Research On The Application Of Historical Bid-Based Estimating For Bid Price Ceiling

Today, Bid Price Ceiling (BPC) of the project invested by the government in China is estimated and confirmed finally according to the budge of construction bid quantity list. The estimating has the following problems:the BPC is much higher than the average bid price, the basis of estimating update slowly and it is not explicit for the level of the BPC comparing to the average bid price. Historical Bid-Based Estimating (HBBE) is commonly used in other countries. The results of HBBE can represent the market level recently for it is based on the recent bid prices. The HBBE has the advantages of fast estimating speed and accuracy. So it can effectively solve the present problems of BPC estimating in China. The application of HBBE was researched in the paper through BPC estimating for the JTN project (the Nanping section of the Beijing-Taipei Expressway).The research was started from the generation and the existing problems of BPC. Then the significance of the research was stated and enlightenment was gained from the research status at home and abroad in the aspect of BPC’s definition, function, existing problems, problem countermeasures and estimate method, etc. According to the theory of bid quantity list and cost estimating, BPC estimating model, which is mainly used to estimate the list item prices, was constructed. Based on the model, HBBE method and historical data analysis theory, the research followed on the application of HBBE for BPC was carried out through the JTN project. Major list items and major cost-affect-factors of typical items were determined based on the combination of literatures, bidding process and messages of the JTN project. The historical bid prices were studied to orientate BPC. In terms of the reasonability and the feasibility, BPC’s basic adjustment coefficient of the JTN project was determined as the average magnitude difference between the highest bid prices and the contract prices from historical projects. Finally, unit BPCs of the typical items in the JTN project were estimated after a series of steps. Those steps included developing the database of historical projects, accessing the appropriate historical data, estimating unit cost prices of different typical items, multiplying the unit prices by the modified coefficient of BPC. The prices including cost and BPC estimated by HBBE method were compared with prices estimated by traditional method and contractor prices to evaluate the effect of HBBE. In HBBE process, different methods were applied to the estimating for different items. Weighted average was simply applied to minor item estimating. When lump sum item was estimated, relative costs must be found. Then concrete data analysis method was determined based on the relationship analysis. Support Vector Regression (SVR) was applied to major item estimating according to major cost-affect-factors. The accuracy of HBBE was not only better than the budge of construction bid quantity list but also better than the estimating used linear regression.The results of cost estimating and BPC estimating showed that accuracy of HBBE was obviously better than the traditional estimating. HBBE can be applied gradually in the bid stage of project construction in China. Individual estimated price may have a significant bias to the actual. The bias can be recognized and reduced by analogy technique to improve the accuracy and practicability of HBBE.