Maintenance Support System

Abstract

To provide a maintenance support system capable of detecting an abnormality of a machine before failure occurs and notifying a user of the abnormality while reducing machine downtime due to subsequent failures by preventing out-of-stock countermeasure parts for repairing a failure caused by the abnormality.SOLUTION: A maintenance support system 1 includes a control unit 100a that is configured so as to, when an abnormality is detected on a machine 10, estimate countermeasure parts for repairing the failure caused by the abnormality and their required quantity, notify a user 30 of the content of the abnormality, identification information of countermeasure parts and required quantity, and when the forecast number of countermeasure parts in stock is less than the forecast number of orders received, generate information to increase the inventory of countermeasure parts.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a maintenance support system that supports appropriate maintenance work on machines. [Background technology]

[0002] For machines that operate continuously for long periods of time, such as construction machinery or wind turbines, it is important to improve the availability rate so that the customer (machine owner, user, agent, or sales office) can maximize their profits. Therefore, a system is needed that can quickly and appropriately replace parts or repair the machine before a breakdown occurs, or that can quickly repair the machine and get it back up and running again if a breakdown does occur. To achieve this, a mechanism is needed that constantly monitors the machine's condition and supports appropriate maintenance work. Hereinafter, a system that supports appropriate maintenance work for machines will be referred to as a "maintenance support system."

[0003] Machine condition monitoring involves using sensors equipped on the machine to collect machine operation information and physical quantities such as temperature and acceleration at a fixed frequency, and analyzing and processing the collected data to monitor the machine's condition and determine whether it is normal or abnormal. The simplest method is to estimate the machine's condition by setting a threshold for the physical quantity. In recent years, abnormality diagnosis methods have been developed that utilize machine learning technology, a field of artificial intelligence technology, to learn (also called "training") from previously collected machine operation data to estimate and determine the current condition. In addition, a method has been disclosed for determining the relationship between the value of a physical quantity and the lifespan deterioration rate, and estimating the remaining lifespan of a machine based on this relationship (Patent Document 1).

[0004] Furthermore, as disclosed in Patent Document 2, a method has been proposed for formulating a maintenance and operation scenario for a machine or component based on an estimated value of the failure risk.

[0005] Furthermore, there is a demand for maintaining an appropriate inventory level for parts necessary for machine maintenance or repair, and as disclosed in Patent Document 3, a method has been proposed in which the demand for maintenance parts is estimated from the results of failure detection and parts inventory is adjusted based on this. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2012-181169 [Patent Document 2] Japanese Patent Application Publication No. 2019-113883 [Patent Document 3] International Publication No. 2013 / 145203 Summary of the Invention [Problem to be solved by the invention]

[0007] As mentioned above, in order to minimize machine downtime due to breakdowns, there is a demand for condition monitoring systems that can detect machine abnormalities and quickly resolve those abnormalities. Furthermore, if an event occurs in which a core part is out of stock or requires a long time for transportation, depending on the time it takes to procure the part, the machine may have to be shut down for an extended period of time. From this perspective, there is a demand for the ability to quickly procure parts necessary for machine maintenance or repair.

[0008] This issue needs to be resolved from both the parts user (customer) and parts provider (manufacturer) sides.

[0009] Users of parts (customers) are required to receive the results of anomaly detection obtained from a condition monitoring system as soon as possible so they can begin preparations for machine maintenance or repair. Therefore, it is important to not only notify customers of the details of the detected anomaly, but also of information regarding the parts required for maintenance or repair and their procurement time, and to provide maintenance inspection recommendations. Patent Document 2 discloses a technique for calculating a machine's health index from machine operation data and formulating a maintenance scenario based on this health index. However, this technique does not take into account the inventory status or procurement time of parts required for machine repair, raising concerns about increased machine downtime due to prolonged parts procurement times.

[0010] Meanwhile, parts providers (manufacturers) are expected to adjust their inventory of countermeasure parts to prevent stockouts. In particular, if a shortage occurs due to inadequate inventory of specialized, irreplaceable parts or parts with long procurement lead times, production of the parts begins after a customer order is received, which can result in a long procurement time and a decrease in machine utilization. To prevent parts from running out of stock, it is necessary to not only plan machine maintenance but also forecast parts demand, including customer part purchase plans in the event of an abnormality. Patent Document 3 discloses a method for estimating demand for maintenance parts from failure probability and adjusting parts inventory based on that estimate. However, because this is a countermeasure method solely implemented by the parts provider (manufacturer), it does not take into account the part purchase plans of the parts users (customers), which can lead to a decrease in the accuracy of parts demand forecasts.

[0011] The present invention has been made in view of such problems, and its purpose is to provide a maintenance support system that detects machine abnormalities and notifies the customer before a breakdown occurs, and prevents the stock of countermeasure parts for repairing the breakdown caused by the abnormality from running out, thereby reducing the downtime of the machine due to subsequent breakdowns. [Means for solving the problem]

[0012] In order to achieve the above object, the present invention provides a maintenance support system having a server for supporting maintenance work of a machine, the server comprising a control device having a calculation function, a storage device for storing data necessary for the calculation processing of the control device, and a communication device for enabling communication between the machine and the control device, the control device receiving operation data including data measured by a sensor mounted on the machine via the communication device and storing it in the storage device, detecting an abnormality of the machine based on the operation data, and when the abnormality is detected, identifying countermeasure parts and the required quantity of parts to repair a failure that is likely to occur after the detection of the abnormality due to the abnormality among parts used in the machine based on the failure and repair history of the machine, and calculate a predicted procurement time, which is a predicted value of the time from when a customer of the machine places an order for the countermeasure part until the countermeasure part arrives at a location designated by the customer, based on parts production data including a production plan and a delivery plan for the countermeasure part, parts procurement data of a parts warehouse including a shipping method and a time required for shipping the countermeasure part, and sales office / agency data including an address of a sales office / agency that handles the countermeasure part, a transportation method of the countermeasure part and a delivery means related to a delivery company, and a procurement history of the countermeasure part by the sales office / agency; The details of the abnormality, the identification information of the countermeasure part, and the required quantity and said estimated lead time via the communication device said customer and notifies the customer of the number of countermeasure parts in stock, the maintenance plan for the machine, and the The aforementioned Production planning and The aforementioned The device calculates a time-series change in the stock forecast number, which is a forecast value of the stock number of the countermeasure parts, based on the arrival plan, and after notifying the abnormality, receives the purchase number of the countermeasure parts input by the customer via the communication device and stores it in the storage device as a purchase history of the countermeasure parts, calculates an order forecast number, which is a forecast value of the number of orders for the countermeasure parts, based on the purchase history, and generates information for increasing the stock of the countermeasure parts when the stock forecast number is lower than the order forecast number.

[0013] According to the present invention configured as described above, it is possible to detect a machine abnormality and notify the customer before a failure occurs, and also to prevent the countermeasure parts for correcting the failure caused by the abnormality from running out of stock, thereby reducing the downtime of the machine due to subsequent failure. [Effects of the Invention]

[0014] According to the present invention, it is possible to detect machine abnormalities and notify the customer before a failure occurs, and to prevent countermeasure parts for correcting the failure caused by the abnormality from running out of stock, thereby reducing the amount of time the machine is down due to subsequent failures. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a diagram illustrating a physical configuration of a maintenance support system according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a functional block diagram of a control device according to the first embodiment of the present invention. [Figure 3] FIG. 2 is a diagram showing a data structure of a storage device according to the first embodiment of the present invention. [Figure 4] FIG. 3 is a diagram showing an example of the transition of the abnormality degree of a machine and threshold values ​​in the first embodiment of the present invention. [Figure 5] FIG. 3 is a diagram showing an example of the data structure of a failure and repair history database in the first embodiment of the present invention. [Figure 6] FIG. 3 is a diagram showing an example of a data structure of a countermeasure component database in the first embodiment of the present invention. [Figure 7] FIG. 4 is a diagram showing an example of a calculation result of a parts inventory estimation unit in the first embodiment of the present invention. [Figure 8] FIG. 4 is a diagram showing an example of a processing procedure of a parts procurement time prediction unit in the first exemplary embodiment of the present invention. [Figure 9] FIG. 4 is a diagram showing an example of a calculation result of a parts procurement time prediction unit in the first embodiment of the present invention. [Figure 10] FIG. 4 is a diagram showing an example of a monitor display of information transmitted to a customer by an alarm transmission unit in the first embodiment of the present invention. [Figure 11] FIG. 2 is a diagram illustrating an example of the configuration of a parts demand forecasting unit in the first embodiment of the present invention. [Figure 12] FIG. 10 is a diagram illustrating a physical configuration of a maintenance support system according to a second embodiment of the present invention. [Figure 13] FIG. 10 is a diagram illustrating a physical configuration of a maintenance support system according to a third embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] A maintenance support system according to an embodiment of the present invention will be described below with reference to the drawings. In the drawings, the same reference numerals are used to designate the same components, and redundant description will be omitted where appropriate. [Example]

[0017] Fig. 1 is a diagram showing the physical configuration of a maintenance support system in a first embodiment of the present invention. The maintenance support system 1 is a system that monitors the state of a machine 10 and supports maintenance work on the machine 10, and includes a server 100 and a storage device 200. The server 100 includes a control device 100a and a communication device 100b. Note that although Fig. 1 shows a hydraulic excavator as an example of the machine 10, the machine 10 is not limited to construction machinery.

[0018] The control device 100a is configured with one or more computers equipped with a power supply, a CPU (Central Processing Unit), a memory, an input / output device, etc. The control device 100a is connected to a network 20 configured with a LAN (Local Area Network), the Internet, etc. via a communication device 100b.

[0019] The storage device 200 is composed of, for example, one or more hard disks. The storage device 200 may be installed within the server 100, or may be connected to the server 100 via the network 20. The storage device 200 stores data related to the processing of the control device 100a in electronic file format or database format such as a relational database. In this embodiment, the data is stored in database format.

[0020] The machine 10 is made up of multiple subunits or parts, and includes various sensors 11 that measure pressure, temperature, etc., a control device 12, and a communication device 13. The control device 12 transmits information (operation data) such as pressure and temperature measured by the sensors 11 to the maintenance support system 1 via the communication device 13.

[0021] Additionally, customers 30 of machine 10 and a parts warehouse 40 that stores parts used in machine 10 are connected via network 20. Parts warehouse 40 manages the production of parts by parts producers 50 and the arrival of parts from parts producers 50 to parts warehouse 40.

[0022] A functional block diagram of the control device 100a is shown in Fig. 2. In Fig. 2, the control device 100a includes a machine condition analysis unit 101, a countermeasure part estimation unit 102, a part inventory estimation unit 103, a part procurement time prediction unit 104, an alarm transmission unit 105, an alarm response storage unit 106, a part demand prediction unit 107, and a part inventory adjustment unit 108.

[0023] 3 shows the data configuration of storage device 200. Storage device 200 stores a machine operation database 201, a failure and repair history database 202, a parts production database 203, a parts procurement database 204, a sales office and agent database 205, an alarm response database 206, a countermeasure parts database 207, a maintenance plan database 208, a parts inventory database 209, and a parts order history database 210. In each figure, databases are abbreviated as "DB."

[0024] The machine condition analysis unit 101 acquires operation data of the machine 10 via the communication device 100b and stores the data in a machine operation database 201. The machine condition analysis unit 101 calculates an index value (health index) indicating the health of the machine 10 from the operation data of the machine 10 using algorithms such as data mining, machine learning, and remaining life assessment. In this embodiment, a condition analysis method using an anomaly detection algorithm will be described.

[0025] In the anomaly detection algorithm, physical quantities called "features" are defined from the operational data of the machine 10, and the distribution of the features obtained from a normal machine 10 is created as a normal model. A feature is, for example, a combination of multiple physical quantities extracted from the operational data of the machine 10, and can be described in vector format. Hereinafter, a combination of multiple features described in vector format will be referred to as a feature vector. The normal model corresponding to the feature vector can be described by the mean (hereinafter referred to as the mean vector) and variance of the feature vector obtained from a normal machine 10. The degree of anomaly of the machine 10 can be measured by the degree to which the newly obtained feature vector deviates from the mean vector. For example, according to a statistical algorithm known as the Mahalanobis-Taguchi method, the degree of anomaly of the machine 10 can be calculated using the following formula:

[0026]

number

[0027] Here, a is the degree of anomaly, x is the newly obtained feature vector, μ is the mean vector, and σ is the standard deviation (square root of the variance) of the feature vector x obtained by the normal machine 10. The numerator on the left side of equation (1) indicates the square of the distance from the new feature vector x to the mean vector μ, and the denominator on the right side indicates the variance of the feature vector x used to calculate the mean vector μ.

[0028] The state of the machine 10 can be analyzed by using the degree of anomaly a as an indicator of the health of the machine 10. A commonly used analysis method is to set a threshold value for the degree of anomaly a, and determine that the machine 10 is abnormal if the degree of anomaly a is equal to or greater than the threshold, and determine that the machine 10 is normal if the degree of anomaly a is less than the threshold. The threshold value for the degree of anomaly a can be determined based on the relationship between the degree of anomaly a and whether or not a subsequent failure occurs.

[0029] 4 shows an example of the transition of the anomaly level a and the threshold value of the machine 10. An algorithm for determining an anomaly in the machine 10 using the anomaly level a and the threshold value can be written in the following pseudocode.

[0030]

number

[0031] When an abnormality is detected in the state analysis of the machine 10, the countermeasure part estimation unit 102 outputs a list (countermeasure part list) of parts (countermeasure parts) required to repair the failure caused by the abnormality. The countermeasure parts can be estimated from information in the failure and repair history database 202. Examples of the failure type (failure mode) include battery deterioration, tooth damage, etc.

[0032] FIG. 5 shows an example of the data structure of the failure / repair history database 202. The failure / repair history database 202 stores failure cases (cases), health indexes at the time of analysis (abnormality level a), the number of days from the time of analysis to failure, failure modes, a list of parts replaced during repair (replacement parts list), and the like. In FIG. 5, in case 1, the health index at the time of analysis is "A1," the number of days from the time of analysis to failure is 15 days, the failure mode is "M1," and the replacement parts list is "PL1." In case 2, the health index at the time of analysis is "A2," the number of days from the time of analysis to failure is 7 days, the failure mode is "M2," and the replacement parts list is "PL2." In case n, the health index at the time of analysis is "An," the number of days from the time of analysis to failure is 28 days, the failure mode is "Mn," and the replacement parts list is "PLn."

[0033] Fig. 6 shows an example of the data structure of countermeasure component database 207. Countermeasure component database 207 stores the component numbers, names, and required quantities of countermeasure components for each failure mode. In the example shown in Fig. 6, five pressure sensors (component number: PAAA-000) and three filters (component number: PAAA-001) are required to deal with failure mode MA, and 20 screws (component number: PXXX-003) are required to deal with failure mode MX.

[0034] The parts inventory estimation unit 103 calculates time-series changes in inventory quantity for each countermeasure part estimated by the countermeasure part estimation unit 102. The parts inventory estimation unit 103 uses information from the parts inventory database 209, the machine 10 maintenance plan database 208, and the parts production database 203 as input data. The parts inventory database 209 contains information such as the inventory quantity of each part handled by the parts warehouse 40, part attributes (part number, name, weight, price, etc.), and past sales figures. The maintenance plan database 208 contains information such as current and future maintenance plans and lists of parts to be used for maintenance for all machines 10 that have a parts supply relationship with the parts warehouse 40. The parts production database 203 contains information such as production plans and receipt plans for each part handled by the parts warehouse 40.

[0035] For a certain part A, if the inventory quantity on the estimated start date (Day0) is NA0, then n Inventory forecast quantity NA Dayn is calculated using the following formula:

[0036]

number

[0037] where IN_A_i is the date i OUT_A_i is the planned number of parts A to arrive on the date Day i This is the planned shipment quantity of part A in the period, and can be estimated from the information in the maintenance plan database 208 and the part demand forecast.

[0038] 7 shows an example of the calculation result of the parts inventory estimation unit 103. In FIG. 7, the calculation result of the parts inventory estimation unit 103 is the estimated start date (date Day 0) to the date Day n This shows the changes in the planned arrival quantity, planned shipment quantity, and forecast inventory quantity of part A up to now.

[0039] The parts procurement time prediction unit 104 calculates a predicted value (predicted procurement time) of the time (procurement time) from when the countermeasure parts are ordered until the countermeasure parts arrive at a location specified by the customer (the machine's work site, a distributor / sales office, or an organization that repairs machines). If there is a wealth of past performance data, it is possible to calculate a highly accurate predicted procurement time by using a machine learning algorithm such as a neural network. In addition, the parts procurement time prediction unit 104 may notify the customer 30 of an estimate of when the countermeasure parts can be procured based on the predicted procurement time (for example, "about one week later," "about one month later," etc.). This allows the customer 30 to know when the countermeasure parts can be procured, making it possible to revise the operation plan for the machine 10 in advance.

[0040] 8 shows an example of the processing procedure of the parts procurement time prediction unit 104 in this embodiment. Each step will be explained below in order.

[0041] Step S401: The parts inventory database 209 is referenced to check the number of countermeasure parts in stock on the order date.

[0042] Step S402: Determine whether the inventory of countermeasure parts is equal to or greater than the order quantity (whether countermeasure parts are in stock). If it is determined that the inventory of countermeasure parts is equal to or greater than the order quantity (Yes), execute step S404. If it is determined that the inventory of countermeasure parts is less than the order quantity (No), execute step S403 and then execute step S404.

[0043] Step S403: Based on the information in the parts production database 203 (production plan and arrival plan for the countermeasure parts), a predicted value (predicted production time) of the time (production time) from when production of the countermeasure parts starts until the countermeasure parts arrive at the parts warehouse is calculated.

[0044] Step S404: Based on the information in the parts procurement database 204 (the shipping method of the countermeasure parts and the time required for shipping, etc.) and the information in the sales office / agency database 205 (the address of each sales office / agency, preferred delivery means, past procurement history of countermeasure parts, etc.), a predicted value (predicted delivery time) of the time it takes for the countermeasure parts to be delivered from the parts warehouse 40 to the location specified by the customer is calculated.

[0045] Step S405: The predicted production time and the predicted delivery time are added together to calculate the predicted procurement time.

[0046] Fig. 9 shows an example of the calculation results of the part procurement time prediction unit 104. In Fig. 9, the calculation results of the part procurement time prediction unit 104 include the inventory status, predicted production time, delivery means, predicted delivery time, and predicted procurement time of each countermeasure part (parts 1 to n).

[0047] The alarm transmission unit 105 transmits, via the communication device 100b, information about the abnormality of the machine 10, a list of countermeasure parts (including quantity, price, etc.) estimated by the countermeasure part estimation unit 102, the estimated procurement time estimated by the part procurement time prediction unit 104, and inspection and maintenance procedures for dealing with the abnormality, together with an alarm, to the customer 30. The alarm can be in the form of a warning sound, a lamp light, a monitor display, an email, a telephone call, a fax, etc.

[0048] FIG. 10 shows an example of a monitor display of alarm information transmitted by the alarm transmission unit 105 to the customer 30. In FIG. 10, a main display frame 301 displays the date and time when the abnormality was detected (anomaly detection date and time), the identification information (machine ID) of the machine 10 in which the abnormality was detected, and the details of the abnormality. In addition, by clicking buttons 302 and 303 arranged in the main display frame 301, inspection and maintenance procedures for dealing with the abnormality and a list of parts that require countermeasures can be displayed in separate display frames 304 and 305. The display items and display method of the alarm information can be changed as appropriate depending on the application software running on the terminal of the customer 30.

[0049] After the alarm transmitting unit 105 transmits an alarm or the like, the alarm response storage unit 106 receives, via the communication device 100b, response information (alarm response data) from the customer 30 in response to the alarm, and stores the information in the alarm response database 206. The alarm response data includes the order date for the countermeasure parts, the number of countermeasure parts purchased, the start date and end date of the machine inspection / repair, etc.

[0050] 11 shows an example of the configuration of the parts demand forecasting unit 107. The parts demand forecasting unit 107 is composed of input data 107a, a learning model 107b, and output data 107c. The input data 107a includes alarm response data acquired by the alarm response storage unit 106, basic machine data such as the model and production year of the machine 10, operation data of the machine 10, maintenance plan data for the machine 10, failure modes, Identify the customer of machine 10 Customer data (sales office and agency data), Includes parts procurement history for parts warehouses The input data 107a includes parts procurement data, parts order history data, etc. The output data 107c includes identification information (e.g., part name) of countermeasure parts related to the failure mode and the number of orders predicted. The learning model 107b uses a supervised machine learning model that can learn the relationship between the input data 107a and the output data 107c. Typical supervised machine learning models include neural networks, decision trees, random forests, and deep learning. The parts demand forecasting unit 107 converts the input data 107a into output data 107c using the learning model 107b, and feeds back the output data 107c to the parts inventory adjustment unit 108 as a parts demand forecast.

[0051] The parts inventory adjustment unit 108 compares the predicted inventory number of countermeasure parts estimated by the parts inventory estimation unit 103 with the predicted order number of countermeasure parts predicted by the parts demand forecasting unit 107. If the predicted order number of a certain countermeasure part exceeds the predicted inventory number, the information (production plan or arrival plan) in the parts production database 203 is changed so that the production or arrival time of the part is brought forward to prevent stockout of the part. That is, the parts inventory adjustment unit 108 in this embodiment generates arrival plan or production plan change information for bringing forward the production or arrival time of the countermeasure part as information for increasing the inventory of the countermeasure part. Note that various information for increasing the inventory of countermeasure parts can be considered depending on the means for increasing inventory. On the other hand, if the predicted order number is significantly lower than the predicted inventory number, the arrival plan or production plan is changed so that the arrival or production time of the part is brought forward to reduce excess inventory of the part.

[0052] (summary) In this embodiment, in a maintenance support system 1 having a server 100 and for supporting maintenance work of a machine 10, the server 100 is provided with a control device 100a having a calculation function, a storage device 200 for storing data necessary for the calculation processing of the control device 100a, and a communication device 100b for enabling communication between the machine 10 and the control device 100a, and the control device 100a receives operation data including data measured by a sensor 11 mounted on the machine 10 via the communication device 100b, stores the data in the storage device 200, detects an abnormality in the machine 10 based on the operation data, and when the abnormality is detected, performs a fault detection for a part used in the machine 10 that is caused by the abnormality. The countermeasure parts for repairing the fault and their required quantity are estimated, the content of the abnormality and the identification information and required quantity of the countermeasure parts are notified to the customer 30 via the communication device 100b, a predicted inventory quantity which is a predicted value of the inventory quantity of the countermeasure parts is calculated, the purchase quantity of the countermeasure parts input by the customer 30 via the communication device 100b after notifying the abnormality is received and stored in the storage device 200 as a purchase history of the countermeasure parts, a predicted order quantity which is a predicted value of the order quantity of the countermeasure parts is calculated based on the purchase history, and when the predicted inventory quantity is lower than the predicted order quantity, information for increasing the inventory of the countermeasure parts is generated.

[0053] According to this embodiment configured as described above, an abnormality in the machine 10 is detected before a failure occurs and notified to the customer 30, and by preventing countermeasure parts for correcting the failure caused by the abnormality from running out of stock, it becomes possible to reduce the downtime of the machine 10 due to a subsequent failure.

[0054] Furthermore, the control device 100a calculates an index value a indicating the health of the machine 10 based on the operation data of the machine 10, and determines that the machine 10 is normal if the index value a is within a predetermined range (for example, less than a predetermined threshold), and determines that there is an abnormality in the machine 10 if the index value a is outside the predetermined range (for example, equal to or greater than a predetermined threshold). This makes it possible to improve the accuracy of detecting abnormalities in the machine 10.

[0055] Furthermore, the storage device 200 stores the failure and repair history of the machine 10, and when the control device 100a detects an abnormality in the machine 10, it estimates the countermeasure parts and the required quantity to deal with the abnormality based on the failure and repair history. This makes it possible to improve the accuracy of estimating the countermeasure parts and the required quantity.

[0056] Furthermore, the storage device 200 stores the inventory quantity of countermeasure parts, the maintenance plan for the machine 10, and the production plan and arrival plan for the countermeasure parts, and the control device 100a calculates the predicted inventory quantity of countermeasure parts based on the inventory quantity of countermeasure parts, the maintenance plan for the machine 10, and the production plan and arrival plan for the countermeasure parts. This makes it possible to improve the calculation accuracy of the predicted inventory quantity.

[0057] Furthermore, the storage device 200 stores parts production data including production plans and delivery plans for the countermeasure parts, parts procurement data including shipping methods and shipping times for the countermeasure parts, and sales office / agency data including addresses of sales offices / agency staff that handle the countermeasure parts, delivery means for the countermeasure parts, and procurement histories of the countermeasure parts. When an abnormality in the machine 10 is detected, the control device 100a calculates a predicted procurement time, which is a predicted value of the time from when the customer 30 places an order for the countermeasure parts until the countermeasure parts arrive at a location specified by the customer 30, based on the parts production data, the parts procurement data, and the sales office / agency data, and notifies the customer 30 of the predicted procurement time via the communication device 100b. This allows the customer 30 to know when the countermeasure parts will be available for procurement, enabling them to revise the operation plan for the machine 10 in advance.

[0058] Furthermore, when the control device 100a detects an abnormality in the machine 10, it notifies the customer 30 of the identification information of a countermeasure part to address the abnormality and the required quantity of the countermeasure part, as well as the price of the countermeasure part and the inspection and maintenance procedure to address the abnormality. Furthermore, when the control device 100a receives response information from the customer 30 regarding the abnormality via the communication device 100b, it stores the response information in the storage device 200. By leaving the response information as a history in this way, the customer 30 can respond quickly to the abnormality in the machine 10 and can also review the inspection and maintenance procedure to address the abnormality based on the response information from the customer 30.

[0059] The storage device 200 also stores basic data including the model and production year of the machine 10, operation data of the machine 10, maintenance plan data for the machine 10, failure modes caused by abnormalities in the machine 10, customer data for the machine 10, parts procurement data for countermeasure parts, and parts order history data including the purchase history of the countermeasure parts, and the control device 100a uses a supervised machine learning model to calculate the predicted number of orders for countermeasure parts based on input data including the operation data, the maintenance plan data, the failure modes, the customer data, the parts procurement data, and the parts order history data. This makes it possible to improve the accuracy of calculating the predicted number of orders for countermeasure parts.

[0060] Furthermore, the storage device 200 stores a production plan and a delivery plan for countermeasure parts to deal with abnormalities in the machine 10, and when the forecasted order quantity of the countermeasure parts exceeds the forecasted inventory quantity, the control device 100a generates change information for the production plan or the delivery plan to bring forward the production time or delivery time of the countermeasure parts as information for increasing the inventory of the countermeasure parts. This makes it possible to increase the inventory quantity of the countermeasure parts based on the production plan or delivery plan for the countermeasure parts. [Example]

[0061] A maintenance support system according to a second embodiment of the present invention will be described with reference to Fig. 12. Fig. 12 is a configuration diagram of the maintenance support system according to this embodiment. The following description will focus on differences from the first embodiment (shown in Fig. 1).

[0062] In FIG. 12, the maintenance support system 1A performs the same processing as in the first or second embodiment for each of a plurality of machines 10 operating at the same site or in the same area.

[0063] According to this embodiment configured as described above, it is possible to detect abnormalities in a plurality of machines 10 operating at the same site or in the same area and notify the customer 30, and also to prevent the countermeasure parts for repairing the failure caused by the abnormality from running out of stock, thereby shortening the downtime of each machine due to a subsequent failure. [Example]

[0064] A maintenance support system according to a third embodiment of the present invention will be described with reference to Fig. 13. Fig. 13 is a configuration diagram of the maintenance support system according to this embodiment. The following description will focus on differences from the first embodiment (shown in Fig. 1).

[0065] 13, the maintenance support system 1B is an application of edge computing technology, and some or all of the functions of the control device 100a of the server 100 in the first embodiment (shown in FIG. 2) are implemented in the control device 12 of the machine 10. The machine 10 also includes a storage device 14. The storage device 14 stores some or all of the databases 201 to 210 (shown in FIG. 3) stored in the storage device 200 in the first embodiment, depending on the functions assigned to the control device 12 of the machine 10.

[0066] According to the maintenance support system 1B in this embodiment configured as described above, it is possible to improve the processing efficiency of the entire system by distributing functions between the control device 100a of the server 100 and the control device 12 of the machine 10.

[0067] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to the above-described embodiments and includes various modifications. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to add part of the configuration of one embodiment to the configuration of another embodiment, or to delete part of the configuration of one embodiment or replace it with part of another embodiment. [Explanation of symbols]

[0068] 1, 1A, 1B...Maintenance support system, 10...Machine, 11...Sensor, 12...Control device, 13...Communication device, 20...Network, 30...Customer, 40...Parts warehouse, 50...Parts producer, 100...Server, 100a...Control device, 100b...Communication device, 101...Machine condition analysis unit, 102...Countermeasure part estimation unit, 103...Parts inventory estimation unit, 104...Parts procurement time prediction unit, 105...Alarm transmission unit, 106...Alarm response memory unit, 107...Parts demand prediction unit, 107a...Input data 107b...learning model, 107c...output data, 108...parts inventory adjustment unit, 200...storage device, 201...machine operation database, 202...fault and repair history database, 203...parts production database, 204...parts procurement database, 205...sales office and agent database, 206...alarm response database, 207...countermeasure parts database, 208...maintenance plan database, 209...parts inventory database, 210...parts order history database.

Claims

1. A maintenance support system having a server for supporting machine maintenance work, The server a control device having a calculation function; a storage device that stores data necessary for the arithmetic processing of the control device; a communication device that enables communication between the machine and the control device; The control device receiving operation data including data measured by a sensor mounted on the machine via the communication device and storing the data in the storage device; Detecting an abnormality in the machine based on the operation data; When the abnormality is detected, countermeasure parts and the required quantity for repairing a failure that is likely to occur after the detection of the abnormality due to the abnormality among the parts used in the machine are identified based on the failure and repair history of the machine, and calculate a predicted procurement time, which is a predicted value of the time from when a customer of the machine places an order for the countermeasure part until the countermeasure part arrives at a location designated by the customer, based on parts production data including a production plan and a delivery plan for the countermeasure part, parts procurement data of a parts warehouse including a shipping method and a time required for shipping the countermeasure part, which are stored in the storage device, and sales office / agency data including an address of a sales office / agency that handles the countermeasure part, a transportation method for the countermeasure part and a delivery means related to a delivery company, and a procurement history of the countermeasure part by the sales office / agency; notifying the customer via the communication device of the details of the abnormality, the identification information of the countermeasure part, the required quantity, and the predicted procurement time; calculating a time series change in a predicted inventory quantity, which is a predicted value of the inventory quantity of the countermeasure part, based on the inventory quantity of the countermeasure part, the maintenance plan for the machine, and the production plan and the receiving plan for the countermeasure part; After notifying the abnormality, the number of countermeasure parts purchased input by the customer via the communication device is received, and the number is stored in the storage device as a purchase history of the countermeasure parts; calculating a predicted order quantity, which is a predicted value of the number of orders for the countermeasure parts, based on the purchase history; When the stock forecast quantity falls below the order forecast quantity, information for increasing the stock of the countermeasure parts is generated. A maintenance support system characterized by:

2. 2. The maintenance support system according to claim 1, The control device calculating an index value indicating the health of the machine based on the operation data; If the index value is within a predetermined range, the machine is determined to be normal; If the index value is outside the predetermined range, it is determined that there is an abnormality in the machine. A maintenance support system characterized by:

3. 2. The maintenance support system according to claim 1, the storage device stores the failure and repair history of the machine, When the control device detects the abnormality, the control device identifies the countermeasure part and the required quantity based on the failure / repair history stored in the storage device. A maintenance support system characterized by:

4. 2. The maintenance support system according to claim 1, the storage device stores the inventory quantity of the countermeasure part, the maintenance plan for the machine, and the production plan and the receiving plan for the countermeasure part, The control device calculates the predicted inventory quantity based on the inventory quantity of the countermeasure part stored in the storage device, the maintenance plan for the machine, and the production plan and arrival plan for the countermeasure part. A maintenance support system characterized by:

5. 2. The maintenance support system according to claim 1, The control device When the abnormality is detected, notifying the customer of the identification information of the countermeasure part, the required quantity, the price of the countermeasure part, and an inspection and maintenance procedure for dealing with the abnormality; When the customer's response information to the abnormality is received via the communication device, the response information is stored in the storage device. A maintenance support system characterized by:

6. 2. The maintenance support system according to claim 1, the storage device stores basic data including the model and production year of the machine, the operation data, maintenance plan data for the machine, a failure mode caused by the abnormality, customer data for identifying the customer of the machine, parts procurement data for the countermeasure parts in a parts warehouse, and parts order history data including the purchase history; The control device calculates the order forecast quantity based on input data including the operation data, the maintenance plan data, the failure mode, the customer data, the parts procurement data, and the parts order history data, using a supervised machine learning model. A maintenance support system characterized by:

Images