Safety rapid evaluation method and system of pump truck upper structure based on combination weighting and agent model

Abstract To accurately and efficiently determine the safety of concrete pump truck upper structure in-service, the method and system for safety performance evaluation of the upper structure based on the combination weighting and the agent model are proposed. Firstly, according to the actual usage of upper structure, the failure status and fault phenomena of key components of the structure are determined. Referring to relevant national standards for pump trucks, the safety evaluation indexes of upper structure are constructed, and the quantitative methods for indexes are designed. Secondly, index weights are calculated by the ranking comparison (RC)-Criteria Importance Through Intercriteria Correlation (CRITIC) combination weighting method. The correlation between the indexes of the evaluation object and the reference indexes at all levels is confirmed by the gray relational analysis (GRA) method and the evaluation level is given. Subsequently, a security performance sample dataset is built. The hyperparameter of gradient boosted decision tree (GBDT) are optimized by the Bayesian optimization algorithm (BO). The agent model for safety performance evaluation of the upper structure is acquired. The safety performance rapid evaluation system of upper structure is developed to realize timely, accurate, and rapid evaluation of the safety performance. Finally, taking the 56X-6RZ concrete pump truck loading system as an example, the scientific and effective methods and systems proposed are verified by comparing it with fuzzy analytic hierarchy process (FAHP), RC, CRITIC, and FAHP-CRITIC, which provided theoretical basis and implementation means for comprehensively identifying and responding to potential safety risks.