Research On Controller Workload Assessing Based On Sector Data Analysis

With the rapid development of China civil aviation transportation industry,the existing airspace tends to be saturated,and even some areas are in an over-saturated state.The sector capacity can no longer meet the existing airspace traffic.Meanwhile,the workload of the controller increase with the traffic growth,some regions or terminal control area controller workload even beyond its regulatory threshold.Therefore,accurately assessing the workload of controllers is the basis for allocating regulatory pressure reasonably,ensuring the safe,efficient and orderly operation of aircraft in the sector and enhancing the sector capacity.Firstly,this paper describes the research results of domestic and foreign scholars on the workload assessing method.It also discusses the implementation process and the advantages and disadvantages of the relevant assessing methods,and clarifies the purpose and significance of the workload assessing.Secondly,according to the relevant researcher's research and controller discussion,use 13 sectors complexity index as the independent variables in the research model,and the self-assessment ATWIT value as the dependent variable.It organize and analyze the data from the implementation of regulatory simulation experiment sector complexity.Finally non-linear and linear studies of the controller's workload are made,to analyze what method can quickly and accurately assess the controller workload.In this paper,three methods are used: BP neural network algorithm uses the original 13 variables as the input layer,and then carries out network model training to obtain the controller's workload evaluation value;the BP neural network algorithm based on principal component analysis and factor analysis is based on the first method to reduce the number of nodes in the input layer,and the accuracy of the evaluation model is improved.Finally,a linear modeling method is used to establish the model,and the error is compared with the mean square error of the first two models.The accuracy of the BP neural network algorithm model based on analysis and principal component analysis is higher.