The Research On Cost Control Of Metallurgy Construction Project

The metallurgical construction project exhibits some features such as long construction period, large-scale investment, huge construction quantities and high degree of complexity. In order to take the project into operation as soon as possible, the CM approach was adopted at the project implementation stage to closely link the design stage up with the construction stage. This determines that the bidding work of metallurgical construction project must be finished in the preliminary design stage. And it usually adopts similar project estimation to estimate the cost of the proposed construction project. In order to accurately estimate the cost of the proposed metallurgical construction and help enterprises to make sensible bidding decisions in the case of shallow drawing, the article deeply researched on how to use similar project estimation method to accurately estimate the cost of a metallurgical project in the preliminary design stage.The research firstly did fuzzy optimization for similar projects in the metallurgical project cost database based on the fuzzy mathematics and interval number theory. The cost of proposed metallurgical project would then be estimated by means of BP neural network.To be specific, this paper firstly introduced how to create the metallurgical project cost database, elaborated the concept, principle and method of work breakdown structure and put forward the suggestions on the work breakdown structure for metallurgical construction project based on construction information classification system. Thus, combined with the cost of the index system, the metallurgical project cost database was created. This database worked as the platform to collect and analyze historical data for metallurgical construction units as well as the source of data for cost estimation model in this paper.Moreover, based on the discussion with experts in the field of metallurgicalengineering cost and the results of the questionnaire, the factors affects the three typesof unit engineering cost were identified using the Interpretative Structural Modeling.According to the fuzzy mathematics theory and interval number theory, these factorswere blurred to have the respective membership degree. The subjective interval weightsderived from Interval-Based AHP were confirmed by sending questionnaire again.Together with the objective weights by means of entropy value method, thecomprehensive interval weights were obtained. The optimal weights of all influencingfactors were confirmed using Multi-attribute Decision Model of Three-ParameterInterval Number, which further gave the closeness degree between similar projects andestimated projects in the database. In this way, fuzzy optimization of similarmetallurgical projects was realized.Finally, the fuzzy optimized historical data were taken as the training sample of BPneural network estimation model, which led to the strong approximation ability ofnetwork and realized the nonlinear mapping between influencing factors andconstruction quantities. Besides, this part set a background of production workshopcalcium carbide factory. Combined the process of metallurgical engineering costestimation, the predictive values were obtained through network output by usingMATLAB.This paper researched into the cost estimation problems of metallurgicalconstruction project under the CM approach. It was found that forecasting precisioncould meet the requirement of bidding work at the metallurgical engineeringpreliminary design stage after the comparative analysis of the estimation model output,which verified the scientificity and practicability of the metallurgical engineering costestimation model. This research not only provides a new way of thinking about biddingdecision-making and cost control for metallurgical construction project under CMapproach, but also offers theoretical support for practice.