JPH0335352A - On-line price deciding system - Google Patents

Abstract

PURPOSE:To automatically decide the proper prices in an on-line system by performing a fuzzy inference based on the prince plans received from a terminal equipment. CONSTITUTION:A host device A is prepared together with plural terminal equipments 1 connected in on-line to the device A. A consumer, a buyer, etc., operates the equipments 1. Thus the buyer transmits a proper price plan to the device A for a specific published commodity. The device A receives the price plans from plural equipments 1 and carries out a fuzzy inference based on these price plans. Then the device A decides the final price in consideration of the distribution state of the shown price plans, the cost prices, etc. In this fuzzy inference the price deciding thought (algorithm) of an expert is not turned into the numerical value but turned into a rule in a human expression. As a result, a proper price can be automatically decided in an on-line system.

Description

Detailed Description of the Invention (a) Industrial Application Field This invention relates to an online pricing system that uses data communication to determine prices and conduct auction sales. (b) Background of the Invention Products for which it is difficult for sellers to determine prices. In the case of products whose prices are easily determined by market prices, auctions and bidding are often used as a means of price determination and sales. Furthermore, when a company goes public or sells increased capital, a system is used in which multiple investors submit prices and a reasonable selling price is determined from these prices.

By the way, in this type of sales method 1 price determination method, it takes skill to decide where to end the auction, which price range is appropriate, etc., and it is difficult to do it by an electronic control device due to the algorithm. Traditionally, this decision was made by experienced experts because it was extremely difficult to make.

However, now that online sales systems such as Teletex are in practical use, it is hoped that the system in which multiple buyers meet at the same location and sell by expert auctioneers will be improved, automated, and moved online. It is rare.

In view of the above-mentioned current situation, it is an object of the present invention to provide an online price determination system that solves the above-mentioned problems by determining the price by using fuzzy inference on the presented price.

(C) Means for Solving the Problems This invention comprises a host device and a plurality of terminals connected online to the host device.
The present invention is characterized in that a price determining means is provided for receiving a price room transmitted from a terminal regarding a specific product, executing fuzzy inference based on this price room, and determining the price of this specific product.

(d) Effects of the Invention In the online price determination system of the present invention, buyers such as consumers and buyers operate terminals. The buyer sends to the host device the price room that is considered reasonable for the particular advertised product.

After the host device receives price information from multiple terminals,
Perform fuzzy inference based on these price rooms. Fuzzy inference determines a final price by inference taking into consideration the distribution state of the presented price room, cost price, etc. Fuzzy inference allows inferences to be made by formulating rules in human expression without converting experts' pricing ideas (algorithms) into numbers, making it possible to automatically determine prices using electronic devices such as computers. . Additionally, since there is no need to gather at a specific venue to bid, anyone who owns a computer or other terminal may be able to participate, allowing consumers to directly participate in price determination. This will allow you to determine a more reasonable price.

(e) Embodiment FIG. 1 is a block diagram showing the configuration of an online price determination system that is an embodiment of the present invention.

A plurality of terminals 1 are connected to the data communication network 2, and are connected to an online transaction processing system 3 through the data communication network 2. The online transaction processing system 3 includes a fuzzy computer 4. A console terminal 6 is connected. An audit subsystem 5 is also connected to the data communication network 2 and also to the online transaction processing system 3. Online transaction processing system 3 Fujii computer 4. Audit subsystem 5 and console terminal 6 constitute host device A. In this system, a buyer operating a terminal device 1 sends a message containing the available purchase price (limit price) and the planned purchase quantity for a specific product, and the host device A, which receives this message, determines the price using fuzzy reasoning. It is a system that

The functions of each device making up this system are as follows.

Terminal L: A device operated by auction participants at home, etc. It receives guidance information such as product presentation from the host device, and accepts inputs such as the limit price (price room) for participating in the auction. Each terminal has a separate ID code, and the host device uses this code to identify the terminal (participant).

Data communication network 2: A device configuring a permanent or temporary communication path between the terminal device 1 and the host device A. Examples include a telephone switched network, a digital switched network, a complex switched network,
There are value-added communication networks, etc.

In this embodiment, it is assumed that the functions of the network interface section (O3I layer 5 and below) on the host device A side are also included.

Online transaction processing system 3: Processes information manually input from terminal 1, executes auction,
File computer 4 data such as sales price before correction.
Fuzzy computer 4: A device that receives antecedent variables from the online transaction processing system 3 and performs fuzzy inference by applying built-in fuzzy inference rules Audit subsystem 5: Bidding A device that objectively analyzes whether or not the dropout is being operated fairly and efficiently, that is, providing a high level of service to participants.

Data of the online transaction processing system 3 is accessed online using a built-in audit identification code, and an audit trail is collected and analyzed offline. The audit report of the analysis results is transferred to a file in the online transaction processing system 3 so that auction participants can access it and check its "fairness."

Console terminal 6: Terminal for operator intervention. The fuzzy computer 4. has multi-windows and uses the relay function of the online transaction processing system 3. It also has the function of inputting and outputting data to and from the audit subsystem 5.

Figure 2 shows online transaction processing system 3
FIG. 2 is a block diagram showing the configuration of FIG. Data communication network 2. Fuzzy computer 4゜Communication control management processor 33 connected to console terminal 6
A plurality of 0LTP subsystems 30, data management files 34, and network management files 35, each corresponding to one product department, are connected to. Also,
The 0LTP subsystem 30 includes an 0LTP processor 31.
It is composed of 0LTP files 32. The functions of these devices are as follows.

0LTP processor 31: This is a well-known online transaction processing processor, and the 0LTP file 3
Along with 2, it has a dual system.

It also has the function of relaying data input/output between the 0LTP file 32 and other processors.

0LTP file 32: This is a well-known online transaction processing file, and as shown in Figure 7, it contains "quantity data by price range", "auction procedure parameter data" for each item, and "item data" and "price data" for each participating sound. "Quantity data" is stored.

Communication control data management processor 33: each 0TL
Serves as a higher-level/preprocessing processor for the P processor, and handles authentication, encryption, and decryption of interface messages with the network, communication with other processors, etc.
．． 7) Execute processing including the function of [. It also handles network diagnosis, reorganization, file recovery, etc. between each processor.

Data management file 34: As shown in Figure 7, the file mainly stores data related to authentication and confidentiality.Network management file 35: 7th file, as shown in Figure 7, the file mainly stores data related to network. The figure is a block diagram showing the configuration of a fuzzy computer (fuzzy inference system) 4. A fuzzy inference processor 41, an inference file 44 (inference control data file 44a, inference case data file 44b), and a man-machine interface processor 45 are connected to an inference management processor 43 connected to the online transaction processing system 3. A fuzzy memory 42 is connected to the fuzzy inference processor 41 .

Fuzzy rules are stored in this fuzzy memory 42 (42a). Further, a development and operation tool file 46 is connected to the man-machine interface processor 45. The functions of these devices are as follows.

Fuzzy inference processor 41: Inference management processor 43
It receives the designation of the rule to be applied and the antecedent variable from , executes fuzzy inference with reference to the fuzzy rules and membership functions stored in the fuzzy memory 42 , and returns the conclusion to the inference management processor 43 . A co-processor architecture is adopted for the inference management processor 43. Another function is to control the storage of rule functions in fuzzy memory.

Fuzzy memory 42: A memory that stores fuzzy rules and membership functions. It has a constant area and a variable area.

Inference management processor 43: Samples the reference area of the TLP file 32 and compares it with the parameters of the file 44 to determine whether to start fuzzy inference. 0 when determining startup
Data is extracted from the TLP file 32, the antecedent variables of the rules are calculated, and the data is transferred to the fuzzy inference processor along with the designation of the ruleset.

File 44 (inference control data file 44a and inference case data file 44b): As shown in FIG.
Inference management data and inference results adopted in the past are stored.

Man-machine interface processor 45: A processor that monitors the execution status of fuzzy inference and supports development and testing of fuzzy rules. Human output is performed through the console terminal 6 window.

Development/operation tool file 46: File that stores development support tools such as editors with built-in syntax such as fuzzy rules, debuggers, analyzers, and tracers. Figure 6 shows the online transaction processing system 3.
3 is a flowchart showing the operation of the 0LTP processor 31 of FIG. First, auction participation guidelines (product contents, access method, etc.) are announced to each terminal 1 via the data communication network 2 (nl). Next, intermediate aggregation time (n2),
It waits until a reference price range announcement command is issued (n6), an auction gathering command is issued (n8), or a message from an auction participant (limit price, quantity, etc.) is received (nil).

Interim aggregation is an operation that executes fuzzy inference based on the messages received so far and announces the final price or reference price range. This operation is periodically executed by a timer.

When the timer announces that it is time for interim aggregation,
Proceed to n2 → n3, create a price-number of sales table from the messages received so far, calculate the average marginal price and each percentile price (n3), perform simulation calculations (n4), and communication control data management processor 33
is transferred to the fuzzy computer 4 via (n5).

Here, the simulation calculation is a calculation to find the maximum total sales price or the maximum total added value price. In other words, the maximum value of selling price (added value) x number of units sold is calculated.

The fuzzy computer 4 infers a reference price range based on this data and notifies the communication control data management processor 33 of the result. The communication control data management processor 33 calculates the total sales amount or the total added value from the reference price range received by the fuzzy computer 4, and when this reaches a predetermined target value, sends an auction termination command to this 0LTP processor. Along with outputting
The fuzzy computer 4 is instructed to infer a fixed price.

If the target value is not reached, the reference price range is output to the 0LTP processor.

When this 0LTP processor receives a reference price range publication command, it proceeds to steps n6→n7. At n7, the reference price range is read out from the communication control data management processor 33, transmitted to each terminal 1, and announced, and the auction continues.

In addition, when an auction termination command is received, the process proceeds from n8 to n9, and the notification that the auction has ended and the final price are announced to each terminal (n9), and individual transactions are processed from each terminal (nl).
O) Finish the movement. For individual transaction processing, payment for one shipment is made by a logistics system (not shown) that is transmitted over the same network.

On the other hand, if the auction participation message is received, nil→n12
In step n12, it is checked whether the format and contents of the message satisfy necessary conditions. The message includes the item, quantity, limit price (the price at which the purchase contract will be automatically entered when the confirmed price falls below this value), and if the limit price falls below the reference price range. If the content does not meet the conditions, it will be removed. If the message is judged to be valid (n l 3), the mask file (OL
TP file) and return to n2.

FIGS. 4 and 5 show the fuzzy rules and membership functions used for the above fixed price inference. The reference price range is calculated manually from the online transaction processing system 3, A: Cost fixed costs, etc. B: Average price/maximum sales price C: Price distribution width ) A correction factor for the maximum price is inferred. The fuzzy rule for this inference is, for example, configured as follows: ``If the maximum total sales price is close to the average price and the mass production effect is high, the price is corrected to be slightly lower than the maximum total sales price.''

The reference price range is inferred as a range in which the fixed price converges, and is inferred using almost the same fuzzy rules and membership functions as in the above fixed price inference.

(f) Effects of the Invention As described above, the online price determination system of this invention determines prices by fuzzy reasoning based on the price proposal sent from the terminal, so that the online price determination system automatically determines prices online. be able to determine a reasonable price. Since buyers can participate in auctions online if they have a terminal, many buyers can participate from anywhere, making pricing and sales extremely efficient.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an online price determination system that is an embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of an online transaction processing system of the online price determination system, and FIG. 3 is a block diagram showing the configuration of an online transaction processing system of the online price determination system. FIGS. 4 and 5 are block diagrams showing the configuration of the fuzzy computer of the price determination system; FIGS. 4 and 5 are diagrams showing the fuzzy rules and membership functions used in fuzzy inference executed by the fuzzy computer; FIG. 7 is a flowchart showing the operation of the communication data management processor of the action processing system, and is a diagram showing data stored in the files and processes included in the online price determination system. A-host device, l-terminal device, 4-fuzzy computer

Claims (1)

[Claims] (1) An online price determination system consisting of a host device and a plurality of terminals connected online to the host device, wherein the host device receives a price proposal sent from the terminals regarding a specific product; An online price determination system characterized by being provided with a price determination means that executes fuzzy inference based on this price proposal and determines the price of this specific product.