Risk And Cause Analysis Of Metro Delay Based On The Improved Functional Resonance Analysis Method

With the development of urban rail transit,metro plays an indispensable role in promoting economic growth and achieving sustainable urban development.As a socialtechnical system with many components,the metro is prone to delay by the coupling interaction of the internal functions,and it’s too difficult to achieve a comprehensive analysis of the system by the traditional accident model.However,the Functional Resonance Analysis Method(FRAM)can provide feasible ideas for explaining the complexity and nonlinearity of the above systems.The objective of the thesis is to describe the functional coupling resonance phenomenon in the metro system,reveal the hidden functional hazards and the causes of train delay.Therefore,the thesis proposed an improved FRAM for risk identification and cause investigation of the metro delay system based on the shortcomings of FRAM in functional identification,network construction,variability coupling characterization and resonance determination.The contents of the thesis are as follows:(1)Construction of FRAM model for metro delay analysis based on 24-TD model and correlation investigation.Firstly,according to the characteristics of the metro system,the improved 24 Model was adopted to set the unsafe event analysis model for train delay(24-TD).Then,24-TD model was utilized to classify the historical operation data of metro to form the results of high-frequency unsafe events.On the basis of the weak links reflected by the statistical results,the FRAM function modules which could characterize the operation process of the metro system were divided,and the catalogue of potential variability in FRAM function performance was established.Finally,from the perspective of redundant link and missing link inspection,the construction process of FRAM network was improved by using the expert questionnaire method.(2)Characterization of FRAM functional coupling based on Monte Carlo simulation and quantified variability.By abstracting the coupling process of functional variability into a numerical operation process,the functional coupling quantification methods for the risk identification and the cause analysis of metro delays were proposed respectively.For the former,the thesis determined the functional variability probability based on the questionnaire method and Triangular Fuzzy Number(TNF)to overcome the generalization of different functional possibilities during the train delay analysis.Furthermore,the thesis identified the risk functions and risk paths that were most prone to coupling and lead to delays through the Monte Carlo simulation.For the latter,the thesis realized the resonance location of the key function of delay by semi-quantitatively characterizing the actual coupling variation and the interaction links of upstream-downstream functions.(3)Case analysis based on the improved FRAM.The FRAM was applied to the risk identification of metro system and the cause investigation of two metro delay events.Firstly,this thesis assigned numerical scores and probabilities to potential variation of functional performance to simulate the possible scores of different functional couplings.By means of delimiting the system tolerance threshold,the result of risk identification reveals five risk coupling path most likely to cause metro delay.Secondly,the actual function variabilities of two delay cases were extracted from the catalogue of potential variability respectively.The consequences of the semi-quantitative calculations show the key functional aggressive paths leading to the delay.Thirdly,the thesis set up barrier protection measures for network links of key function.Finally,the effectiveness of the improved FRAM is verified by comparing the results with the official report.The improved FRAM which provides feasible ideas for the safe production of metro systems,can not only be utilized to identify metro delay risks of the system to accomplish daily supervision and management,but also be exploited to investigate the actual delay causes to achieve systematic delay prevention.