Research On Architecture Based Reliability Assessment Of TPIS System

TPIS( Travelsky Passenger Information and Service System) is a core system focused by the nation and the Civil Aviation Authority, with 21 large multifunction and architecture complex subsystems. It plays an important role in the field of civil aviation. Reliability of the TPIS system is very important to ensure normal service provided to airlines, airports, air traffic control, agents and passengers.In terms of module reliability, constituent source, performance and functionality of modules in each TPIS subsystem are quite different. Based on the DTMC model, idea of Black Box is integrated into the White Box model. A competitive match selection method was put forward for module reliability instead of a single fitting algorithm. The Bayes estimation uses prior information to better understand the statistical characteristics of failure data. The Bayes estimation model were incorporated into three candidate models, NHPP, Shchneidwind model and Goel-Okumoto model, to get module reliability. Failure statistics features are fully grasped. Reliability fitted parameters for each module is closer with the true value, so that the final results of the assessment are more authentic.In terms of the interaction mode between the modules, for the existing model assumptions, failure of modules should be independent and inter-module migration should be accord with Markov random property. With TPIS subsystems complex structure, there are failures with dependencies and deterministic module migrations. In order to achieve a particular function, the interaction patterns between modules presented diversity. Accordingly,special module interaction patterns of TPIS system has been classified and analyzed. Then each special module interaction pattern has been individually analyzed to meet the model assumes of the corresponding mode of transformation process. The difference defect between the original model and TPIS real system has been make up, so that the final results of the assessment will be more accurate.Finally the TPIS system reliability assessment and critical parameters has been analyzed,and key modules of the system are pointed out. On the one hand, key modules of the system with a large influence on the reliability has been listed. On the other hand risking key modules of TPIS system are listed. Thereby feasible technical references to increased system reliability have been provided.