstract process in the upper layer make use of and analyze the performance of the lower process.There are multiple processes which have different operation mechanism and schedule strategies in a complex enterprise.Each process can be flexibly simulated according to the shedule strategies.So, during complex enterprise process Simulation a sub-simulation-model should be built for each sub-process as well as a simulation engine for each sub-model.Through these engines' collaboration,we can achieve the process simulation and obtain general performance of the complex enterprise.This thesis researches the extent-coordinative simulation algorithm between the process ontologies to achieve the distributed simulaton of complex enterprise process.The multi-agent technique can perfectly support the distributed simulation of complex system.A multi-agent based distributed simulation algorithm is proposed for the complex enterprise process simulation.Synchronization algorithm is a common technique for synchronizing the simulation time clock of all simulation engines.Conservative mechanism is the most useful synchronization algorithm.By improving the traditional conservative mechanism,this thesis proposes a method to calculate the Lookahead in process agent,and designs the synchronization algorithm for the distributed simulation of the complex enterprise process to achieve the flexible simulation and analyzation of enterprise process.In addition, this thesis designs the distributed architecture for complex process simulation system.