Research On Supervision For The Electronic Bidding Of Railway Construction Driven By Big Data

Railway is the main artery of the national economy,key infrastructure and major livelihood projects,which plays an important role in the economic and social development of our country.In recent years,the construction scale of railway engineering has been huge under the goal of building a strong country in transportation,and the construction bidding market has been active.The construction bidding of railway engineering included in the public resource trading market has adopted electronic bidding to varying degrees.Traditional bidding supervision methods are difficult to meet supervisory requirements,and there is an urgent need for innovation in supervisory methods to make supervision more effective.The massive data accumulated in the electronic bidding process has laid the data foundation,in the context of big data technology as an important means of innovative supervision methods,to study bidding supervision and early warning by means of big data technology,which is of great significance for realizing scientific,accurate and efficient intelligent supervision of electronic bidding for railway construction.The research object is supervision for the electronic bidding of railway construction in this dissertation.The early warning models are built from the perspective of bidding participants,to forewarn the illegal activities of bidding participants,on the basis of analyzing the supervision framework of railway construction electronic bidding driven by big data.Firstly,sort out the current situation of the research object,including the process and advantages of electronic bidding,links and key points of supervision,as well as the existing problems and development trends of supervision.Secondly,study the connotation,mode and content of supervision for railway construction electronic bidding driven by big data.Then,analyze the application path of big data technology in bidding supervision,and build a conceptual model of railway construction electronic bidding supervision driven by big data.Finally,according to the different participants involved in the bidding,the early warning models are built for the tendentious and restrictive behaviors of tenderers,from the aspect of the tenderer’s tendency,financial requirement,and limited maximum bid price,based on the network relationship and the principle of small probability;the early warning models are built for the unfair competition behaviors of bidders,from the aspect of colluding behaviors in bidding,bid price,production capacity and financial status,based on some methods such as community network structure and function fitting;the early warning models are built for the perform-duty behaviors of bid evaluation experts,from the aspect of abnormal scores,scoring reliability,and evaluation capability of bid evaluation experts,based on some methods such as the decision entropy and the logsig function,realizing the early warning of abnormal behaviors of tenderers,bidders and bid evaluation experts respectively,which can provide some effective reference for the activities of supervision and management for the bidding administrative supervision department.The innovations of this dissertation are the following:(1)The conceptual model of supervision for the electronic bidding of railway construction driven by big data is built,clarifying the basic principles of using the big data technology to supervise and manage the bidding activities.According to the big data processing process,from the aspects of data source,data collection,data processing,data storage,data analysis and data visualization,the application path of big data technology in electronic bidding supervision is studied.Then,on the basis of analyzing the supervision elements for the electronic bidding of railway construction,the conceptual model is built for the railway construction electronic bidding supervision driven by big data,from the dimensions of supervision subject,supervision objectives and principles,supervision objects and content,supervision mode,supervision system,and supervision technology.(2)The early warning models for the tendentious and restrictive behaviors of tenderers are built,realizing the supervision and early warning for the abnormal behaviors of tenderers.On the basis of analyzing the supervision content of tendentious and restrictive behaviors of tenderers,the tendency degree of tenderers is calculated from the perspective of the relationship network,and the concept of the dependence degree of bidders for tenderers is introduced,to build the early warning model of the tendentious behaviors of tenderers;the content of the restrictive behaviors of tenderers is analyzed,from the financial requirements and limited maximum bid price set by the tenderer,and the small probability method is used to analyze and calculate the early warning indicators thresholds of the operating income and working capital in the financial requirements,as well as the early warning indicators thresholds of the fund descent rate and saving rate of the limited maximum bid price,to forewarn the unreasonable restrictive conditions set by the tenderer.(3)The early warning models for unfair competition behaviors of bidders are built,realizing the supervision and early warning for the unfair competition behaviors of bidders.On the basis of analyzing the pattern of manifestations for unfair competition behaviors of bidders,the community network structure and function fitting method are used to analyze and forewarn bidders’ colluding behaviors in bidding;the bid price that too low or too high is monitored through of the sequencing of bidders’ bid price,the unbalanced bidding is analyzed under the condition of correcting the deviation value,and the similar bid price among bidders is identified based on the vector angle cosine value;the C-D production function method is used to build the bidder’s production capacity model,and the bidder’s financial performance evaluation model is built based on the efficiency coefficient method,which are used to supervise and forewarn the production capacity and financial status of bidders,in order to analyze the contractual capacity of bidders.(4)The early warning models for the quality of expert scores and evaluation ability of bid evaluation experts are built,realizing the supervision and early warning for the perform-duty behaviors of bid evaluation experts.On the basis of analyzing the perform-duty behaviors of bid evaluation experts,the quality of expert scores and evaluation ability of bid evaluation experts are emphatically studied,and the quality of expert scores is analyzed from two aspects of abnormal scores and scoring reliability.Firstly,the mathematical statistics methods are used to detect abnormal scores of bid evaluation experts,the controversial,customary,and tendentious factors that affect scoring deviation are analyzed,and the early warning path for abnormal scores is designed;secondly,the decision entropy method is used to build a score reliability model,to supervise and forewarn the score reliability of bid evaluation experts;then,the evaluation ability model is built based on the logsig function,through the results of the scores reliability analysis and the number of evaluation for the bidding experts,to supervise and forewarn the evaluation ability of bid evaluation experts.