KR102451291B1 - E-bidding consulting system based on competitor prediction - Google Patents

Abstract

Disclosed is an e-bidding consulting system based on competitor prediction. The competitor prediction-based e-bidding consulting system according to the present invention applies a deep learning model to competitors expected to participate in the current bid announcement, and the estimated assessment rate (or expected bid) of each of the competitors for the current bid announcement A deep learning operation unit for extracting , and a predictive analysis unit for predicting the expected successful bid evaluation rate (or expected successful bid price) for the current bid announcement based on the expected evaluation rate of each of the competitors, and outputting the prediction result.

Description

E-BIDDING CONSULTING SYSTEM BASED ON COMPETITOR PREDICTION

The present invention predicts the expected successful bid assessment rate (or expected successful bid price) for the current bid announcement based on information on past bid announcements similar to the current bid announcement published in the electronic bidding system, and provides the prediction result to the customer (user) It relates to a competitor prediction-based e-bidding consulting system that can be provided.

In general, the process of selecting a project performer for a project ordered by the government, local governments, and various public institutions is performed by electronic bidding. Such e-bidding refers to a bidding method in which a person can participate in a bidding for procurement of goods or facility construction through the Internet without having to visit a bidding site and submit bidding documents. From September 2002, government agencies, local governments, and government-invested institutions have been obliged to disclose tender notices on the Nara Marketplace. In addition, a bid execution officer of a public institution may prepare a bid notice through a bid management system, electronically receive a bid form from a company, perform automatic opening, and then select a successful bidder.

A conventional e-bidding process will be described as follows. First, public institutions that wish to conduct business publish an electronic bidding notice. The notice includes the name of the notice, notice number, region, license, base amount, estimated price, assessment rate, expected price assessment range, bid closing date and time, and opening date and time. Here, the basic amount is the construction amount of the project investigated by the ordering party, and is the sum of the value-added tax, excluding the cost of government-related materials, and refers to the price that is the basis for calculating the 15 multiple preliminary prices for the planned price.

In addition, the estimated price is used to determine the eligibility criteria by excluding VAT and procurement fees based on the amount calculated in the budget before the expected price is determined. Minimum bid price ratio, created to prevent bad construction and specified in the tender notice. The evaluation rate is variously set to a value up to four decimal places within the range of -3 to +3.

In addition, the predetermined price assessment range refers to a range for calculating a plurality of preliminary prices based on the basic amount and determining the predetermined price. In addition, the plurality of preliminary prices are divided into 15 different prices prepared by the ordering authority based on the basic amount for the purpose of determining the expected price, and divided into sections in the estimated price assessment range according to the pre-determined cost calculation and the method of preparing the expected price. 15 are randomly generated. It is divided into an ordering party that discloses the plurality of preliminary prices and an ordering party that does not disclose the plurality of preliminary prices.

The bidders who want to bid for the project disclosed in the bid notice will participate in the bid by calculating the bid price, and the bidders participating in the bid will randomly select a plurality of two preliminary prices at the time of bidding. Then, the bid management system selects the predetermined price by arithmetic average of the plurality of preliminary prices selected the most among the plurality of preliminary prices selected by the bidders, that is, the number of the most drawn numbers.

Meanwhile, in the prior art, in order to understand a competitor's assessment rate, among the past bid announcements, bid announcements with the same conditions, such as construction type, ordering party, region, license, and scheduled price assessment range, are checked one by one to determine the competitor's There is a problem in that considerable time and effort are inevitably required because it is necessary to determine whether or not to bid and the assessment rate.

In particular, since the evaluation rates of competitors are formed differently depending on regions, licenses, ordering sites, etc., there is a problem in that it is very difficult to accurately grasp the bidding tendency of competitors. In addition, there is a problem in that the prediction accuracy of the competitor's evaluation rate is lowered according to the subjectivity of the e-bidding analyst who analyzes the competitor's evaluation rate.

As a related prior art, there is Republic of Korea Patent Publication No. 10-2007-0057584 (Title of the Invention: Electronic Bidding Point Calculation System and Method, Publication Date: 2007.06.07.).

The purpose of the present invention, that is, the problem to be solved, is to provide a competitor prediction-based e-bidding consulting system that can improve the accuracy and reliability in predicting the successful bid assessment rate (or successful bid price) for the current bid notice published in the e-bidding system. will provide

The problem to be solved by the present invention is not limited to the problem(s) mentioned above, and another problem(s) not mentioned will be clearly understood by those skilled in the art from the following description.

The above object, according to the present invention, by applying a deep learning model to the competitors expected to participate in the current bid announcement Deep to extract the estimated assessment rate (or expected bid price) of each of the competitors for the current bid announcement Competitor prediction, comprising: a running operation unit; and a predictive analysis unit for predicting the expected successful bid evaluation rate (or expected successful bid price) for the current bid announcement based on the expected evaluation rates of each of the competitors and outputting the prediction result is achieved by the e-bidding consulting system based.

Preferably, competitors expected to participate in the current bid announcement are selected based on information on past bid announcements similar to the current bid announcement, and the information on the past bid announcements and the It may further include a model generator for generating the deep learning model based on the actual assessment rate information (or actual bid information) of competitors.

Preferably, the predictive analysis unit may predict the expected successful bid assessment rate (or expected successful bid price) by applying the expected assessment rate (or expected bid price) of each of the competitors to a probability distribution function.

Preferably, the probability distribution function is generated by repeatedly simulating according to the successful bid assessment rate (or successful bid price) calculation rule applied in the electronic bidding system of the current bid announcement, or information on the actual successful bid assessment rate of the past bid announcements (or actual It may be a normal distribution function generated based on successful bid price information).

Preferably, the predictive analysis unit distributes the expected assessment rate (or expected bid price) of each of the competitors as an X-axis value of the probability distribution function, and the current bid for each of the expected assessment rates (or expected bids) of the competitors By calculating the probability of the expected successful bid evaluation rate (or the expected successful bid price) of the announcement, the top N (natural numbers) having a high probability may be estimated as the expected successful bid evaluation rate (or the expected successful bid price).

Preferably, the probability of becoming the expected successful bid assessment rate (or expected successful bid price) of the current bid announcement is, in the probability distribution function, the expected assessment rate (or expected bid price) for each of the expected assessment ratios (or expected bid prices) of the competitors. and an area value calculated by integrating a range between the nearest and other expected assessment rates (or expected bid prices) having a smaller value.

The present invention extracts the expected assessment rate (or estimated bid price) of the current bid announcement for each of the competitors expected to participate in the current bid announcement based on information on past bid announcements similar to the current bid announcement, and extracts the extracted competitor By predicting the expected successful bid evaluation rate (or expected successful bid price) of the current bid announcement based on each of the expected evaluation rates and providing them to customers (users) using the e-bidding consulting system of the present invention, the successful bid evaluation rate in the e-bidding system It can improve the accuracy and reliability of predictions for (or winning bids).

1 is a diagram illustrating a network configuration of a competitor prediction-based e-bidding consulting system according to an embodiment of the present invention.
2 is a block diagram illustrating a detailed configuration of a competitor prediction-based e-bidding consulting system according to an embodiment of the present invention.
3 is a diagram illustrating an example of calculating a probability (area value) for each value of the X-axis (expected evaluation rate of competitors) in a normal distribution function in order to predict an expected successful bid evaluation rate according to an embodiment of the present invention. .

Advantages and/or features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but will be embodied in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and common knowledge in the art to which the present invention pertains It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout.

In addition, the preferred embodiment of the present invention to be carried out below is already provided in each system function configuration in order to efficiently describe the technical components constituting the present invention, or a system function normally provided in the technical field to which the present invention pertains. The configuration is omitted as much as possible, and the functional configuration to be additionally provided for the present invention will be mainly described. If a person of ordinary skill in the art to which the present invention pertains, it will be possible to easily understand the functions of the conventionally used components among the functions omitted not shown below, and also the components omitted as described above. Relationships between elements and components added for purposes of the present invention will also be clearly understood.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating a network configuration of a competitor prediction-based e-bidding consulting system according to an embodiment of the present invention, and FIG. 2 is a diagram illustrating the detailed configuration of an e-bidding consulting system according to an embodiment of the present invention. It is a block diagram.

Referring to FIG. 1 , the e-bidding consulting system 120 based on competitor prediction according to an embodiment of the present invention includes a customer terminal 130 including a smart phone and a personal computer through a communication network including an Internet communication network and a mobile communication network. ) can be implemented in the form of an accessible server.

Competitor prediction-based e-bidding consulting system 120 according to an embodiment of the present invention provides information on actual assessment rates (or actual evaluation rates) of competitors who have participated in similar bidding announcements in the past with respect to the current bidding advertisements published in the e-bidding system 110 . Bid information) may predict the expected successful bid assessment rate (or expected successful bid price) for the current bid announcement, and provide the prediction result to the customer terminal 130 .

Here, the e-bidding system 110 typically includes an e-bidding system of the Nara Market operated by the Korean Public Procurement Service.

Meanwhile, in this specification, the term 'anticipated successful bid evaluation rate' is used to include the 'expected successful bid price'. In addition, in the e-bidding system, the 'success bid assessment rate' is a number typically displayed to four decimal places, and is a concept that matches the 'success bid price' expressed as an amount in a one-to-one correspondence. That is, if the 'auction assessment rate' is determined, the 'success bid price' matching it is also determined. In the same vein, competitors' 'actual assessment rate' and 'expected assessment rate' include 'actual bid price' and 'estimated bid price', respectively. Hereinafter, embodiments of the present invention will be described in detail using only the expression 'evaluation rate' to include 'a successful bid price' or a 'bid price'.

Referring to FIG. 2 , the competitor prediction-based e-bidding consulting system 120 according to an embodiment of the present invention includes a model generation unit 210 , a deep learning operation unit 220 , a prediction analysis unit 230 , and a control unit 240 . ) may be included.

The model generator 210 is deep learning based on information on past bid announcements similar to or related to the current bid announcement published in the electronic bidding system 110 and the actual assessment rate information of competitors participating in the past bid announcements. You can create a model (Deep Learning Model).

Specifically, the model generation unit 210 selects competitors expected or expected to participate in the current bid announcement based on information on past bid announcements similar to the current bid announcement, and then selects information on past bid announcements and the past. A deep learning model can be created based on the actual assessment rate information of competitors for bid announcements.

Alternatively, competitors expected to participate in the current bid announcement are selected based on information on past bid announcements similar to the current bid announcement, but may be selected through a configuration or process other than the model generator 210 .

Here, the 'deep learning model' predicts the expected assessment rate of competitors for past bid announcements based on information about past bid announcements while changing a plurality of algorithms and a plurality of data sets by combining them according to a predetermined rule. After doing this, by executing a myriad of feedback processes that compare this with competitors' actual assessment rates for past bid announcements, the combination of algorithms and data sets that perform best, that is, the best fit, can be selected.

The deep learning operation unit 220 may extract the expected assessment rate of each of the competitors for the current bid announcement by using the deep learning model generated by the model generation unit 210 as described above. That is, the deep learning operation unit 220 may apply the deep learning model to the competitors expected to participate in the current bid announcement to extract the expected assessment rate of each of the competitors for the current bid announcement.

Here, competitors expected to participate in the current bid announcement may search for past bid announcements similar to the current bid announcement, and may be selected based on information about the searched past bid announcements. In addition, past bid announcements similar to the current bid announcement may be provided, for example, as past bid announcements that meet the current bid announcement's work type, ordering party, region, license, and estimated price assessment range.

3 is a diagram illustrating an example of calculating a probability (area value) for each value of the X-axis (expected evaluation rate of competitors) in a normal distribution function in order to predict an expected successful bid evaluation rate according to an embodiment of the present invention. .

2 and 3, the predictive analysis unit 230 predicts the expected successful bid evaluation rate for the current bid announcement based on the expected evaluation rate of each of the competitors extracted by the deep learning operation unit 220 as above, and , the prediction result can be output.

Preferably, the predictive analysis unit 230 may predict the expected successful bid evaluation rate by applying the expected evaluation rate of each of the competitors to the pre-generated probability distribution function. Although the normal distribution function is used as the probability distribution function in predicting the expected successful bid assessment rate in the present embodiment, the present invention is not limited thereto, and other types of probability distribution functions may be used.

Specifically, the predictive analysis unit 230 distributes the estimated assessment rates of each of the competitors as the X-axis values of the normal distribution function as shown in FIG. 3, and the predicted successful bid assessments of the current bid announcement for each of the expected assessment rates of the competitors. It is possible to calculate the probability of becoming a bid rate and estimate the top N (natural numbers) with a high probability as the expected successful bid evaluation rate. In addition, the prediction analysis unit 230 may output a prediction result regarding the expected successful bid assessment rate and provide it to the customer terminal 130 .

For convenience of explanation, in FIG. 3, the expected ejaculation rates (x1, x2, x3, x4, x5, x6) of each of the six competitors are displayed as the X-axis values of the normal distribution function. It is common for the predicted assessment rates to be displayed at close intervals on the X-axis of the normal distribution function.

Here, the probability of becoming the expected evaluation rate for successful bids in the current bid notice is calculated by integrating the range between the expected evaluation rate and other expected evaluation rates that are closest and have a smaller value for each expected evaluation rate of competitors in a normal distribution function. It can correspond to the value of the area.

For example, in FIG. 3 , the area values (s1, s2, s3) for three three expected ejaculation rates (x2, x4, x6) out of six expected ejaculation rates (x1, x2, x3, x4, x5, x6) ) is s2 > s1 > s3, the predictive analysis unit 230 may predict that the expected assessment rate corresponding to s2 has the highest probability of being the expected successful bid assessment rate of the current bid notice.

In conclusion, the predictive analysis unit 230 calculates the area in the normal distribution function for each of the expected evaluation rates of competitors, estimates the top N items with high area values as the expected successful bid evaluation rates, and outputs the result to the customer terminal ( 130) can be provided.

As described above, in the present invention, the probability of being the expected successful bid evaluation rate of the current bid announcement calculated by the predictive analysis unit 230 means only the probability of the successful bid evaluation rate (or successful bid price) expected in the electronic bidding system 110 . Rather, it means a probability that also considers the distribution of assessment rates (expected assessment rates) to be submitted by competitors who are expected to participate in the current bid announcement. In general, the successful bid assessment rate (or successful bid price) announced in the e-bidding system 110 is often located in the central section of the normal distribution function, but in this central section, the assessment rates submitted by competitors are very densely and intensively distributed. In reality, it can be seen that the probability of winning a successful bid in relations with competitors is low.

In other words, the predictive analysis unit 230 of the present invention provides not only the probability of the successful bid evaluation rate (or successful bid price) expected in the electronic bidding system 110, but also the evaluation rate (expected evaluation rate) to be submitted by competitors expected to participate in the current bid announcement. ) in the current bid announcement in which the customer (user) using the e-bidding consulting system 120 of the present invention participated by predicting the expected successful bid assessment rate (or the expected successful bid price) for the current bid announcement by calculating the probability considering the distribution of This will increase your chances of actually winning the bid.

On the other hand, the normal distribution function applied by the predictive analysis unit 230 is a probability distribution function, and it can be generated by repeatedly simulating a number of times according to the successful bid assessment rate (or successful bid price) calculation rule applied in the electronic bidding system of the current bid notice. . Alternatively, the normal distribution function applied by the predictive analysis unit 230 may be generated based on actual successful bid assessment rate information of past bid announcements similar to the current bid announcement.

As shown in FIG. 2 , the control unit 240 includes a competitor prediction-based e-bidding consulting system 120 according to an embodiment of the present invention, that is, the aforementioned model generation unit 210 , extraction unit 220 , and prediction analysis unit. You can control the overall operation of 230, etc.

The components of the present invention described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, the components described in the embodiments may include a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), a programmable logic unit (PLU), a microprocessor, or an instruction. It may be implemented using one or more general purpose computers or special purpose computers, such as any other device capable of executing and responsive to . The processing element may execute an operating system (OS) and one or more software applications executing on the operating system. A processing element may also access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, a processing element is sometimes described as being used, but one of ordinary skill in the art will recognize that the processing element may include a plurality of processing elements and/or a plurality of types of processing elements. it can be seen that there is For example, the processing element may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.

Software may comprise a computer program, code, instructions, or combination of one or more of these, and may configure the processing element to operate as desired or instruct the processing element, independently or in combination, to operate as desired. The software and/or data may be embedded in any type of machine, component, physical device, virtual device, computer storage medium or device, or transmitted signal wave, to be interpreted by or to provide instructions or data to the processing element. It may be embodied permanently or temporarily. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The present invention is not limited to the above-described embodiments, and it is apparent to those skilled in the art that various modifications and variations can be made without departing from the spirit and scope of the present invention. Accordingly, it should be said that such modifications or variations shall fall within the scope of the claims of the present invention.

110: e-bidding system
120: e-bidding consulting system
130: customer terminal
210: model generation unit
220: deep learning operation unit
230: predictive analysis unit
240: control unit

Claims (6)

a deep learning operation unit for extracting the estimated assessment rate (or expected bid price) of each of the competitors with respect to the current bid announcement by applying a deep learning model to the competitors expected to participate in the current bid announcement; and
A predictive analysis unit for predicting the expected successful bid evaluation rate (or expected successful bid price) for the current bid announcement based on the expected evaluation rate of each of the competitors and outputting the prediction result;
Competitors expected to participate in the current bid announcement are selected based on information on past bid announcements similar to the current bid announcement,
A model generator for generating the deep learning model based on the information on the past bid announcements and the actual assessment rate information (or actual bid information) of the competitors for the past bid announcements; further comprising,
The predictive analysis unit,
The expected assessment rate (or expected bid price) of each of the competitors is applied to a probability distribution function to predict the expected successful bid assessment rate (or expected successful bid price),
The expected assessment rate (or expected bid price) of each of the competitors is distributed as the X-axis value of the probability distribution function, and the expected assessment rate (or expected bid price) of the current bid announcement for each of the expected assessment ratios (or expected bids) of the competitors ( or the predicted successful bid price), and the top N (natural numbers) with high probability are estimated as the expected successful bid assessment rate (or expected successful bid price),
The probability of being the expected successful bid assessment rate (or expected successful bid price) of the current bid notice is,
As the probability considering the distribution of the assessment rate (or bid price) expected to be submitted by the competitors in the current bid announcement, the expected assessment rate (or the expected assessment rate) of each of the competitors in the probability distribution function A competitor prediction-based e-bidding consulting system, characterized in that it corresponds to an area value calculated by integrating a range between an estimated bid) and another expected assessment rate (or expected bid) that is closest and has a smaller value.
delete delete The method of claim 1,
The probability distribution function is
Generated by repeatedly simulating according to the rules for calculating the successful bid assessment rate (or successful bid price) applied in the electronic bidding system of the current bid announcement, or generated based on the actual successful bid assessment rate information (or actual successful bid information) of the past bid announcements Competitor prediction-based e-bidding consulting system, characterized in that it is a normal distribution function. delete delete

Images