Research On The Application Of Bayesian Network In Quality Control Of Railway Project

Railway is the artery of national economy, the country’s major infrastructure and thepopularly traffic tools, play an important role in the country’s sustained economic growth andeconomic security, can’t be replaced by other modes of transportation. In recent years, China’srailway construction usher in the timing of the high-speed development, but at the same time,the railway project has the characteristics with large scale, large amount of investment, longconstruction period, complicated techniques, numerous involved participants, high risk,complex management. Many factors may affect quality of project and quality control isdifficult. The occurring of quality problems fundamentally due to lack of quality managementIt’s need to innovate the project quality control methods. On the basis of in-depth study of therailway project quality control, The paper study on the application of Bayesian network inquality control of railway project. By means of the advantage of Bayesian network inuncertainty analysis as well as the powerful information processing function, Regard wholeproject construction as the starting point, to dynamically control the quality of project.Firstly, In order to implementing effective process quality control for railway project, thispaper proposes that by controlling the state of quality in the process of construction to controlthe project quality, then proposes and define the concept of the state of quality, analyze itsmeaning, and study the analysis process to the state of quality.Secondly, on the basis of in-depth study on the characteristics of Bayesian networks aswell as application in related fields. The paper apply the Bayesian network theory to qualitycontrol of the railway projects, proposes the quality control method of railway project basedon Bayesian network, then analyze the application process. Bayesian network is tried to applyto quality control of railway project, realizing dynamic control and process control to railwayprojects and making up for the deficiencies of traditional quality control methods.Thirdly, the application of the Bayesian network model must be within a certain rangeand follow certain operational mechanism. By dividing the quality Control process of railwayproject stage by stage, process by process and level by level, the quality control system ofrailway project and the operating mechanism of the Bayesian network model are determined.Fourthly, The Construction of Bayesian network model of quality control of railwayproject is divided into two phases: First, the structure of the Bayesian network, Second, thedetermination of Bayesian network parameters. The concrete building process: Combiningexpertise and questionnaire data, using Bayesian network structure learning and parameterlearning to assist, applying GeNIeVer.2.0to software to achieve. Through the application ofthis model, realizing prediction and diagnosis of the quality status of the railway project, Besides, When analytical object of the quality status finish construction, it also can diagnosisquality problems.Finally, in order to clarify the specific application of the model above and testing thepracticality of the model, this paper take concrete railway project as an example to introduceapplication of the model.