Data-Driven Task And Role Reassignment Method In The Business Process

To improve the efficiency of the execution of enterprise business,more and more information management systems are used.And the track of the operation of manufacturing system is recorded by event logs,which can be used to support improvement of the system execution.Process mining technology is came into being at this time,which aims to find import information from event logs to support business process management,process analysis,process improvement,etc.This is the theory and technical background of the thesis.Individuation and diversification of customer demand restricts the efficiency improvement of manufacturing system.Diversification of demand results in task diversification,and resources(originators and equipment,etc.)need to take more kinds of tasks.The difference of resource behavior affects the efficiency of task execution;this is which the thesis is based on.And the method of role and task reassignment is proposed to improve the efficiency of system execution.First,information of process,originators and time recorded by event logs are used to get some originator behavior based on statistics: the differences of the task kinds of individual originator who perform,the efficiency of different originators who perform same task,the load of originators,the task preference of originator who perform,cooperation preference of originators and cooperation efficiency of originators.They are the basis of role and task reassignment.Then,the role and task reassignment method based on the differences of originator behavior is proposed and the optimization model of the least time of process execution and the balance of the load of originator for multiple-objective is established.Based on NSGA? algorithm,the solution of optimization model is proposed.Tasks and roles in the business process are reassigned through a case study of event log in business process mining field,and the changes are analyzed.Finally,based on historical data,it also has been to verify the accuracy of the algorithm to support the future task assignment by simulation technology.