The Estimation About Flight Support Service Time Of Hub Airport

For the convenience of airport to publish the accurate information about flight status to passengers in time and aviod extreme events,estimating the service time of airport flight support was needed.In this research the service time of airport flight support was studied from two aspects(static and dynamic)respectively.On the one hand,considering the flight support service procedure is a mixed procedure of job shop and fixed site,and with characteristics of time window constrains and resource demand difference,a model of flight support service procedure based on vehicle routing problem with time windows(VRPTW)was built.For the strong NP nature of vehicle routing problems,a two phase hybrid heuristic algorithm based on greedy algorithm and tabu search was proposed.And then,applied it to the actual operation data of a large domestic hub airport,implementing the support service time estimation under the conditions of flight density change,vehicle number change and flight model change.In the end,accuracy test demonstrated that the model and algorithm could estimate the flight support service time of hub airport as well as flight status,effectively.On the other hand,the service time of airport flight support based on dynamic Bayesian network was studied in a static method of research and analysis,and an estimation model of flight support service time based on Bayesian network(BN)was proposed.The knowledge of aviation experts and the machine learning to historical data were combined by the model,and the incremental learning characteristic of BN was used to adjust the model dynamically,so as to adapt itself to new conditions and constantly update the service time estimates of flight support.When using the data selected from a large domestic hub airport information system,to train the BN model by the algorithm expectation maximization(EM),the test results was obtained.The analysis of experimental results and model evaluation show that the proposed method can effectively estimate the service time of flight support and has higher accuracy.In addition,sensitivity analysis demonstrates that the flight density during flight arrival time has the strongest influence on flight support service time.