Study On Human Reliability Of High-Speed Railway Dispatching And Commanding System Under Abnormal Conditions

As a nerve center for high-speed railway transport system, dispatching and commanding system is responsible for organizing and commanding high-speed train operation safely, punctually and efficently. To guarantee high-speed rail operation safety and improve operational efficiency, Centralized Traffic Control System (CTC system) and Chinese Train Control System are used in China’s high-speed railway. Under normal conditions, because there is no traffic adjustment problems, CTC system control train route automatically according to train phase adjust plan. Train dispatchers’main job is monitoring the technical condition of trains and equipment with very few human intervention. Dispatchers and equipment are in good working condition and have a high level of system reliability. While when system under abnormal conditions such as equipment malfuction, equipment maintenance, natural disasters, bad weather or various unexpected situations in the process of train running etc., the tasks can’t be accomplished by transport plan and normal procedures. The probability of human errors of dispatchers increases significantly than normal conditions due to lack of knowledge and experience, job stress, inappropriate emergency drills and other factors.In this thesis, based on system theory, cognitive psychology, safety engineering and human reliability theory, a systematic human error identification and human reliability analysis method of high-speed rail train dispatchers(including assistant dispatchers) under abnormal conditions is built by using literature analysis, interpretation structure model, analytic network process, performance evaluation matrix, empirical analysis, evidence theory and Bayesian networks. The main research work can be concluded as follows:(1) Firstly, the structure and function of high-speed railway dispatching and commanding system, dispatching job setting, dispatching equipment and train dispatcher’s tasks are described and analyzed. Then abnoramal conditions in high-speed rail transport is defined and a statistical analysis of abnoramal conditions is made through the relevant data. From the coupling relationship between train dispatcher’s working ability and task requirements, their changes and influencing factors are discussed and then the human risk mechanism of train dispatchers under abnoramal conditions is proposed.(2) By summarizing existing human error classification methods, the advantages, disadvantages and applicability of methods have been reviewed. Existing methods mainly focus on dominant human error identification, but have great difficulties in latent human error identification of decision-making and planning stage. A high-speed railway train dispatching skill-rule-knowledge-based (SRK) cognitive model is established combining human cognitive process model and skill-rule-knowledge-based behavioral theory. Then a new human error classification method for human error identification is proposed. Finally, a case example of temporary speed restriction in high-speed railway train dispatching is presented and the detailed human error types is given. Through interviews with train dispatchers, the practicality of the method is demonstrated, especially in human error identification of decision-making and planning stage.(3) A full-set performance shaping factors(PSF) system is constructed from the collection and review of existing PSF taxonomies. Considering the major characteristics of train dispatching tasks under abnormal conditions, a complete PSF system of high-speed rail train dispatchers is built. Due to the presence of the complex relationship between PSF and human error, the hierarchy structure of PSF is analyzed and a multilevel hierarchical directed graph of PSF is built using interpretation structural model (ISM). According to the interdependence of various factors, the PSF influence weight is calculated by analytic network process (ANP). Then the superiority of the ANP method is demonstrated by comparative analysis of the results of ANP and AHP. A questionnaire is designed to conduct human reliability influencing factors investigation of high-speed train dispatchers based on PSF system. The performance evaluation matrix approach is applied to explore the key factors contribute to human reliability. The results show that workload under abnormal conditions, understanding and awareness of abnormal events, safety reports and feedback, procedures operability are key factors and need to improve, is high-speed rail dispatching urgent need to improve the people were a key factor in the risk.(4) By summing up existing human error probability calculation methods, a new method to quantify the probability of human error for high-speed rail train dispatchers is proposed through the integrated use of Bayesian networks, Evidence theory and SLIM method. A case example of dispatching order release in temporary speed restriction task is presented to demonstrate the practicality of proposed approach. Error probability, error detectability, error importance and accident severity are used to evaluate human error risk. Due to the uncertainties and imprecisions, the risk factors are characterized with intuitionistic triangular fuzzy numbers. Four factors’objective weights are calculated based on intuitionistic fuzzy entropy. At last, a human error risk prioritization model for high-speed railway dispatchers based on intuitionistic triangular fuzzy TOPSIS is constructed.