Risk Management Of Quality Accidents Of Pumped Concrete Based On Bayesian Networ

Driven by the social economy in recent years,the construction engineering industry has made remarkable achievements,and the quality of the project has become an important concern for construction enterprises seeking development.In the whole construction process of engineering projects,concrete placement is a crucial link,and its construction quality has a significant impact on the construction quality of engineering projects.Especially with the rapid development of high-rise buildings and large bridges,pumped concrete has become the main way of concrete project construction.However,the construction of pumped concrete is not a simple task,and due to the increasing height of pumping and the gradual increase in construction difficulty,quality accidents such as pipe blocking and pump blocking occur from time to time,showing the uncertainty of the quality of concrete pumping construction.Therefore,it is necessary to establish a scientific uncertainty inference model for concrete pumping quality accidents to monitor and predict the probability of risk occurrence of concrete pumping quality accidents and to achieve rapid discrimination of the causes of concrete pumping quality accidents.In this paper,a concrete pumping quality accident diagnosis model is constructed based on Bayesian networks,and the scientific rationality of the model is verified by using the Nujiang River Special Bridge project as a practical case.Specifically,this study firstly used literature analysis method to sort out the key influencing factors of concrete pumping failure,combined with expert knowledge and construction experience,and established a concrete pumping quality accident risk factor influence index system.Secondly,DEMATEL and ISM methods were used to construct a causal chain model of the influencing factors of concrete pumping quality accidents,and a hierarchical structure diagram of the influencing factors of concrete pumping quality accident risk was drawn to provide a causal structure basis for the Bayesian network model.On this basis,the study transforms the ISM hierarchical structure diagram into a Bayesian network topology and performs parameter learning through expert knowledge,thus constructing a concrete pumping quality accident diagnosis model based on Bayesian network.Through model inference,the most causal nodes and causal chains of concrete pumping quality accidents can be identified.Finally,the feasibility of the Bayesian network model for concrete pumping quality accident diagnosis is verified through an empirical study of the Nujiang River Special Bridge project,and specific recommendations for risk avoidance are proposed based on the model diagnosis inference results.In summary,the Bayesian network-based risk inference and cause diagnosis model for quality accidents has wide practical application value.The model can help construction units control potential risks in advance and effectively avoid property damage and human casualties caused by accidents.In addition,the model can provide scientific and effective guidance after an accident occurs and quickly discern the cause of the accident,thus shortening the quality accident processing cycle.Therefore,this Bayesian network-based quality accident risk inference and cause diagnosis model has great potential for application in actual project construction and is worthy of being widely promoted and used.