Research On Scheduling Expert System Of Airport Ground Staff Based On The Rule-based Reasoning

With the rapid growth in passenger traffic of our country’s civil aviation industry,higher requirements have been placed on airport ground handling and service levels.A key decision problem in ground handling is staff scheduling.Especially,when faced with massive flight delays,a quality staff scheduling is pivotal in providing guaranteed levels of operations and services.Therefore,using intelligent technology to optimize the scheduling plan,achieving optimal and automatic staff scheduling is significant from both practical and theoretical viewpoints.First,based on the characteristics of airport ground staff scheduling,this paper split the scheduling process into three parts: demand forecasting,pre-scheduling,and dynamic adjustment of the pre-schedule.Then,using rule-based reasoning method to study the problem of automatic scheduling.The main contributions can be summarized as follows.(1)The problem of forecasting the demand of airport ground staff is studied.According to the characteristics of the check-in post demand,a model of the average waiting time of passengers arriving under a non-Poisson distribution is established,which is used to predict the required number of check-in crews for guaranteed flights.Then,to analyze the task guarantee process,established a task graph model.With this model,transformed the task merging problem into a minimum path coverage problem,then solve it by the bipartite graph maximum matching algorithm,and thus determine the tasks that need to be merged.Finally,the number of different staff required by the airport ground service center is chosen by guaranteeing the airport’s peak period requirements.(2)Based on the rule-based reasoning method,the problem of pre-scheduling of airport ground staff is studied.First,it sorted out the constraints of the airport staff scheduling problem,and adopted the method of rules to express various constraints,and established the knowledge base of the pre-scheduling expert system.Then,an expert system reasoning engine using the forward BFS reasoning method,based on the three modules of pre-processing,selection allocation and conflict elimination,which verifies the feasibility of the pre-scheduling expert system.Finally,by comparing the variance of the historical workload of the staff in the schedule,the system can ensure that the workload is divided fairly.(3)The dynamic adjustment of the scheduling plan under extreme flight conditions is studied.First,it analyzes the non-executable conditions of the pre-scheduling plan,and propose a dynamic adjustment method of the scheduling plan based on abnormal moments.Then a knowledge base for dynamic adjustment of the scheduling plan is established,and thus pre-scheduling expert system gets improved.With the optimization objective being minimum changes of the pre-scheduling plan,a mathematical programming model for the dynamic adjustment of the scheduling plan is established.Finally,the branch and bound algorithm,genetic algorithm and scheduling expert system are used for simulation and comparative analysis.The simulation results show that even under extreme flight conditions,the proposed expert system adapts the staff scheduling more quickly.