Evaluation Model of Airport Bus Operation Quality Based on Projection TOPSIS

This paper introduces a novel evaluation model of airport bus operation quality based on projection TOPSIS, which solves the problems of insufficient description of airport bus operation quality and the problems of weak classification characteristics and poor flexibility of evaluation results when the traditional approach to ideal solution sorting method is used to evaluate airport bus operation. In our newly proposed method, we first establish an evaluation index system according to the needs of business scenarios, then use the projection TOPSIS method to obtain evaluation results with good and bad grading features, and combine the natural breakpoint method to construct a projection TOPSIS evaluation model, and final training neural network to learn the evaluation empirical knowledge of the projection TOPSIS evaluation model, for getting a more flexible evaluation model. Our model is based on the airport bus data of a large domestic airport. The results show that compared with the traditional TOPSIS method, the evaluation results of the projection TOPSIS method generally have better classification accuracy under the commonly used classifiers, and the accuracy reaches 96.5%, the highest improvement 11.5%, which can clearly and intuitively reflect the operation quality of airport buses.