Vehicle price evaluation system and vehicle price evaluation method

Abstract

The future condition progression of the vehicle will be evaluated over a continuous time axis. [Solution] The vehicle price evaluation system 10 includes a vehicle operation information acquisition unit 21 and a price decline rate curve calculation unit 30. The vehicle operation information acquisition unit 21 acquires the number of years of operation of the used car to be evaluated. When the number of years of operation is input to the price decline rate curve calculation unit 30 calculates a price decline rate curve that predicts the buying and selling price of the used car to be evaluated at a time after the input. The price decline rate curve calculation unit 30 includes a trained model that has been trained using machine learning to output a price decline rate curve that is a function of the number of years of operation and the buying and selling price, using the number of years of operation of multiple used cars, the buying and selling prices of multiple used cars, and whether or not there are buying and selling contracts for multiple used cars.

Description

Technical Field

[0001] The present invention relates to a technique for evaluating the price of a vehicle.

Background Art

[0002] Patent Document 1 describes a vehicle inspection system. The vehicle inspection system of Patent Document 1 determines the vehicle state of the vehicle to be inspected by using the inspection result (first inspection data) by the vehicle itself of the vehicle to be inspected and the inspection result (second inspection data) by a third party (inspection device).

[0003] At this time, the vehicle inspection system described in Patent Document 1 determines the presence or absence of a failure part of the vehicle to be inspected.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the prior art such as Patent Document 1, only the presence or absence of a failure at the time of inspection can be known, and the future state transition of the vehicle cannot be determined on a continuous time axis. For this reason, in the prior art such as Patent Document 1, it has been impossible to determine at what timing after the inspection time the target vehicle (used vehicle) needs to be repaired or should be sold.

[0006] Therefore, an object of the present invention is to provide a technique for evaluating the future state transition of a vehicle on a continuous time axis.

Means for Solving the Problems

[0007] The vehicle price evaluation system of this invention comprises a vehicle operation information acquisition unit and a price depreciation rate curve calculation unit. The vehicle operation information acquisition unit acquires the number of years of operation of the used car to be evaluated. When the number of years of operation is input to the price depreciation rate curve calculation unit, it calculates a price depreciation rate curve that predicts the buying and selling price of the used car to be evaluated at a time after the input. The price depreciation rate curve calculation unit includes a trained model that has been trained to output a price depreciation rate curve which is a function of the number of years of operation and the buying and selling price, using the number of years of operation of multiple used cars, the buying and selling prices of multiple used cars, and whether or not there are buying and selling contracts for multiple used cars.

[0008] The price trends of used cars are closely related to the changes in their condition. For example, when a used car starts to experience many breakdowns, its price tends to drop significantly. By generating a trained model based on this relationship, and using this model as input for the number of years the vehicle has been in operation, it is possible to obtain a continuous curve showing the future price trends of the vehicle after the evaluation point, and to evaluate the future condition changes of the vehicle on a continuous time axis.

[0009] The vehicle price evaluation system of this invention comprises a vehicle ownership information acquisition unit and a price depreciation rate curve calculation unit. The vehicle ownership information acquisition unit acquires the number of years a used car to be evaluated has been owned by a single owner. When the number of years owned by a single owner is input to the price depreciation rate curve calculation unit, it calculates a price depreciation rate curve that predicts the buying and selling price of the used car to be evaluated at a time after the input. The price depreciation rate curve calculation unit includes a trained model that has been trained to output a price depreciation rate curve which is a function of the number of years owned by a single owner and the buying and selling price, using the number of years owned by a single owner of multiple used cars, the buying and selling prices of multiple used cars, and whether or not there are buying and selling contracts for multiple used cars.

[0010] The price trends of used cars are related to the number of years they have been owned by a single owner. For example, the period until a vehicle starts experiencing frequent breakdowns tends to be roughly the same as the number of years it has been owned by a single owner. And when a used car starts experiencing frequent breakdowns, its price drops significantly. By generating a trained model based on this relationship, and using this trained model as input for the number of years owned by a single owner, it is possible to obtain a continuous curve showing the price trends of the vehicle being evaluated from the evaluation point onward, making it possible to evaluate the future condition changes of the vehicle on a continuous time axis. [Effects of the Invention]

[0011] According to this invention, the future state changes of a vehicle can be evaluated on a continuous time axis. [Brief explanation of the drawing]

[0012] [Figure 1] Figure 1 is a functional block diagram of a vehicle price evaluation system according to the first embodiment of the present invention. [Figure 2] Figure 2 is a functional block diagram of a learning model according to the first embodiment of the present invention. [Figure 3] Figure 3 is a table showing an example of explanatory variables for the learning algorithm according to the first embodiment of the present invention. [Figure 4] Figure 4 is a graph showing an example of a price decline rate curve according to the first embodiment of the present invention. [Figure 5] Figure 5 is a functional block diagram showing another example of a vehicle price evaluation system according to the first embodiment of the present invention. [Figure 6] Figure 6 is a flowchart showing an example of a method for generating a trained model for generating a price decline rate curve in the vehicle price evaluation method according to the first embodiment of the present invention. [Figure 7] Figure 7 is a functional block diagram of a vehicle price evaluation system according to a second embodiment of the present invention. [Figure 8] Figure 8 is a table showing an example of explanatory variables for the learning algorithm according to the second embodiment of the present invention. [Figure 9]FIG. 9 is a functional block diagram of a vehicle price evaluation system according to a third embodiment of the present invention. [Figure 10] FIG. 10 is a graph showing an example of a price decline rate curve according to a third embodiment of the present invention. [Figure 11] FIG. 11 is a functional block diagram of a vehicle price evaluation system according to a fourth embodiment of the present invention. [Figure 12] FIG. 12 is a graph showing an example of a price decline rate curve according to a fourth embodiment of the present invention.

Embodiments for Carrying Out the Invention

[0013] [First Embodiment] A vehicle price evaluation system and a vehicle price evaluation method according to a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a functional block diagram of a vehicle price evaluation system according to a first embodiment of the present invention. FIG. 2 is a functional block diagram of a learning model according to a first embodiment of the present invention. FIG. 3 is a table showing an example of explanatory variables of a learning algorithm according to a first embodiment of the present invention. FIG. 4 is a graph showing an example of a price decline rate curve according to a first embodiment of the present invention.

[0014] As shown in FIG. 1, the vehicle price evaluation system 10 includes a vehicle operation information acquisition unit 21, a price decline rate curve calculation unit 30, and a high decline period calculation unit 40.

[0015] The vehicle operation information acquisition unit 21 acquires the operation years of a plurality of used vehicles. The operation years are the number of years from when the vehicle started operating as a new vehicle until the time of the sales transaction. For example, the operation start date and time registered as a new vehicle are input from the vehicle inspection certificate, and the elapsed time from the operation start date and time at the evaluation time point is used as the operation years. Alternatively, the production date and the date of shipment from the factory may be used as the operation start date and time, or the date when the vehicle was handed over to the user may be used as the operation start date and time. However, by using the registration date on the vehicle inspection certificate, a simple and unified reference value can be obtained. Also, if automated driving progresses and the engine operation time can be acquired as log data, the actual operation time of this engine may be used as the operation years. At this time, the vehicle operation information acquisition unit 21 acquires the identification information (vehicle ID) in association with each of the plurality of used vehicles.

[0016] Next, the model creation method of the price decline rate curve calculation unit 30 will be described. As shown in FIG. 2, the operation years of a plurality of used vehicles, their prices, and the transaction information are input as data, and a model of the transition of the transaction price with respect to the operation years is created. The price indicates the transaction price when the sale and purchase are concluded, and the asking price when the sale and purchase are not concluded.

[0017] For example, in the example shown in FIG. 3, for the used vehicle with vehicle ID1, it is acquired that the operation years are 5 years, and it is acquired that the sale and purchase were concluded at a sale and purchase price of 1 million yen. Similarly for vehicle IDs 2, 3, and 4, the operation years are acquired, and the sale and purchase price and the fact that the sale and purchase were concluded are acquired.

[0018] For vehicle ID5, the operation years are acquired, the sale and purchase price (in this case, the asking price), and the fact that the sale and purchase were not concluded are acquired.

[0019] Furthermore, regarding the new vehicle as vehicle ID0, the operation years of 0 are acquired, and the sale and purchase price (asking price) is acquired as the new vehicle price.

[0020] Since vehicle IDs 0 to 5 are information regarding different vehicles, information on various operation years can be collected in a short period without multiplying by the operation years in years.

[0021] The price decline rate curve calculation unit 30 is composed of a model that represents the relationship between the number of years of operation and the price decline rate curve based on the information in Figure 3. More specifically, the price decline rate curve calculation unit 30 takes the number of years of operation, the buying and selling price, and whether or not a transaction was completed as inputs, and uses a trained model that outputs a price decline rate curve which is a function of the number of years of operation and the buying and selling price (the price at which a transaction was completed) to calculate the price decline rate curve.

[0022] As shown in Figure 3, the trained model is generated by a learning algorithm 300 that uses a survival analysis method with operating years, trading price, and whether or not a trade was executed as explanatory variables. In this case, whether or not a trade was executed corresponds to whether or not an event occurred according to the survival analysis method. That is, if a trade was executed, the event value is set to "1", and if a trade was not executed, the event value is set to "0".

[0023] Then, as the learning algorithm 300, for example, the Cox proportional hazards model is used. In this Cox proportional hazards model, the model is trained using hazard ratios. That is, by learning the change in the degree to which a buy / sell transaction is likely to occur compared to the case where all event values ​​of the explanatory variables are 0, a trained model is generated.

[0024] The generation of this trained model allows for the creation of a price decline curve f1, which is a function of the number of years in operation and the buying and selling price, as shown in Figure 4. This price decline rate is expressed by the formula (new car price - used car price) / new car price, and represents the degree of price decline from the new car price.

[0025] Furthermore, this price decline curve f1 includes both vehicles that have been sold (used cars) and vehicles that have not been sold (used cars), including whether or not a sale was completed, as part of the learning process. Therefore, it can use a large number of explanatory variables for training. Consequently, the accuracy of the price decline curve f1 is improved compared to a price decline curve calculated using only vehicles that have been sold.

[0026] Furthermore, the price decline curve f1 does not represent the price at a single point in the future, but rather the price over a continuous period of time (operating time). Therefore, the price decline curve f1 can also represent the trend of price fluctuations (decline) as the operating time of a vehicle increases.

[0027] The price decline rate curve calculation unit 30 stores the trained models generated in this manner in advance. In this case, as will be described later, if the trained models are generated based on vehicle types, for example, the price decline rate curve calculation unit 30 may store a trained model for each vehicle type, or it may be a trained model that calculates the price decline rate curve for the input vehicle type using vehicle type information as an input parameter.

[0028] The vehicle price evaluation system 10X may include a vehicle-related information acquisition unit 22 in addition to the vehicle operation information acquisition unit 21. Figure 5 is a functional block diagram showing another example of a vehicle price evaluation system according to the first embodiment of the present invention. Specifically, as input to the price decline rate curve calculation unit 30, for example, in addition to the vehicle type and model number, classifications such as small cars, medium cars, large cars, trucks, etc., displacement classifications, automobile manufacturer classifications, automobile manufacturer nationality classifications, etc., can also be adopted.

[0029] The vehicle operation information acquisition unit 21 outputs vehicle operation information, including the vehicle ID and years of operation, to the price decline rate curve calculation unit 30. The vehicle-related information acquisition unit 22 outputs vehicle-related information, including at least one piece of information such as the vehicle ID, vehicle type, model number, engine displacement, and automobile manufacturer, to the price decline rate curve calculation unit 30.

[0030] The price depreciation curve calculation unit 30, upon receiving input of the operating years of the user's vehicle (for example, a used car owned by an individual, a used car owned by a car dealership, or a used car that an auction company is planning to put up for auction), uses a trained model to calculate a price depreciation curve appropriate for this vehicle. This price depreciation rate is the rate of decrease from the new car price, but it is also possible to calculate the rate of decrease over time from the present moment (when the user inputs the information), allowing the user to understand future depreciation rates. In other words, Users can accurately understand and evaluate price fluctuations (decreased prices) of their owned vehicles based on future time elapsed. Furthermore, users can understand and evaluate price changes over time, rather than just the price at a single point in time.

[0031] With the configuration and processing described so far, the vehicle price evaluation system 10 can evaluate the future changes in the vehicle's condition over a continuous time axis. However, by including the high / low period calculation unit 40, the vehicle price evaluation system 10 can achieve the following further effects.

[0032] The high / low period calculation unit 40 receives the relationship between the number of operating years and the price decline rate obtained from the price decline rate curve calculation unit 30 as a numerical input. The high / low period calculation unit 40 calculates the magnitude of the change in the price decline rate in response to the change in the number of operating years. In other words, the high / low period calculation unit 40 calculates the derivative of the price decline rate curve (the inflection point of the price decline rate curve).

[0033] The high / low rate period calculation unit 40 obtains the number of operating years in which the differential value is greater than or equal to a threshold from this calculation result. The high / low rate period calculation unit 40 outputs the obtained number of operating years. Alternatively, it may subtract the current number of operating years from the number of operating years in which the price decline rate will be high and output the remaining number of years until the price decline rate becomes high. The high / low rate period may be selected as the point in the future trend of the decline rate in which the decline rate will be highest, even if the differential value is not greater than or equal to a threshold.

[0034] By using this configuration and processing, users can easily identify periods when prices are likely to drop significantly. This allows users to, for example, determine when to sell to maximize their profits.

[0035] (Method for generating price decline curves) Figure 6 is a flowchart showing an example of a method for generating a trained model for generating a price decline rate curve in the vehicle price valuation method according to the first embodiment of the present invention. The specific details of each step in the flowchart shown in Figure 6 are described in the above-mentioned explanation of the configuration, so they are omitted unless further explanation is necessary.

[0036] The vehicle price evaluation system 10 acquires vehicle operation information, including the number of years in operation (S11). The vehicle price evaluation system 10 acquires vehicle sales information, including the sales price and whether or not a sales transaction has been completed (S12).

[0037] The vehicle price evaluation system 10 takes vehicle operation information and vehicle sales information as inputs (explanatory variables) and generates a trained model that outputs a price depreciation curve based on years of operation minus the closing price (S13).

[0038] Furthermore, when the vehicle price evaluation system 10 receives input of the operating years from the user, it generates and outputs a price depreciation rate curve corresponding to this input. In addition, based on this price depreciation rate curve, the vehicle price evaluation system 10 calculates and outputs periods of high depreciation rates.

[0039] While vehicle valuations are typically determined based on mileage, this invention uses years of operation. For example, in electric vehicles (EVs), batteries deteriorate over time due to free discharge, resulting in lower vehicle prices. Even in the case of gasoline vehicles, years of operation are used to account for deterioration that correlates with time, rather than just mileage, such as the aging of engine oil, paint due to UV rays, rubber gaskets, and idling time. Furthermore, in the configuration of this embodiment, the presence or absence of a completed sale is also considered when generating the trained model used for vehicle valuation. This allows for the incorporation of knowledge that is difficult to obtain using only years of operation, such as the expertise of used car buyers familiar with the buying and selling of used cars, into the trained model.

[0040] [Second Embodiment] A vehicle price evaluation system and a vehicle price evaluation method according to a second embodiment of the present invention will be described with reference to the figures. Figure 7 is a functional block diagram of the vehicle price evaluation system according to the second embodiment of the present invention. Figure 8 is a table showing an example of explanatory variables for the learning algorithm according to the second embodiment of the present invention.

[0041] As shown in Figure 7, the vehicle price evaluation system 10A according to the second embodiment differs from the vehicle price evaluation system 10 according to the first embodiment in that it includes a vehicle additional information acquisition unit 23 and in the trained model in the price decline rate curve calculation unit 30A. The other components of the vehicle price evaluation system 10A are the same as those of the vehicle price evaluation system 10, and the explanation of the similar parts will be omitted.

[0042] The vehicle price evaluation system 10A includes a vehicle additional information acquisition unit 23. The vehicle additional information acquisition unit 23 includes a vehicle driving information acquisition unit 231 and a vehicle maintenance information acquisition unit 232.

[0043] The vehicle driving information acquisition unit 231 acquires vehicle driving information, including the vehicle ID and mileage. The vehicle driving information acquisition unit 231 outputs the vehicle driving information to the price decline rate curve calculation unit 30A.

[0044] The vehicle maintenance information acquisition unit 232 acquires vehicle maintenance information, including the vehicle ID and the number of maintenance procedures performed to maintain the vehicle's condition, such as the number of oil changes. The vehicle maintenance information acquisition unit 232 outputs the vehicle maintenance information to the price decline rate curve calculation unit 30A.

[0045] The information acquired by the vehicle driving information acquisition unit 231 and the vehicle maintenance information acquisition unit 232 corresponds to the events in the additional information of the present invention.

[0046] The price decline rate curve calculation unit 30A generates a trained model that outputs a price decline rate curve, which is a function of the number of years of operation and the buying and selling price, using vehicle operation information (number of years of operation in the case of Figure 8), vehicle sales information (price and whether or not a sale was made in the case of Figure 8), vehicle driving information (mileage in the case of Figure 8), and vehicle maintenance information (number of oil changes in the case of Figure 8) as explanatory variables (inputs). The price decline rate curve calculation unit 30A calculates the price decline rate curve using this trained model.

[0047] With this configuration, the vehicle price evaluation system 10A can learn using additional vehicle information and calculate a price decline rate curve with higher accuracy. Furthermore, this allows the vehicle price evaluation system 10A to calculate with greater accuracy when prices will fall significantly.

[0048] Furthermore, vehicle driving information is not limited to mileage, but may include at least one of the following items related to vehicle driving: frequency of driving, driving area, frequency of high-load driving such as driving uphill, frequency of sudden steering, quality of road surface, frequency of high-speed driving, etc. These can be recorded by attaching sensors or the like to the vehicle. The vehicle maintenance information acquisition unit 232 can then acquire vehicle driving information by reading this record.

[0049] Furthermore, the vehicle maintenance information should include at least one of the following items related to vehicle maintenance, such as the number of times various maintenance and inspections are performed, the frequency of various maintenance and inspections (e.g., oil change frequency), the presence or absence of damage, repair history, the number of car washes, the frequency of car washes, and the vehicle storage environment. These can be recorded as appropriate. The vehicle maintenance information acquisition unit 232 can then acquire the vehicle maintenance information by reading this record.

[0050] [Third Embodiment] A vehicle price evaluation system and a vehicle price evaluation method according to a third embodiment of the present invention will be described with reference to the figures. Figure 9 is a functional block diagram of the vehicle price evaluation system according to the third embodiment of the present invention. Figure 10 is a graph showing an example of a price decline rate curve according to the third embodiment of the present invention.

[0051] As shown in Figure 9, the vehicle price evaluation system 10B according to the third embodiment differs from the vehicle price evaluation system 10A according to the second embodiment in that it includes a trained model in the price decline rate curve calculation unit 30B, processing in the high decline period calculation unit 40B, and a price decline factor identification unit 50B. The other components of the vehicle price evaluation system 10B are the same as those of the vehicle price evaluation system 10A, and the explanation of the similar parts will be omitted.

[0052] The vehicle price evaluation system 10B includes a price decline rate curve calculation unit 30B, a high decline period calculation unit 40B, and a price decline factor identification unit 50B.

[0053] The price decline rate curve calculation unit 30B classifies specific information in the vehicle supplementary information into multiple categories according to the conditions. The price decline rate curve calculation unit 30B uses vehicle operation information (in the case of Figure 8, years of operation), vehicle sales information (in the case of Figure 8, price and whether or not a sale was made), vehicle driving information (in the case of Figure 8, mileage), and vehicle maintenance information (in the case of Figure 8, number of oil changes) as explanatory variables (inputs) to generate a trained model for each classification that outputs a price decline rate curve, which is a function of years of operation and sales price.

[0054] For example, the price decline curve calculation unit 30B classifies the input (explanatory variable) based on whether the oil change frequency is once or more per year, or less than once per year. The price decline curve calculation unit 30B learns using input sets that satisfy the oil change frequency being once or more per year, and outputs the price decline curve f31 (see Figure 10). The price decline curve calculation unit 30B learns using input sets that satisfy the oil change frequency being less than once per year, and outputs the price decline curve f32 (see Figure 10).

[0055] The price peak / decline period calculation unit 40B detects inflection points (positions with large differential values) in the price decline rate curve for each classification. The price peak / decline period calculation unit 40B calculates the price peak / decline period for each classification from the detected inflection points.

[0056] For example, in the above case, the price fluctuation timing calculation unit 40B calculates the price fluctuation timings for cases where oil changes are performed at least once a year, based on the price decline rate curve f31. The price fluctuation timing calculation unit 40B also calculates the price fluctuation timings for cases where oil changes are performed less than once a year, based on the price decline rate curve f32.

[0057] The price peak / fall timing calculation unit 40B outputs the price peak / fall timing for each category to the price fall factor identification unit 50B.

[0058] The price decline factor identification unit 50B compares the timing of price increases and decreases for each classification to identify how the classified information is influencing the price decline. For example, if the timing of price increases and decreases differs between classifications, the price decline factor identification unit 50B identifies that the classified information is a major factor in the price decline. In the above case, if the timing of price increases and decreases occurs earlier for oil changes performed less than once a year compared to oil changes performed more than once a year, the price decline factor identification unit 50B identifies that the frequency of oil changes is a major factor in the price decline.

[0059] The price decline factor identification section 50B may identify the factor based on the difference in the declining price, or it may identify the factor based on the time difference and price difference between the high and low periods.

[0060] This allows the vehicle price evaluation system 10B to notify users of factors that could cause a price decrease, for example, enabling users to pay attention to how they use and maintain their vehicles.

[0061] In the configuration and processing described above, the vehicle price evaluation system 10B identified the factors causing the price decline by classifying one type of information (event) included in the additional information. However, the vehicle price evaluation system 10B may also classify multiple types of information to identify the factors causing the price decline. Furthermore, the vehicle price evaluation system 10B can also compare the price decline rate curves of multiple pieces of information to identify which piece of information has the greatest impact as a factor causing the price decline.

[0062] [Fourth Embodiment] A vehicle price evaluation system and a vehicle price evaluation method according to a fourth embodiment of the present invention will be described with reference to the figures. Figure 11 is a functional block diagram of the vehicle price evaluation system according to the fourth embodiment of the present invention. Figure 12 is a graph showing an example of a price decline rate curve according to the fourth embodiment of the present invention.

[0063] As shown in Figures 11 and 12, the vehicle price evaluation system 10C according to the fourth embodiment differs from the vehicle price evaluation system 10 according to the first embodiment in that it uses vehicle ownership information instead of vehicle operation information. The other configurations and processes of the vehicle price evaluation system 10C are the same as those of the vehicle price evaluation system 10, and explanations of the similar parts will be omitted.

[0064] The vehicle price evaluation system 10C includes a vehicle ownership information acquisition unit 24 and a price decline rate curve calculation unit 30C.

[0065] The vehicle ownership information acquisition unit 24 acquires the vehicle ID and the length of time the vehicle has been owned by a single owner as vehicle ownership information. The vehicle ownership information acquisition unit 24 outputs the vehicle ownership information to the price decline rate curve calculation unit 30C.

[0066] The price decline rate curve calculation unit 30C takes the length of the one-owner period, the buying and selling price, and whether or not a transaction was completed as inputs, and calculates the price decline rate curve f4, which is a function of the length of the one-owner period and the buying and selling price as shown in Figure 12, using a pre-trained model.

[0067] The one-owner period ends when the user who owned the vehicle decides to part with it. When a user decides to part with a vehicle, there are usually several reasons, one of which is likely deterioration or malfunction of the vehicle.

[0068] Therefore, by using the length of time the vehicle has been owned by a single owner instead of the number of years it has been in operation, the vehicle price evaluation system 10C can achieve the same effects as the vehicle price evaluation system 10.

[0069] In the above example, we explained how to create a trained model using data on operating hours, sales prices, and vehicle information, and how to use this trained model to predict future price declines based on operating hours. The trained model doesn't necessarily have to be used indefinitely; it can also be updated by further training it with price information and whether or not a sales contract was concluded for used cars whose price decline curves were calculated using this system. [Explanation of Symbols]

[0070] 10, 10A, 10B, 10C, 10X: Vehicle price evaluation system 21: Vehicle Operation Information Acquisition Unit 22: Vehicle-related information acquisition unit 23: Vehicle Additional Information Acquisition Unit 24: Vehicle Ownership Information Acquisition Department 30, 30A, 30B, 30C: Price decline rate curve calculation unit 40, 40B: High / low fall period calculation section 50B: Identifying Factors Contributing to Price Decline 231: Vehicle Driving Information Acquisition Unit 232: Vehicle Maintenance Information Acquisition Department 300: Learning Algorithms

Claims

1. A vehicle operation information acquisition unit that acquires the operating years of the used car to be evaluated, The system includes a price decline curve calculation unit that, upon inputting the aforementioned operating years, calculates a price decline curve predicting the buying and selling price of the used car being evaluated in the time period after the input. The price decline rate curve calculation unit includes a trained model that has been trained to output a price decline rate curve which is a function of the number of years of operation and the sales price, using the number of years of operation of multiple used cars, the sales prices of the multiple used cars, and whether or not there are sales contracts for the multiple used cars. Vehicle price evaluation system.

2. The system includes a high / low period calculation unit that detects the price inflection point in the price decline curve and calculates information about the inflection point that occurs after the time the operating years are input, and the high / low period of the price after the time the operating years are input. The vehicle price evaluation system according to claim 1.

3. Further input vehicle information, including the make and model of the used car being evaluated. The aforementioned trained model has been further trained to include the vehicle information. The vehicle price evaluation system according to claim 1.

4. The vehicle includes a vehicle supplementary information acquisition unit that acquires events that occurred in the aforementioned used car as vehicle supplementary information, The price decline rate curve calculation unit is: A trained model is used that takes the aforementioned vehicle additional information as one of its inputs and outputs the price decline rate curve. A vehicle price evaluation system according to any one of claims 1 to 3.

5. The aforementioned vehicle additional information includes a plurality of the aforementioned events, The price decline rate curve calculation unit is: A trained model is used that classifies the aforementioned events into multiple types and outputs the price decline rate curve for each classification. The vehicle price evaluation system according to claim 4.

6. A high / low price period calculation unit that detects inflection points in the price decline rate curve for each of the aforementioned classifications and calculates the high / low price periods for each of the aforementioned classifications from the inflection points, A price decline factor identification unit compares the periods of price increases and decreases for each of the aforementioned classifications and identifies events that cause price increases and decreases based on the comparison results, The vehicle price evaluation system according to claim 5, comprising:

7. The vehicle ownership information acquisition unit acquires the number of years the used car being evaluated has been owned by the same person, The system includes a price depreciation curve calculation unit that, upon inputting the number of years of single ownership, calculates a price depreciation curve that predicts the buying and selling price of the used car being evaluated in the time period after the input. The price decline rate curve calculation unit includes a trained model that has been trained to output a price decline rate curve which is a function of the number of years of single ownership and the sales price, using the number of years of single ownership of multiple used cars, the sales prices of the multiple used cars, and whether or not there is a sales contract for the multiple used cars. Vehicle price evaluation system.

8. The aforementioned trained model is trained using a survival analysis method. A vehicle price evaluation system according to any one of claims 1 to 7.

9. The vehicle operation information acquisition step involves obtaining the operating years of the used car to be evaluated, The system includes a step of calculating a price decline curve, in which, once the operating years are entered, it calculates a price decline curve that predicts the buying and selling price of the used car being evaluated in the time period after the time the operating years were entered. The step of calculating the price decline curve utilizes a pre-trained machine learning model that outputs a price decline curve, which is a function of the years of use and the sales price, using the years of use of multiple used cars, the sales prices of the multiple used cars, and whether or not there are sales contracts for the multiple used cars. Vehicle price valuation method.

10. The vehicle ownership information acquisition step involves obtaining the number of years the used car being evaluated has been owned by the same person, The system includes a step of calculating a price depreciation curve, in which, when the number of years of single ownership is entered, the system calculates a price depreciation curve that predicts the buying and selling price of the used car being evaluated in the time period after the time of input. The step of calculating the price decline curve utilizes a pre-trained machine learning model that outputs a price decline curve which is a function of the number of years of single ownership and the sales price, using the number of years of single ownership for multiple used cars, the sales prices of the multiple used cars, and whether or not there is a sales contract for the multiple used cars. Vehicle price valuation method.

11. The aforementioned trained model is trained using a survival analysis method. The vehicle price evaluation method according to claim 9 or claim 10.

Images