Research On Large-scale Construction Objectives Deviation Timely Monitoring Method And Warning Model

In 2014, with the put forward of "new normal", engineering construction industry entering into a critical period of adjustment. Face the trend of "Economic development pattern changed from the speed type to quality efficiency and intensive growth type ", we need to use new ideas to improve the engineering construction industry development pattern. Owing to the high investment and long schedule, the construction process of large-scale consturction projects need more dynamic monitoring, and attaches great importance to the quality, safety performance, building energy conservation, such as the balanced development of multi-objective, strengthen the project management level of refinement. The construction owner is the total organizers of construction projects, and shoulder important management responsibility. However, in recent years, the construction project objective management research at home and abroad mostly focus on the individual objective planning process or implementation process optimization research, few can stand in the perspective of owner, and discuss how to strengthen the supervision and management of the whole construction process. For construction owner who attaches great importance to the management and technology innovation, how to dynamically and comprehensive grasp the project status in the process of construction, how to determine the severity of the deviation between actual status and plan, how to make effective and fair rewards or punishment to the other participants of the project, are all the management difficulties owner faced in the large-scale consturction projects implementation, which is also the focus of this paper.This paper argues that if the construction objectives consistent with project plan, is an important basis for judging project’s health. Therefore, on the basis of a large number of related research on engineering construction plan, control, early warning, this paper puts forward the construction objectives grid monitor mode which take construction owner as the performer of the monitoring. Different with the traditional top-down linear type objective monitor mode, grid monitor mode highlights the timing and comprehensive of the monitoring. In the process of building the structure of the monitoring, transverse monitor dimension is the embodiment of the timely monitoring idea. This dimension concentrate the monitoring whole process into several key monitor milestones and auxiliary monitor milestone, balancing the needs of dynamic and practical. In the process of building the longitudinal objective dimension, we divided the objectives need to be monitored into two categories:the Evaluative objective and March objective, and further list objectives in each kind, to make the monitoring scope clearer. The longitudinal and transverse dimensions constitute the construction objective monitoring grid structure. On the basis of this structure, the objective deviation monitor, grade warning of deviation, and warning response will be further implemented.In the research of objective deviation monitoring, this paper puts forward "deviation value" as the quantitative tools of deviation degree, and four indicators to measure the objective deviation value:Ideal objective plan value, Limit objective plan value, objective timely status value, objective corrected status value. As the objective value of Evaluative objective is relatively independent, stable, and can only get through evaluative; March objective value has the characteristics of continuous, unidirectional change and relatively fixed amount. Therefore, the four indicators’calculation between two types of objective are in big difference. Due to the universality need of overall objective monitoring method, in this paper, two kinds of objective deviation value calculation method have been designed respectively. Aim at evaluative objective, this paper presents a multiple subject evaluative method based on Choquet integral operator. This method adopts the objective implementation evaluative opinions of owner, designer, contractor, supervisor, and eliminate the correlation between indicators and appraises through the Choquet integral operator, to obtain more accurate deviation value results. For the March objective, the deviation value calculation method is proposed based on the multi-level buffer plan. We take out the security in original objective plan, and obtain ideal state and limit state of the objective. Objective deviation value is the result of objective timely monitoring, and the data source of the deviation graded warning model.In the study of grading warning model of the objective deviation, since we take the objective deviation monitoring results as data sources, so in the process of transform the concrete deviation value into abstract warning level, the Evaluative objective and March objective are studied dividedly either. For Evaluative objective, this paper constructed the grading warning model based on kurtosis method-Fisher discriminate method. We selected sample data, and gained preliminary deviation value threshold segmentation through the Fisher discriminate method, and using the kurtosis to improve the discontinuous problem, to get the final threshold segmentation results. By comparing the timely deviation value with the threshold value interval, we can get the warning level. For the March objective, the deviation buffer contrast model was proposed on the basis of multi-level buffer plan, divided the buffer consumption interval, and achieved the graded warning. After the completion of the grading warning model, we do the example analysis, and demonstrated the whole process of building monitoring grid structure, deviation monitoring and grading warning. This will make readers understand the research idea clearly.Finally, this paper designed the expert support system warning response mechanism based on the results of grading warning. This mechanism associates the objectives, milestones, deviation level, and factors leads to deviation through a brief database design. Collect the corrective recommendations of expert, forming a corrective knowledge base. When grading warning comes, through the intelligent reasoning, this mechanism can comprehensive analysis the factors caused objective deviation, and giving advice to the owner, help them to make decision to correct the deviation. To enhance the applicability of the response mechanism, this paper add the feedback and improvement function in the mechanism, constantly adjust the content of knowledge base according to the using effect in actual engineering. The expert support system warning response mechanism is a complement of objective deviation monitoring and grading warning, which can minimize the loss caused by the objective deviation. Through the correct suggestions, it can change the continuous expanding trend of objective deviation, and achieve the closed loop operation of the whole research.