The railway artificial intelligence competition is an important way to promote the innovation and application development of railway artificial intelligence technology. The innovation, advanced, independent and application of railway artificial intelligence technology can be effectively promoted by holding the competition. In this process, it is very important to establish the evaluation criteria and theoretical methods for the algorithm, the outcome plan and the team performance. In this paper, a scientific evaluation mechanism is designed for the process of railway artificial intelligence competition, and a multi-dimensional comprehensive evaluation index system is established. Using multi-criteria decision making theory, a comprehensive evaluation model of competition results based on AHP-entropy weight-TOPSIS (technique for order preference by similarity to ideal solution optimization) method is constructed, and the process design is implemented based on the railway artificial intelligence competition platform to screen out excellent teams. Finally, through the application case analysis, it is proved that the evaluation body of this paper can effectively evaluate the performance of the railway artificial intelligence competition team, and the calculation results are consistent with the actual situation. The research results can provide technical reference and theoretical support for the evaluation of the results of the subsequent railway artificial intelligence competition.
Research on railway artificial intelligence competition evaluation system based on AHP-entropy weight-TOPSIS
Ninth International Conference on Electromechanical Control Technology and Transportation (ICECTT 2024) ; 2024 ; Guilin, China
Proc. SPIE ; 13251
2024-08-28
Conference paper
Electronic Resource
English
SAGE Publications | 2024
|Evaluation of smart port development level based on entropy weight TOPSIS method
British Library Conference Proceedings | 2022
|GRA + TOPSIS optimal trajectory completion method based on entropy weight
European Patent Office | 2024
|