Vehicle maintenance prediction system, vehicle maintenance prediction method, and program

Abstract

To predict appropriate replacement time of a consumable item of a vehicle.SOLUTION: A vehicle maintenance prediction system 1 which predicts replacement time of a consumable item being an inspection object of vehicle maintenance, acquires sensor data for learning of the consumable item and a wear degree of the consumable item, generates a learning model by learning association data in which the acquired sensor data for learning of the consumable item and the wear degree of the consumable item are associated, acquires prescribed sensor data of the consumable item, estimates the wear degree of the consumable item corresponding to the acquired prescribed sensor data of the consumable item by use of the generated learning model, and predicts the replacement time of the consumable item.SELECTED DRAWING: Figure 1

Description

Technical Field

[0004] , , , , , ,

[0001] The present invention relates to an effective technique for predicting the replacement timing of consumables to be inspected in vehicle maintenance.

Background Art

[0002] In recent years, in the logistics industry, technologies related to vehicle maintenance of vehicles such as trucks used for delivery have attracted attention. For example, in Patent Document 1, based on the maintenance data of other vehicles and statistical data, by calculating an exchange recommendation degree indicating the degree to which parts used in the vehicle to be diagnosed should be promoted for replacement, for parts that deteriorate over time in the vehicle, a technique for presenting an appropriate replacement timing according to the situation of each vehicle is disclosed. In addition, in Patent Document 2, by utilizing telematics to appropriately grasp the deterioration state of vehicle consumables, formulating an appropriate maintenance plan, evaluating the maintenance plan based on the maintenance results according to the plan, and providing feedback to the next maintenance plan, a technique for supporting operations related to preventive maintenance of vehicles is disclosed.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] For vehicles such as trucks used for delivery, the vehicle administrator needs to manage the vehicle consumables. In addition, since the replacement timing of vehicle consumables varies depending on the driving characteristics of the driver, the loading state of the cargo, the driving environment, conditions (long distance / route delivery), etc., it is required to predict an appropriate replacement timing by AI (Artificial Intelligence). However, the technologies described in Patent Documents 1 and 2 cannot predict the appropriate replacement time using AI.

[0005] The inventors recognized the need for a vehicle maintenance prediction system, a vehicle maintenance prediction method, and a program that can use AI to predict the appropriate replacement timing for vehicle consumables.

[0006] Therefore, the present invention aims to provide a vehicle maintenance prediction system, a vehicle maintenance prediction method, and a program that can predict the appropriate replacement timing for consumable parts of a vehicle. [Means for solving the problem]

[0007] The present invention relates to a vehicle maintenance prediction system for predicting the replacement timing of consumables subject to inspection during vehicle maintenance, comprising: a learning data acquisition unit that acquires sensor data of the consumables for learning purposes and the degree of wear of the consumables; a generation unit that generates a learning model by learning associated data that links the acquired sensor data of the consumables for learning purposes and the degree of wear of the consumables; a prediction data acquisition unit that acquires predetermined sensor data of the consumables; and a prediction unit that uses the generated learning model to estimate the degree of wear of the consumables in relation to the acquired predetermined sensor data of the consumables and predicts the replacement timing of the consumables. An estimation unit that estimates the work time according to the replacement time and consumables; an output unit that outputs the predicted replacement time of the consumables, the estimated work time, and inventory information of the consumables; and a replacement date and time notification unit that, according to the outputted replacement time of the consumables, the work time, the inventory information of the consumables, and the work schedule, checks whether the vehicle maintenance can be performed at the time of replacement, makes a provisional reservation for the vehicle maintenance if there is an available date and time and inventory, and notifies the customer of the provisional reservation date and time as the replacement date and time of the consumables. We provide a vehicle maintenance prediction system equipped with the following features.

[0008] According to the present invention, a vehicle maintenance prediction system for predicting the replacement timing of consumables subject to inspection during vehicle maintenance acquires sensor data of the consumables for learning purposes and the degree of wear of the consumables, generates a learning model by learning associated data that links the acquired sensor data of the consumables for learning purposes and the degree of wear of the consumables, acquires predetermined sensor data of the consumables, and uses the generated learning model to estimate the degree of wear of the consumables in relation to the acquired predetermined sensor data of the consumables, thereby predicting the replacement timing of the consumables. [Effects of the Invention]

[0009] According to the present invention, it becomes possible to predict the appropriate replacement time for consumable parts of a vehicle. [Brief explanation of the drawing]

[0010] [Figure 1] This is a diagram illustrating the overview of the vehicle maintenance prediction system 1. [Figure 2] This diagram shows the functional configuration of the vehicle maintenance prediction system 1. [Figure 3] This diagram shows the learning model generation process performed by the vehicle maintenance prediction system 1. [Figure 4] This diagram schematically shows an example of training data for a belt used in learning. [Figure 5] This diagram schematically shows an example of training data for tires used in learning. [Figure 6] This diagram shows the replacement timing prediction process performed by the vehicle maintenance prediction system 1. [Figure 7] This diagram schematically shows an example of a condition matching notification screen. [Figure 8] This diagram schematically illustrates the contents of the belt prediction results. [Figure 9] This diagram schematically illustrates the content of the tire prediction results. [Figure 10] This diagram shows the maintenance reservation process performed by the vehicle maintenance prediction system 1. [Figure 11] This diagram schematically illustrates an example of a delay notification. [Figure 12] This diagram schematically shows an example of a reservation confirmation. [Figure 13] This diagram schematically shows an example of quotation data. [Figure 14] This diagram shows the vehicle maintenance process performed by the vehicle maintenance prediction system 1. [Modes for carrying out the invention]

[0011] Hereinafter, with reference to the accompanying drawings, embodiments for implementing the present invention (hereinafter referred to as embodiments) will be described in detail. In the following drawings, the same elements are denoted by the same numbers or symbols throughout the description of the embodiments.

[0012] [Basic Concepts / Basic Configuration] FIG. 1 is a diagram for explaining the outline of the vehicle maintenance prediction system 1. The vehicle maintenance prediction system 1 is a system for predicting the replacement timing of consumables of the vehicle 5, including at least a maintenance prediction device 10 for predicting the replacement timing of consumables of the vehicle 5 and a maintenance reservation device 20 for executing various processes at a work site for vehicle maintenance.

[0013] In this embodiment, as a premise, a predetermined sensor described later is attached to the vehicle 5, and this predetermined sensor detects predetermined sensor data.

[0014] The processing steps for the case where the vehicle maintenance prediction system 1 predicts the replacement timing of consumables to be inspected for vehicle maintenance of the vehicle 5 will be described based on FIG. 1. First, the maintenance prediction device 10 acquires sensor data of consumables for learning and the wear condition of this consumable (step S1). Consumables are items to be inspected for vehicle maintenance. For example, belts such as V-belts and fan belts, tires, lights such as light bulbs and LEDs (Light Emitting Diodes), support parts for supporting the truck bed such as towing fittings, and compressors for the freezer. Sensor data is, for example, vibration sensor data, distance sensor data (detection by acoustic sensors or optical sensors), image sensor data, illuminance sensor data (detection results by voltmeters, ammeters, etc. provided on the circuit can also be used), and pressure sensor data. The wear condition is, for example, the looseness of the belt, the wear condition of the tire groove, the wear condition of the light, the deterioration condition of the support part, and the refrigerant filling amount. The maintenance prediction device 10 acquires sensor data detected by various sensors attached to vehicle 5 or other vehicles, along with the degree of wear of various consumables determined by maintenance workers while performing maintenance on vehicle 5 or other vehicles, or sensor data of various consumables stored by an external server or the like, along with the degree of wear of these consumables, as sensor data for consumables to be used for learning.

[0015] The maintenance prediction device 10 generates a learning model by learning associated data that links acquired sensor data of consumables for learning with the degree of wear of the consumables (step S2). Examples of learning methods include machine learning using supervised learning, unsupervised learning, reinforcement learning, etc., and deep learning using convolutional neural networks, recurrent neural networks, long- and short-term memory, etc. In this embodiment, machine learning using supervised learning will be explained as an example. The maintenance prediction device 10 generates associated data by associating the acquired sensor data for each consumable item with the degree of wear of each consumable item. The maintenance prediction device 10 generates associated data by associating vibration sensor data with the degree of belt looseness, by associating distance sensor data or image sensor data with the degree of tire tread wear, by associating illuminance sensor data with the degree of light wear, by associating vibration sensor data with the degree of deterioration of support parts, and by associating pressure sensor data with the amount of refrigerant charged. The reason for using pressure sensor data for the compressor in the freezer is that pressure drop due to refrigerant deterioration, leakage due to aging, and refrigerant leakage due to holes in the compressor and / or heat exchanger and its pathways (for example, caused by impacts from pebbles, etc.) are suspected. The maintenance prediction device 10 learns associated data for each consumable and generates a learning model based on the learning results. The maintenance prediction device 10 generates a learning model for each of the consumables mentioned above. At this time, the maintenance prediction device 10 learns sensor data, the degree of wear, and the replacement timing of the consumable, and generates a learning model based on the learning results. The replacement timing refers to the period from several weeks to several days before the day on which it is predicted that the consumable will reach a wear value at which it will malfunction. Figure 1 shows examples of training data used for learning, specifically belts and tires.

[0016] The maintenance prediction device 10 acquires predetermined sensor data for consumables (step S3). The predetermined sensor data is one or more combinations of the sensor data that generated the associated data described above. The maintenance prediction device 10 acquires predetermined sensor data for consumables from sensors attached to the vehicle 5. In Figure 1, vibration sensor data from the belt and distance sensor data from the tires are shown as examples of consumables.

[0017] The maintenance prediction device 10 uses the generated learning model to estimate the degree of wear of a consumable item based on predetermined sensor data of that consumable item, and predicts when the consumable item should be replaced (step S4). The maintenance prediction device 10 uses a belt learning model to estimate the degree of belt wear based on acquired belt vibration sensor data. The maintenance prediction device 10 also uses a tire learning model to estimate the degree of tire tread wear based on acquired tire distance sensor data. The maintenance prediction device 10 uses a belt learning model to predict the timing of belt replacement based on the estimated wear of the belt. The maintenance prediction device 10 also uses a tire learning model to predict the timing of tire replacement based on the estimated wear of the tire treads.

[0018] The maintenance prediction device 10 outputs the predicted replacement timing for consumables to the maintenance reservation device 20, and the maintenance reservation device 20 then makes a reservation for maintenance work on the consumables based on this prediction result.

[0019] According to this vehicle maintenance prediction system 1, it is possible to predict the appropriate replacement timing for vehicle consumables.

[0020] [Functional Configuration] Based on Figure 2, the functional configuration of the vehicle maintenance prediction system 1 will be explained. The vehicle maintenance prediction system 1 is a system comprising a maintenance prediction device 10 and a maintenance reservation device 20, each connected via a network 3 such as a public telephone network to enable data communication. The vehicle maintenance prediction system 1 may include a driver's terminal held by the vehicle driver, a manager's terminal held by the vehicle manager, a maintenance worker's terminal held by the maintenance worker, and other terminals and devices. In this case, the vehicle maintenance prediction system 1 executes the processing described later using one or more combinations of the maintenance prediction device 10, the maintenance reservation device 20, and other terminals and devices.

[0021] The maintenance prediction device 10 is a computer or personal computer with server functionality that predicts when vehicle consumables need to be replaced. The maintenance prediction device 10 may be implemented using, for example, one computer, or it may be implemented using multiple computers, such as a cloud computer. In this specification, a cloud computer may be one that uses any computer scalably to perform a particular function, or one that includes multiple functional modules to implement a certain system and uses those functions in any combination.

[0022] The maintenance prediction device 10 includes a control unit comprising a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), etc., and a communication unit comprising a device for enabling communication with other terminals and devices, a learning data acquisition unit 11 for acquiring sensor data of learning consumables and the degree of wear of these consumables, and a prediction data acquisition unit 12 for acquiring predetermined sensor data of consumables. Furthermore, the maintenance prediction device 10 includes a data storage unit, such as a hard disk, semiconductor memory, storage medium, or memory card, as its storage unit. Furthermore, the maintenance prediction device 10 includes, as a processing unit, various devices that perform various processes, a generation unit 13 that learns sensor data of learning consumables and associated data that links these consumables to the consumables and generates a learning model, and a prediction unit 14 that uses the learning model to estimate the degree of wear of the consumables in relation to predetermined sensor data of the consumables and predicts when the consumables should be replaced.

[0023] In the maintenance prediction device 10, the control unit reads a predetermined program and, in cooperation with the communication unit, implements a learning data acquisition module, a prediction data acquisition module, a condition matching notification module, a prediction result output module, a quotation data approval acceptance module, a maintenance unique number output module, a vehicle entry information acquisition module, a consumables notification module, and an invoice notification module. Furthermore, in the maintenance prediction device 10, the control unit loads a predetermined program, and in cooperation with the memory unit, realizes a learning model memory module. Furthermore, in the maintenance prediction device 10, the control unit reads a predetermined program and, in cooperation with the processing unit, realizes an associated data generation module, a learning model generation module, an inference module, a prediction module, a condition judgment module, a maintenance unique number generation module, a work schedule registration module, a maintenance unique number matching module, a maintenance completion judgment module, and an invoice issuance module.

[0024] The maintenance reservation device 20 is a computer or personal computer with server functions that performs various processes in a workshop where vehicle maintenance is carried out. The maintenance reservation device 20, like the maintenance prediction device 10 described above, may be implemented on a single computer, for example, or on multiple computers, such as a cloud computer.

[0025] The maintenance reservation device 20, like the maintenance prediction device 10 described above, includes a control unit 21 with a CPU, GPU, RAM, ROM, etc., a communication unit 22 with a device for enabling communication with other terminals and devices, a storage unit with a data storage unit, and a processing unit 23 with various devices for executing various processes.

[0026] In the maintenance reservation device 20, the control unit 21 reads a predetermined program and, in cooperation with the communication unit 22, realizes a prediction result acquisition module, a delay notification module, a reservation details notification module, and a quotation data approval acceptance module. Furthermore, in the maintenance reservation device 20, the control unit 21 reads a predetermined program and, in cooperation with the processing unit 23, realizes a work schedule confirmation module, a work time estimation module, an availability date and time confirmation module, an inventory confirmation module, a maintenance reservation module, a delivery date and time confirmation module, a delay judgment module, and an estimate data issuance module.

[0027] [Tracking model generation process for maintenance prediction device 10] Based on Figure 3, the learning model generation process performed by the maintenance prediction device 10 will be explained. Figure 3 is a flowchart showing the learning model generation process performed by the maintenance prediction device 10. The processes performed by each module described above will be explained in conjunction with this process. Note that this learning model generation process is a detailed explanation of the learning data acquisition process (step S1) and the learning model generation process (step S2) described above.

[0028] First, the training data acquisition module acquires training data (sensor data of training consumables and the degree of wear of the consumables) (step S10). As mentioned above, consumables are items that are subject to inspection during vehicle maintenance, and include belts, tires, lights, support parts, and the compressor for the refrigeration compartment. Sensor data, as mentioned above, includes vibration sensor data, distance sensor data, image sensor data, illuminance sensor data, and pressure sensor data. The degree of wear, as mentioned above, includes the degree of belt looseness, the degree of tire tread wear, the degree of light wear, the degree of deterioration of support parts, and the amount of refrigerant filled. The learning data acquisition module acquires sensor data detected by various sensors attached to the vehicle, and the degree of wear of various consumables determined by the maintenance worker while performing maintenance on the vehicle, as learning data. The learning data acquisition module may acquire sensor data directly from the various sensors, or it may acquire sensor data via other terminals or devices. The learning data acquisition module may acquire the degree of wear of various consumables from the degree of wear of various consumables that is input by the maintenance worker's terminal, or it may acquire this by directly receiving input of the degree of wear of these consumables. Furthermore, the learning data acquisition module acquires sensor data of various consumables stored on an external server, etc., as well as the degree of wear and tear of these consumables. Furthermore, the number and types of consumables, sensor data, and the number and types of wear and tear in the training data acquired by the training data acquisition module can be changed as appropriate, not limited to the examples described above.

[0029] The association data generation module generates association data that associates sensor data of the consumables used for learning with their degree of wear (step S11). The association data generation module associates the sensor data of each consumable in the acquired training data with the degree of wear of each consumable, and generates association data. The association data generation module generates association data for each acquired sensor data. Specifically, the association data generation module generates association data by associating vibration sensor data with the degree of belt looseness, by associating distance sensor data or image sensor data with the degree of tire tread wear, by associating illuminance sensor data with the degree of light wear, by associating vibration sensor data with the degree of deterioration of support parts, and by associating pressure sensor data with the amount of refrigerant filled.

[0030] The learning model generation module learns associated data and generates a learning model (step S12). As described above, learning methods include machine learning using supervised learning, unsupervised learning, reinforcement learning, etc., as well as deep learning using convolutional neural networks, recurrent neural networks, long- and short-term memory, etc. In this embodiment, the case of machine learning using supervised learning will be described. Needless to say, the present invention is also applicable to other learning methods. The learning model generation module performs supervised learning based on the generated associated data. The learning model generation module uses the generated associated data and the replacement timing of this consumable, based on maintenance manuals, etc., as supervised data for machine learning. As mentioned above, the replacement timing refers to the period from several weeks to several days before the day on which it is predicted that the consumable will reach a wear value at which it will malfunction.

[0031] The machine learning process performed by the learning model generation module will be explained using belt and vibration sensor data, and tire and distance sensor data as examples. When the learning model generation module performs machine learning on a belt, it uses supervised data such as vibration frequency (f) and vibration velocity (cm / s) from vibration sensor data, the degree of belt slack, and the timing of belt replacement. Similarly, when the learning model generation module performs machine learning on a tire, it uses supervised data such as mileage from distance sensor data, the degree of tire tread wear, and the timing of tire replacement. For other consumables and sensor data, just like with belts and tires, machine learning is performed using sensor data, wear and tear, and replacement timing as supervised data. The learning model generation module then generates learning models for each of the other consumables based on the results of this learning.

[0032] [Training data] Based on Figures 4 and 5, the training data used by the learning model generation module for training will be explained. Figure 4 is a schematic diagram showing an example of training data for belts used by the learning model generation module for training. Figure 5 is a schematic diagram showing an example of training data for tires used by the learning model generation module for training. In Figure 4, the learning model generation module uses the following as training data: vibration frequency (f) "A, C, ...", vibration velocity (cm / s) "B, D, ...", belt slack "large, small, ...", and belt replacement timing "replace immediately, after 6 months (after 2000 km of driving), ...". The learning model generation module uses vibration frequency "A", vibration velocity "B", slack "large", and replacement timing "replace immediately" as training data, and similarly uses the other parameters as training data. Furthermore, in Figure 5, the learning model generation module uses the following as training data: mileage "20,111km, 18,222km, ...", tire tread wear (distance to the wear indicator) "4.5mm, 5.3mm, ...", and tire replacement timing "replace immediately, after 6 months (after 2000km of driving), ...". The learning model generation module uses mileage "20,111km", tire tread wear "4.5mm", and replacement timing "replace immediately" as training data, and similarly uses the other data as training data.

[0033] Returning to Figure 3, we will continue explaining the learning model generation process. The learning model memory module stores the learning model (step S13). The learning model memory module stores the learning model for each consumable generated.

[0034] The above describes the learning model generation process.

[0035] [Replacement timing prediction processing of maintenance prediction device 10] Based on Figure 6, the replacement timing prediction process performed by the maintenance prediction device 10 will be explained. Figure 6 is a flowchart showing the replacement timing prediction process performed by the maintenance prediction device 10. The processes performed by each module described above will be explained in conjunction with this process. This replacement timing prediction process is a detailed explanation of the prediction data acquisition process (step S3) and the consumable replacement timing prediction process (step S4) described above.

[0036] The prediction data acquisition module acquires sensor data of consumables (step S20). The consumables and their sensor data in this process are the same as those used in the learning model generation process described above. The prediction data acquisition module may acquire sensor data directly from various sensors attached to the vehicle, or it may acquire this sensor data via other terminals or devices. In this embodiment, a belt and tires are described as examples of consumables. Vibration sensor data is described as an example of belt sensor data. Distance sensor data is described as an example of tire sensor data.

[0037] The estimation module estimates the degree of wear of the consumables (step S21). The estimation module uses the learning model generated by the process in S12 described above to estimate the degree of wear of the consumables in relation to the acquired sensor data. The estimation module estimates the degree of wear of the consumables corresponding to the sensor data acquired this time that matches or approximates the sensor data in the learning model. If the consumable is a belt, the estimation module uses a belt learning model to estimate the degree of belt slack in relation to the amplitude and vibration velocity in the acquired vibration sensor data. Similarly, if the consumable is a tire, the estimation module uses a tire learning model to estimate the degree of tire tread wear in relation to the mileage in the acquired distance sensor data. The prediction module, similarly for other consumables, uses a consumable learning model to predict the degree of wear of the consumables based on the acquired sensor data.

[0038] The prediction module predicts when consumables will need to be replaced (step S22). The prediction module uses the learning model generated by the process in step S12 described above to predict when consumables will need to be replaced based on the estimated wear and tear of the consumables. The prediction module predicts the replacement time for the consumables that corresponds to the wear and tear of the consumables estimated in this step that matches or approximates the wear and tear of the consumables in the learning model. If the consumable is a belt, the prediction module uses a belt learning model to predict when the belt should be replaced based on the estimated degree of belt slack. Similarly, if the consumable is a tire, the prediction module uses a tire learning model to predict when the tire should be replaced based on the estimated degree of tire tread wear. The prediction module, similarly for other consumables, uses a learning model of the consumables to predict when to replace them based on their estimated wear and tear. Furthermore, the prediction module's prediction of consumable replacement timing may be used as new training data to update the learning model. The prediction module will then use the updated learning model to predict the replacement timing of consumables for future replacements.

[0039] The condition determination module determines whether the replacement time meets predetermined conditions (step S23). The predetermined conditions are, for example, that the number of days until the replacement time is within a predetermined number of days (for example, within 1 day, within a few days, within a week).

[0040] If the condition determination module determines that the replacement time meets the predetermined conditions (step S23 YES), the condition matching notification module notifies that the replacement time meets the predetermined conditions (step S24). If the number of days until the replacement time is within a few days, the condition matching notification module notifies the administrator or driver of the need to replace the consumables as soon as possible. The condition matching notification module transmits this condition matching notification to the administrator terminal or driver terminal. The administrator terminal or driver terminal receives this condition matching notification and displays the condition matching notification shown in Figure 7 on its display unit. By displaying this condition matching notification on the administrator terminal or driver terminal, the condition matching notification module notifies the administrator or driver that the replacement time meets the conditions. The administrator or driver will understand that the replacement time meets the predetermined conditions by viewing the condition matching notification displayed on this terminal. After completing this process, the maintenance prediction device 10 executes the process of step S25, which will be described later. Furthermore, if the condition determination module determines that the replacement timings of multiple consumables meet predetermined conditions, the condition matching notification module can also be configured to group together the consumables that are nearing their replacement time and notify the administrator or driver of the earliest replacement time among them. This approach encourages the bulk replacement of consumables and prevents them from being forgotten, while also reducing costs by allowing the continued use of consumables that are still usable and have a long time until they need replacing.

[0041] [Condition Match Notification Screen] The condition matching notification screen will be explained based on Figure 7. Figure 7 is a schematic diagram showing an example of the condition matching notification screen displayed on the administrator terminal. In Figure 7, the administrator terminal displays a condition-matching notification indicating that the fan belt needs to be replaced urgently because the remaining time until replacement is within a few days. The administrator terminal displays the name of the consumable to be serviced, the date and time until replacement, the status of the consumable, and a message as part of the condition-matching notification. The condition-matching notification module notifies the administrator that the replacement time has been met by displaying this condition-matching notification on the administrator terminal.

[0042] Returning to Figure 6, we will continue explaining the replacement timing prediction process. If the condition determination module determines that the replacement timing does not meet the predetermined conditions (step S23 NO), the prediction result output module outputs the degree of wear of the consumables and the replacement timing to the maintenance reservation device 20 (step S25). The prediction result output module associates the vehicle identifier (ID, name, license plate, management number, etc.) shown in Figures 8 and 9 with the consumable identifier (ID, name, model, serial number, management number, etc.), the degree of wear, and the replacement timing, and outputs the prediction result to the maintenance reservation device 20. The maintenance reservation device 20 receives the outputted prediction result and uses this received prediction result in the maintenance reservation process described later. If the consumable is a belt, the prediction result output module outputs the prediction result by associating the vehicle ID, consumable ID, degree of looseness, and replacement time. Similarly, if the consumable is a tire, the prediction result output module outputs the prediction result by associating the vehicle ID, consumable ID, degree of tire tread wear, and replacement time. The prediction result output module outputs prediction results for other consumables as well, associating the vehicle ID, consumable ID, degree of wear of the consumable, and replacement timing. Furthermore, the prediction result output module can also be configured to group together consumables that are nearing their replacement date if multiple consumables requiring replacement are predicted, change the replacement dates of the other consumables to match the replacement date of the consumable with the earliest replacement date, and output the wear status of these consumables and the changed replacement dates to the maintenance reservation device 20. This approach encourages the bulk replacement of consumables and prevents them from being forgotten, while also reducing costs by allowing the continued use of consumables that are still usable and have a long time until they need replacing.

[0043] [Output of prediction results] Based on Figures 8 and 9, the output content of the prediction results output by the prediction result output module will be explained. Figure 8 is a schematic diagram showing the prediction results for a belt output by the prediction result output module. Figure 9 is a schematic diagram showing the prediction results for a tire output by the prediction result output module. In Figure 8, the prediction result output module associates the vehicle ID "12345", the consumable ID "234567", the degree of looseness "Severe", and the replacement timing "Replace immediately", and outputs this information to the maintenance reservation device 20. Furthermore, in Figure 9, the prediction result output module associates the vehicle ID "12345", the consumable ID "345678", the tire tread wear degree "5.8 mm", and the replacement timing "14 months later (replace after 4700 km of driving)" and outputs this information to the maintenance reservation device 20.

[0044] The content of the prediction results output by the prediction result output module is not limited to the example described above; it is sufficient if it outputs at least the degree of wear of consumables and the replacement timing for those consumables. Furthermore, the prediction result output module can be configured to output the prediction results to itself instead of to the maintenance reservation device 20. In this case, the maintenance prediction device 10 can perform the processing that the maintenance reservation device 20 would normally perform in the processing described later.

[0045] The above is the replacement timing prediction process.

[0046] [Service reservation processing] Based on Figure 10, the maintenance reservation process performed by the maintenance reservation device 20 will be explained. Figure 10 is a flowchart showing the maintenance reservation process performed by the maintenance reservation device 20. The processes performed by each module described above will be explained in conjunction with this process. Although this maintenance reservation process involves the transmission and reception of data between the maintenance reservation device 20 and the administrator terminal via the maintenance prediction device 10, for the sake of clarity, it will be described as if the maintenance reservation device 20 and the administrator terminal directly transmit and receive data.

[0047] The prediction result acquisition module acquires the degree of wear and replacement timing of consumables (step S30). The prediction result acquisition module acquires the degree of wear and replacement timing of consumables by receiving the degree of wear and replacement timing of consumables output by the maintenance prediction device 10 through the processing in step S25 described above. At this time, the reservation result acquisition module also acquires the vehicle ID and consumable ID.

[0048] The work schedule confirmation module checks the work schedule (step S31). The work schedule confirmation module refers to a pre-registered work schedule database that associates the date and time the maintenance work will be performed, the vehicle ID of the vehicle the maintenance work will be performed on, and the details of the maintenance work, and confirms the work schedule by checking for dates and times when there are no other maintenance works (available dates and times).

[0049] The work time estimation module estimates the work time (step S32). The work time estimation module estimates the work time based on the acquired consumable ID. The work time estimation module estimates this work time by referring to a work time database or the like, which has been pre-associated consumable IDs with work time.

[0050] The availability confirmation module checks for available dates and times based on the work time (step S33). The availability confirmation module then identifies available dates and times from the confirmed work schedule that meet the estimated work time.

[0051] The inventory check module determines whether consumables are in stock at the available time (step S34). The inventory check module refers to the inventory database where the consumable inventory is registered and the work schedule database to determine whether the consumables to be replaced are in stock at the checked available time.

[0052] If the inventory check module determines that consumables are in stock for an available date and time (step S34 YES), the maintenance reservation module makes a provisional reservation for maintenance (step S35). A provisional reservation is a temporary reservation of an available date and time, and this available date and time is reserved for a predetermined period (for example, the validity period of the quotation data described later). If approval from the administrator is not received by the time the predetermined period has elapsed through the process in step S42 described later, the reservation is canceled. If approval from the administrator is received through the process in step S42 described later, the reservation is maintained. The maintenance reservation module provisionally reserves vehicle maintenance by registering the vehicle's ID and the maintenance work details for the confirmed available date and time. After executing this process, the maintenance reservation device 20 executes the process of step S40 described later.

[0053] On the other hand, if the inventory check module determines that there is no stock of consumables during an available time slot (step S34 NO), the delivery date and time check module checks the delivery date and time of the consumables that are out of stock (step S36). The delivery date and time check module refers to the order history of the consumables that are out of stock and confirms the delivery date and time of these consumables.

[0054] The delay determination module determines whether the delivery date and time are earlier than the replacement date and time (step S37). The delay determination module compares the confirmed availability date and time with the confirmed delivery date and time of the consumables and determines whether the delivery date and time are earlier than the replacement date and time. In other words, this process is in which the delay determination module determines whether the out-of-stock consumables will be replenished by the replacement date and time.

[0055] If the delay determination module determines that the delivery date and time is earlier than the replacement date and time (step S37 YES), the maintenance reservation module makes a provisional reservation for maintenance (step S38). The maintenance reservation module makes a provisional reservation for vehicle maintenance by registering the vehicle's vehicle ID and the maintenance work details at the confirmed available date and time. After executing this process, the maintenance reservation device 20 executes the process of step S40 described later. At this time, if there is no stock of multiple consumables, the maintenance reservation module makes a provisional reservation for maintenance at the available date and time corresponding to the consumable with the latest delivery date and time.

[0056] On the other hand, if the delay determination module determines that the delivery date and time is not earlier than (later than) the exchange date and time (step S37 NO), the delay notification module notifies that the delivery date and time is later than the exchange date and time (step S39). The delay notification module outputs a delay notification to the administrator terminal indicating that the delivery date and time is later than the exchange date and time. The administrator terminal receives this delay notification and displays the delay notification shown in Figure 11 on its display unit, etc. By displaying this delay notification on the administrator terminal, the delay notification module notifies the administrator that the delivery date and time is later than the exchange date and time. By doing so, administrators can view delay notifications displayed on their administrator terminals and understand that consumables will not be delivered by the scheduled replacement date, making it easier to consider appropriate countermeasures. After executing this process, the maintenance reservation device 20 terminates the maintenance reservation process.

[0057] [Delay Notification] Based on Figure 11, the content of the delay notification output by the delay notification module will be explained. Figure 11 is a schematic diagram showing an example of a delay notification displayed on the administrator terminal. In Figure 11, the administrator terminal displays the name of the consumable item to be serviced, "Belt," its inventory status, "Out of Stock," the scheduled work date and time, "2021 / 01 / 29," the scheduled delivery date and time, "2021 / 01 / 31," and a message indicating that the replacement will not be completed on time. By viewing this delay notification, the administrator can understand that the item will not be delivered by the replacement date and time, and can consider taking appropriate measures such as suspending the operation of the vehicle as soon as possible. The content of the delay notification is not limited to the example above and can be changed as needed; at the very least, it should notify that the delivery date is later than the exchange date.

[0058] Returning to Figure 10, we will continue explaining the maintenance reservation process. The processing after the execution of steps S35 and S38 described above will now be explained. The quotation data generation module generates quotation data (step S40). The quotation data generation module refers to a cost database in which costs are pre-registered for each maintenance task and calculates the amount required for the current maintenance task. If there are multiple consumables to be used in the maintenance task, a discount amount is set according to the number of items, and this amount is subtracted from the amount required for the maintenance task to calculate the final estimated amount. This discount amount can be changed as appropriate, such as by setting the amount to vary depending on the number of items, setting the amount to be fixed depending on the number of items, or other settings. The quotation data generation module generates quotation data that includes details of the maintenance work, the amount required for the maintenance work, the discount amount (which can be omitted if there is only one consumable item), and the total amount. Please note that the format and content of the quotation data can be changed as needed.

[0059] The reservation notification module notifies the administrator of the reservation details and the quotation data (step S41). The reservation notification module notifies the administrator of the provisionally reserved reservation details and the issued quotation data. The reservation notification module outputs the work date and time on which the maintenance work will be performed, the vehicle ID, the maintenance work details, and the issued quotation data to the administrator terminal. The administrator terminal receives this reservation details and quotation data and displays the reservation notification shown in Figures 12 and 13 on its display unit. By displaying this reservation notification on the administrator terminal, the reservation notification module notifies the administrator of the reservation details and quotation data. By viewing the reservation notification displayed on the administrator terminal, the administrator can understand the scheduled work date and time, the vehicle ID of the vehicle requiring maintenance, and the consumables.

[0060] [Reservation Details Notification] Based on Figure 12, the reservation details notified by the reservation notification module will be explained. Figure 12 is a schematic diagram showing an example of a reservation notification displayed on the administrator terminal. In Figure 12, the administrator terminal displays the reservation details, including the vehicle ID "41235", the scheduled work date and time "2021 / 01 / 29 13:00-14:00", and the names of the consumables to be serviced, "belt, tire". The administrator terminal accepts input from the administrator (for example, inputting into the estimate data confirmation icon) and displays the estimate data. By viewing this reservation notification, the administrator can understand the scheduled work date and time, the vehicle ID of the vehicle requiring maintenance, and the consumables.

[0061] [Quotation Data Notification] Based on Figure 13, the quotation data notified by the reservation notification module will be explained. Figure 13 is a schematic diagram showing an example of quotation data displayed on the administrator terminal. In Figure 13, the administrator terminal displays the cost of the maintenance work in the reservation as estimate data. Since the estimate data includes multiple consumables such as belts and tires as part of the maintenance work, the discount amount is also displayed. The administrator terminal receives input from the administrator (for example, input operation on the estimate data approval icon), approves the estimate data, and outputs an approval notification to the maintenance reservation device 20 indicating that it has been approved. Upon receiving this approval notification, the maintenance reservation device 20 accepts the administrator's approval of the estimate data. By viewing this estimate data, the administrator can understand the amount required for the maintenance work.

[0062] Returning to Figure 10, we will continue explaining the maintenance reservation process. The quotation data approval receiving module accepts approval of the quotation data (step S42). The quotation data approval receiving module accepts that the administrator has approved this quotation data. The administrator terminal accepts input from the administrator approving the quotation data. The administrator terminal outputs an approval notification to the maintenance reservation device 20 indicating that the administrator has approved the quotation data. Upon receiving this approval notification, the quotation data approval receiving module accepts approval of the quotation data from the administrator. Furthermore, if the quotation data approval module does not accept approval for the quotation data (including cases where it is rejected), this maintenance reservation process will be terminated.

[0063] The maintenance reservation module makes a formal reservation for maintenance (step S43). A formal reservation is the official reservation of an available date and time that was provisionally reserved through the processes in steps S35 and S38 described above.

[0064] The above is the maintenance reservation process.

[0065] [Vehicle maintenance procedures] Based on Figure 14, the vehicle maintenance processing performed by the maintenance prediction device 10 will be explained. Figure 14 is a flowchart showing the vehicle maintenance processing performed by the maintenance prediction device 10. The processing performed by each module described above will be explained in conjunction with this processing.

[0066] The quotation data approval receiving module accepts approval of the quotation data (step S50). The quotation data approval receiving module accepts that the administrator has approved the quotation data. The administrator terminal accepts input from the administrator approving the quotation data. The administrator terminal outputs an approval notification to the maintenance prediction device 10 indicating that the administrator has approved the quotation data. Upon receiving this approval notification, the quotation data approval receiving module accepts approval of the quotation data from the administrator. The process in step S50 is executed at the same time as the process in step S42 described above. That is, when the process in step S42 described above is performed, the administrator terminal outputs a notification to both the maintenance prediction device 10 and the maintenance reservation device 20 indicating that the estimate data has been approved. Furthermore, if the quotation data approval module does not accept approval for the quotation data (including cases where it is rejected), the vehicle maintenance process will be terminated. Furthermore, upon receiving this approval, the maintenance prediction device 10 may notify the vehicle driver that vehicle maintenance will be performed. This notification may include information related to the vehicle maintenance, such as the vehicle ID, the date and time of the work, and the details of the work.

[0067] The maintenance-specific number generation module generates a maintenance-specific number (step S51). The maintenance-specific number is a unique number assigned to each maintenance job. The maintenance-specific number generation module generates the maintenance-specific number according to the work date and time, work details, and vehicle ID of the vehicle performing the maintenance work, as found in the approved estimate data. This maintenance-specific number is associated with the work date and time, work details, and vehicle ID.

[0068] The work schedule registration module registers a maintenance-specific number in the work schedule (step S52). The work schedule registration module registers this maintenance-specific number in the work schedule database at the date and time corresponding to the work date and time.

[0069] The maintenance-specific number output module outputs a maintenance-specific number to the maintenance reservation device 20 (step S53). The maintenance-specific number output module transmits the maintenance-specific number to the maintenance reservation device 20. The maintenance reservation device 20 receives this maintenance-specific number and registers it in the work schedule. The maintenance reservation device 20 registers the maintenance-specific number in the work schedule by associating this maintenance-specific number with the work schedule database where the maintenance has been officially reserved, as described in step S43 above.

[0070] The vehicle entry information acquisition module acquires vehicle entry information indicating that a vehicle has entered the maintenance yard (step S54). The vehicle entry information includes the vehicle ID. The vehicle entry information acquisition module acquires vehicle entry information from the maintenance reservation device 20, maintenance worker terminal, driver terminal, or administrator terminal, etc. When a vehicle enters the maintenance yard, the maintenance reservation device 20, etc., transmits the vehicle entry information to the maintenance prediction device 10. The vehicle entry information acquisition module acquires the vehicle entry information by receiving this vehicle entry information. If the vehicle entry information acquisition module has not acquired vehicle entry information, this process will be repeated.

[0071] The maintenance-specific identification number verification module verifies the maintenance-specific identification number (step S55). The maintenance-specific identification number verification module compares the vehicle ID in the vehicle entry information with the vehicle ID associated with the maintenance-specific identification number, and verifies the maintenance-specific identification number of the vehicle to be serviced this time.

[0072] The consumables notification module notifies the maintenance worker of the consumables to be used (step S56). The consumables notification module outputs the consumables associated with the verified maintenance-specific number to the maintenance worker's terminal. The maintenance worker's terminal receives these consumables and displays them on its own display unit. The consumables notification module notifies the maintenance worker of these consumables by displaying them on the maintenance worker's terminal. The maintenance worker will then be able to identify the consumables to be used for maintenance by viewing the consumables displayed on the maintenance worker's terminal.

[0073] The maintenance completion determination module determines whether or not the maintenance work has been completed (step S57). The maintenance completion determination module makes this determination based on whether or not it has received a maintenance work completion notification or the like from the maintenance worker's terminal indicating that the maintenance work has been completed. If the maintenance completion determination module determines that the maintenance work has not been completed (step S57 NO), the maintenance completion determination module repeats this process until it determines that the maintenance work has been completed.

[0074] On the other hand, if the maintenance completion determination module determines that the maintenance work is complete (step S57 YES), the invoice issuance module issues an invoice (step S58). The invoice issuance module issues an invoice in a predetermined format based on the amount and other details stated in the approved estimate data.

[0075] The invoice notification module notifies the administrator terminal of the invoice (step S59). The invoice notification module outputs the issued invoice to the administrator terminal. The administrator terminal receives this invoice and displays it on its display unit, etc. The invoice notification module notifies the administrator of this invoice by displaying it on the administrator terminal. The administrator will be able to understand the payment amount for the maintenance work by viewing the invoice displayed on the administrator terminal.

[0076] The above outlines the procedures to be followed during vehicle maintenance.

[0077] The means and functions described above are realized by a computer (including the CPU, information processing unit, and various terminals) reading and executing a predetermined program. The program may be provided, for example, in the form of a computer delivered via a network (SaaS: Software as a Service) or as a cloud service. Alternatively, the program may be provided in the form of a computer-readable recording medium. In this case, the computer reads the program from the recording medium, transfers it to an internal or external recording device, records it, and executes it. Alternatively, the program may be pre-recorded on a recording device (recording medium) and provided to the computer from that recording device via a communication line.

[0078] Although embodiments of the present invention have been described above, the present invention is not limited to these embodiments. Furthermore, the effects described in the embodiments of the present invention are merely a list of the most preferred effects arising from the present invention, and the effects of the present invention are not limited to those described in the embodiments.

[0079] (1) A vehicle maintenance prediction system that predicts the replacement timing of consumables subject to inspection during vehicle maintenance (e.g., belts, tires, lights, support parts, compressors for the refrigeration compartment), A learning data acquisition unit (e.g., learning data acquisition unit 11, learning data acquisition module) acquires sensor data of the consumables for learning purposes (e.g., vibration sensor data, distance sensor data, image sensor data, illuminance sensor data, pressure sensor data) and the degree of wear of the consumables (e.g., belt looseness, tire tread wear, light wear, support component deterioration, refrigerant charge amount), A generation unit (for example, generation unit 13, learning model generation module) generates a learning model by learning associated data that links the acquired sensor data of the learning consumables with the degree of wear of the said consumables, A predictive data acquisition unit (for example, a predictive data acquisition unit 12, a predictive data acquisition module) that acquires predetermined sensor data of the consumables, A prediction unit (e.g., prediction unit 14, prediction module) uses the generated learning model to estimate the degree of wear of the consumables in relation to predetermined sensor data of the consumables and predicts when the consumables should be replaced. A vehicle maintenance prediction system equipped with the following features.

[0080] According to invention (1), it becomes possible to predict the appropriate replacement time for consumable parts of a vehicle.

[0081] (2) A prediction result notification unit (e.g., a condition matching notification module) notifies the user if the prediction result of the replacement time of the consumables satisfies a predetermined condition (for example, the number of days until the replacement time is within a predetermined number of days), The vehicle maintenance prediction system described in (1) further comprises the following:

[0082] According to the invention in (2), it becomes easier for the recipient to understand that the predetermined conditions have been met, and they can then consider taking appropriate action.

[0083] (3) The consumable item is a belt, The learning data acquisition unit acquires vibration sensor data of the belt for learning and the degree of looseness of the belt. The generation unit learns association data that links the vibration sensor data of the learning belt with the degree of looseness of the belt, and generates a learning model. The prediction data acquisition unit acquires vibration sensor data of the belt, The prediction unit uses the generated learning model to estimate the degree of looseness of the belt based on the acquired vibration sensor data of the belt, and predicts when the belt should be replaced. (1) The vehicle maintenance prediction system described above.

[0084] According to invention (3), it becomes possible to predict the appropriate time for replacing the belt.

[0085] (4) The consumable item is a tire, The aforementioned learning data acquisition unit acquires distance sensor data of the tire or image sensor data of the tire tread for learning purposes, and the degree of wear of the tire tread. The generation unit generates a learning model by learning associated data that links the distance sensor data of the tire used for learning or image sensor data of the tire groove with the degree of wear of the tire groove. The predictive data acquisition unit acquires distance sensor data of the tire or image sensor data of the tire groove, The prediction unit uses the generated learning model to estimate the degree of wear of the tire grooves based on the acquired distance sensor data of the tire or image sensor data of the tire grooves, and predicts when the tire should be replaced. (1) The vehicle maintenance prediction system described above.

[0086] According to invention (4), it becomes possible to predict the appropriate time to replace tires.

[0087] (5) An output unit (for example, a prediction result output module) that outputs the replacement timing of the consumable predicted by the prediction unit and the consumable, A replacement date and time notification unit (for example, a reservation details notification module) checks whether vehicle maintenance is possible at the time of replacement, based on the replacement date of the consumables output, and notifies the date and time of replacement of the consumables. The vehicle maintenance prediction system described in (1) further comprises the following:

[0088] According to invention (5), it becomes possible to determine the date and time when maintenance can be performed.

[0089] (6) The replacement date and time notification unit also notifies the quotation data for the replacement of the consumables. (5) The vehicle maintenance prediction system described above.

[0090] According to the invention of (6), it becomes easier to grasp the costs related to the replacement of consumables.

[0091] (7) A vehicle maintenance prediction method performed by a vehicle maintenance prediction system that predicts the timing of replacement of consumables subject to inspection for vehicle maintenance, Steps include acquiring sensor data of the consumables used for learning and the degree of wear of the consumables (for example, step S10), Steps include: generating a learning model by learning associated data that links the acquired sensor data of the consumables used for learning with the degree of wear of the consumables (for example, step S12); The steps include acquiring predetermined sensor data of the consumables (for example, step S20), Using the generated learning model, the process involves estimating the degree of wear of the consumable item in relation to predetermined sensor data of the consumable item, and predicting the replacement time of the consumable item (for example, step S22). A vehicle maintenance prediction method comprising the following features.

[0092] (8) A computer that predicts when consumables subject to inspection during vehicle maintenance will replace: Steps to acquire sensor data of the consumables used for learning and the degree of wear of the consumables (for example, step S10), Steps to generate a learning model (for example, step S12): Learn associated data that links the acquired sensor data of the consumables used for learning with the degree of wear of the consumables. Steps to acquire predetermined sensor data of the consumables (for example, step S20), Using the generated learning model, the process involves estimating the degree of wear of the consumable item in relation to predetermined sensor data of the consumable item, and predicting the replacement time of the consumable item (for example, step S22). A computer-readable program for executing a command. [Explanation of symbols]

[0093] 1. Vehicle Maintenance Prediction System 3 Network 5 vehicles 10 Maintenance prediction device 11. Training data acquisition unit 12. Data acquisition unit for prediction 13 Generation part 14 Prediction Section 20 Maintenance reservation device 21 Control Unit 22 Communications Department 23 Processing Unit

Claims

1. A vehicle maintenance prediction system that predicts the replacement timing of consumable parts subject to inspection during vehicle maintenance, A learning data acquisition unit that acquires sensor data of the consumables for learning purposes and the degree of wear of the consumables, A generation unit that generates a learning model by learning associated data that links the acquired sensor data of the consumables used for learning with the degree of wear of the consumables, A predictive data acquisition unit that acquires predetermined sensor data of the consumables, A prediction unit that uses the generated learning model to estimate the degree of wear of the consumables in relation to predetermined sensor data of the consumables and predicts when the consumables should be replaced. A prediction unit that estimates the work time according to the replacement time and consumables, An output unit that outputs the predicted replacement time for the consumable, the estimated work time, and inventory information for the consumable; A replacement date and time notification unit checks whether vehicle maintenance is possible at the time of replacement, based on the outputted replacement time of the consumable, the work time, the inventory information of the consumable, and the work schedule, and if there is an available date and time and inventory, it provisionally reserves the vehicle maintenance and notifies the customer of the provisionally reserved date and time as the replacement date and time of the consumable; A vehicle maintenance prediction system equipped with the following features.

2. A prediction result notification unit notifies the user if the predicted replacement time for the aforementioned consumables meets predetermined conditions, The vehicle maintenance prediction system according to claim 1, further comprising:

3. The vehicle maintenance prediction system according to Claim 2, wherein the prediction result notification unit determines that the replacement timing of a plurality of consumables of the vehicle meets predetermined conditions, groups together the consumables that are nearing replacement time from among the plurality of consumables, and notifies the administrator or driver of the earliest replacement time among them.

4. The aforementioned consumable is a belt, The learning data acquisition unit acquires vibration sensor data of the belt for learning and the degree of looseness of the belt. The generation unit learns association data that links the vibration sensor data of the learning belt with the degree of looseness of the belt, and generates a learning model. The prediction data acquisition unit acquires vibration sensor data of the belt, The prediction unit uses the generated learning model to estimate the degree of looseness of the belt based on the acquired vibration sensor data of the belt, and predicts when the belt should be replaced. The vehicle maintenance prediction system according to claim 1.

5. The aforementioned consumable is a tire. The aforementioned learning data acquisition unit acquires distance sensor data of the tire or image sensor data of the tire tread for learning purposes, and the degree of wear of the tire tread. The generation unit generates a learning model by learning associated data that links the distance sensor data of the tire used for learning or image sensor data of the tire groove with the degree of wear of the tire groove. The predictive data acquisition unit acquires distance sensor data of the tire or image sensor data of the tire groove, The prediction unit uses the generated learning model to estimate the degree of wear of the tire grooves based on the acquired distance sensor data of the tire or image sensor data of the tire grooves, and predicts when the tire should be replaced. The vehicle maintenance prediction system according to claim 1.

6. The vehicle maintenance prediction system according to claim 1, wherein the replacement date and time notification unit also notifies the including the data of the estimate for the replacement of the consumables.

7. A vehicle maintenance prediction method performed by a vehicle maintenance prediction system that predicts the replacement timing of consumables subject to inspection during vehicle maintenance, A step of acquiring sensor data of the consumables used for learning and the degree of wear of the consumables, The steps include: generating a learning model by learning associated data that links the acquired sensor data of the consumables used for learning with the degree of wear of the consumables; The steps include: acquiring predetermined sensor data of the consumable item; Using the generated learning model, the process involves estimating the degree of wear of the consumable item in relation to predetermined sensor data of the consumable item, and predicting the replacement time for the consumable item. The steps include: estimating the work time according to the replacement time and consumables; Steps include outputting the predicted replacement time for the consumable, the estimated work time, and inventory information for the consumable; The steps include: checking whether vehicle maintenance is possible at the time of replacement, based on the outputted replacement time of the consumable, the work time, the inventory information of the consumable, and the work schedule; making a provisional reservation for the vehicle maintenance if there is an available date and time and inventory; and notifying the customer of the provisional reservation date and time as the replacement date and time of the consumable; A vehicle maintenance prediction method comprising the following features.

8. A computer that predicts when consumable parts to be replaced during vehicle maintenance inspections will A step of acquiring sensor data of the consumables used for learning and the degree of wear of the consumables, A step of generating a learning model by learning associated data that links the acquired sensor data of the consumables used for learning with the degree of wear of the consumables, A step of acquiring predetermined sensor data for the consumables, Using the generated learning model, estimate the degree of wear of the consumable item based on predetermined sensor data of the consumable item, and predict when the consumable item should be replaced. A step of estimating the work time according to the aforementioned replacement time and consumables, A step of outputting the predicted replacement time for the consumable, the estimated work time, and the inventory information for the consumable. The steps include: checking whether vehicle maintenance is possible at the time of replacement based on the outputted replacement time of the consumable, the work time, the inventory information of the consumable, and the work schedule; making a provisional reservation for the vehicle maintenance if there is an available date and time and inventory; and notifying the customer of the provisional reservation date and time as the replacement date and time of the consumable; A computer-readable program for executing a command.

Images