Font Size: a A A

Research On Calculation Method Of Bid Cost Price Based On Artificial Neural Networks

Posted on:2020-09-21Degree:MasterType:Thesis
Country:ChinaCandidate:D R LuoFull Text:PDF
GTID:2392330590496644Subject:Architecture and civil engineering
Abstract/Summary:PDF Full Text Request
Since 2003,the model of valuation with bill quantity has been formally implemented in China,and the bidding model based on this model has been gradually promoted and continuously improved,which is in line with the international institution.However,there are still many problems in the practice of bidding in China.Among them,the harm of winning a bid at a low price is very serious.Some bidders participate in bidding at a price lower than the cost,and obtain improper profits through high-price claims or reducing the quality of the project.This behavior not only makes the bidder suffer huge losses,seriously wastes social resources,but also destroys the market environment of fair competition in the whole construction industry.Article 33 of BidTender Law of the People ? s Republic of China clearly stipulates that bidders shall not bid at a price lower than the cost.However,there is no uniform standard and method for whether the bid price is lower than the cost price.It can only refer to quotations of other bidders or tender control prices to make a rough judgement and make inquiries.It is inefficient and difficult to identify the unbalanced quotation of bidders.In order to solve the problem of lacking quantitative standard in the process of bidding evaluation,this paper analyses the difference between the average cost of society and the individual cost of enterprises,and puts forward that the cost price should be defined on the basis of the lowest cost of society.On this basis,this paper puts forward the method of calculating the cost price.Firstly,the cost of construction prediction model based on artificial neural network is established.Through the analysis of relevant literature,two kinds of index systems are constructed,which are project characteristics and project characteristics.The model is trained with 20 sets of historical project cost settlement data,and the cost of the proposed project is predicted,including the total quotation and the comprehensive unit price.Then,through the proportion and characteristics of labor,material,machinery and management cost in the comprehensive unit price,the comprehensive unit price is adjusted reasonably,and the cost price of the comprehensive unit price and the total quoted cost price are calculated.In bidding evaluation,the quotation in the list is compared with the calculated cost price item by item,which serves as a criterion to determine whether the bidding price is lower than the cost price,and provides a quantitative basis for the bid evaluation committee.Bid evaluation experts can make inquiries to bidders for list items below cost price,which is conducive to controlling the risk of unbalanced quotation of bidders,safeguarding the legitimate rights and interests of bidders,safeguarding the basic principles of fair,fair and open bidding market,and providing a good start for the smooth progress of the project.
Keywords/Search Tags:Bill-of-Quantity, Tendering and Bidding, Cost, Neural Networks
PDF Full Text Request
Related items