WO2016125248A1

WO2016125248A1 - Maintenance assistance system, maintenance assistance method, and maintenance assistance program

Info

Publication number: WO2016125248A1
Application number: PCT/JP2015/052941
Authority: WO
Inventors: 峻行羽渕; 羽根　慎吾; 靖英森; 藤城　孝宏; 丈士白井
Original assignee: 株式会社日立製作所
Priority date: 2015-02-03
Filing date: 2015-02-03
Publication date: 2016-08-11

Abstract

A memory stores, among losses which a user incurs through maintenance of components which are installed in a device, a first loss index which represents an extent of a first loss, and a second loss index which represents an extent of a second loss. The first loss index is an index that a loss arising prior to faults with the components is lower than the loss arising after the faults with the components, and the second loss index is an index that the loss arising prior to the faults with the components is greater than the loss arising after the faults with the components. A processor acquires a plurality of fault probabilities of the components at a plurality of times during a period in which the components are used. Using the acquired plurality of fault probabilities, the first loss index, and the second loss index, the processor computes predicted values of the loss arising when maintaining the components for a plurality of candidate timings at which the maintenance is carried out during the period in which the components are used, and outputs the computed plurality of predicted values.

Description

Maintenance support system, maintenance support method, and maintenance support program

The present invention relates to a maintenance support system, a maintenance support method, and a maintenance support program.

When using equipment such as a gas engine, elevator, mining machine, construction machine, or computer, it is necessary to maintain the equipment in order to keep these equipment operating. One of the effective technologies for maintaining equipment is to detect signs of failure by diagnosing device abnormalities, and to provide a means of repair from failures to the administrator. Has been.

Also, in the prior art, a system for making a maintenance plan for equipment has been proposed (see, for example, Patent Document 1). According to the summary of Patent Document 1, “the operation load / simulation state of the construction machine is simulated by the operation simulation means based on the production operation condition, and then the cumulative load for each part according to the operation / work state is calculated. A system for predicting the life of each part based on "is described.

International Publication No. 2005/106139

When generating a maintenance plan using conventional technology, the maintenance plan is generated using the life of the equipment assumed as one constant. However, because the life of equipment varies from one piece of equipment to another, the maintenance plan generated by the prior art is not always appropriate for each real piece of equipment. .

The object of the present invention is to provide information for making an appropriate maintenance plan based on the actual value related to the lifetime of the equipment.

In order to solve the above-mentioned problems, the present invention is a maintenance support system comprising a processor and a memory, and the memory is a first disadvantage among the disadvantages suffered by a user due to maintenance of components mounted on a device. A first disadvantageous index indicating a degree of profit and a second disadvantageous index indicating a second degree of disadvantage are stored, and the first disadvantageous index is generated before failure of the component. The disadvantage is lower than the disadvantage that occurs after the failure of the component, and the second disadvantage index is that the disadvantage that occurs before the failure of the component is less than the disadvantage that occurs after the failure of the component The processor is configured to obtain a plurality of failure probabilities of the component at a plurality of time points during a period in which the component is used, the acquired plurality of failure probabilities, and the first disadvantage index; And using the second disadvantage index A predicted value of a disadvantage that occurs when the part is maintained is calculated at a plurality of candidate timings where maintenance is performed during a period in which the part is used, and the calculated plurality of predicted values are output. A maintenance support system.

According to the present invention, information for making an appropriate maintenance plan can be provided.

Issues, configurations, and effects other than those described above will be clarified by the following description of the embodiments.

1 is a block diagram illustrating a maintenance system according to a first embodiment. 2 is a functional block diagram of a maintenance server according to Embodiment 1. FIG. FIG. 6 is an explanatory diagram illustrating a failure history according to the first embodiment. It is explanatory drawing which shows the customer information of Example 1. FIG. It is explanatory drawing which shows the maintenance information of Example 1. FIG. It is explanatory drawing which shows the output information of Example 1. FIG. It is a flowchart which shows the process for calculating the maintenance cost of Example 1, an operation loss, and a lifetime loss. It is explanatory drawing which shows transition of the failure probability according to the abnormality degree of Example 1. FIG. It is explanatory drawing which shows transition of the abnormality degree according to the elapsed time of Example 1. FIG. It is explanatory drawing which shows transition of the failure probability according to the elapsed time of Example 1. FIG. It is explanatory drawing which shows distribution of the result of having multiplied the failure probability of Example 1 and the maintenance cost. It is explanatory drawing which shows distribution of the result of multiplying the failure probability of Example 1 and the operation loss. It is explanatory drawing which shows distribution of the result of having multiplied the failure probability of Example 1, and lifetime loss. It is a flowchart which shows the process which calculates the predicted value of the disadvantage in the maintenance of Example 1. It is explanatory drawing which shows the processing result of the maintenance cost in the maintenance timing of Example 1, an operation loss, and a lifetime loss. It is explanatory drawing which shows the processing result which calculated | required the total value of the maintenance cost in the maintenance timing of Example 1, operation loss, and lifetime loss. It is explanatory drawing which shows the screen which displays the disadvantage by the maintenance after a failure of Example 1. FIG. It is explanatory drawing which shows the screen which displays the disadvantage according to the maintenance timing of Example 1. FIG. It is explanatory drawing which shows the example of application of the maintenance system of Example 1. FIG. FIG. 3 is a sequence diagram illustrating an operation of the first embodiment. It is explanatory drawing which shows the failure log | history of Example 2. FIG. It is a flowchart which shows the process which calculates the predicted value of the disadvantage in the maintenance of Example 2. It is explanatory drawing which shows distribution of the failure probability according to the component hour meter of Example 2. FIG. It is explanatory drawing which shows distribution of the future failure probability of Example 2. FIG. It is explanatory drawing which shows the surplus table contained in the customer information of Example 3. FIG. It is a flowchart which shows the process which calculates the predicted value of the disadvantage in the maintenance of Example 3.

Hereinafter, examples will be described with reference to the drawings.

FIG. 1 is a block diagram illustrating a maintenance system according to the first embodiment.

The maintenance system of the first embodiment includes a maintenance server 100, an excavator 140, a truck 130, an on-vehicle device 150, a network 160, and a user terminal 170. The excavator car 140 and the truck 130 are devices (hereinafter referred to as target devices) that are maintained by the maintenance system of the first embodiment. The target devices shown in FIG. 1 are the excavator 140 and the truck 130, but may be a pipe, a generator, an engine, or the like, or a device such as a computer.

At least one vehicle-mounted device 150 is installed in each part of the excavator car 140 and the truck 130. The vehicle-mounted device 150 is connected to each component of the excavator car 140 and the truck 130 and has sensors for measuring vibration, temperature, pressure, and the like. Moreover, the onboard equipment 150 has a processor and memory.

The vehicle-mounted device 150 acquires the degree of abnormality indicating the type of abnormality that has occurred in the component and the strength of the abnormality, using the measurement value measured by the sensor. Here, the vehicle-mounted device 150 may detect an abnormality occurring in the component by any method. The vehicle-mounted device 150 may detect an abnormality when, for example, the difference between the standard value held in advance and the measured value measured by the sensor is larger than a predetermined threshold value.

Further, the vehicle-mounted device 150 may acquire the type of abnormality and the degree of abnormality by any method. For example, the vehicle-mounted device 150 may compare a measurement value measured by a sensor with a standard value held in advance, and calculate a greater degree of abnormality as the difference between the measurement value and the standard value increases. Moreover, the onboard equipment 150 may specify the type of abnormality based on the type and measurement value of the measured value determined to be abnormal and information on the abnormality that is held in advance.

In addition, the vehicle-mounted device 150 acquires the elapsed time that has elapsed since the component was installed in the target device, from the hour meter provided in the component. And the onboard equipment 150 sends the elapsed time which the acquired hour meter shows to the maintenance server 100. FIG.

The vehicle-mounted device 150, the user terminal 170, and the maintenance server 100 are connected via the network 160. The network 160 may be any network, for example, the Internet, a LAN, or a WAN.

The user terminal 170 is used by a maintenance system administrator or a maintenance person (hereinafter referred to as a maintenance person) to receive an instruction to the maintenance server 100 or output data transmitted from the maintenance server 100. The user terminal 170 includes an output device such as a processor, a memory, a display, or a printer, and an input device such as a keyboard or a mouse. The user terminal 170 may be a device in which an input device and an output device are integrated, such as a tablet terminal.

The maintenance server 100 is a device that generates information for the maintenance person to determine the maintenance timing based on the information collected from the vehicle-mounted device 150. The maintenance server 100 includes physical devices such as an I / O (Input / Output) interface 111, a memory 112, a communication device 113, a CPU 114, and an auxiliary storage device 120.

The I / O interface 111 is an interface for connecting to an input device and an output device. The maintenance server 100 may be directly connected to the input device and the output device via the I / O interface 111 instead of the network 160.

Then, an instruction from the maintenance person may be received using a directly connected input device, or data may be output using a directly connected output device. The following processing of the user terminal 170 may be executed by an input device and an output device that are directly connected to the maintenance server 100.

The communication device 113 is an interface for connecting to the network 160. The maintenance server 100 is connected to the vehicle-mounted device 150 and the user terminal 170 via the communication device 113.

The memory 112 is a storage device that temporarily holds data and programs. The CPU 114 is an arithmetic device and a control device, and may be at least one processor, for example. The CPU 114 executes a program using the memory 112.

The memory 112 includes a ROM that is a nonvolatile storage element and a RAM that is a volatile storage element. The ROM stores an immutable program (for example, BIOS). The RAM is a high-speed and volatile storage element such as a DRAM (Dynamic Random Access Memory), and temporarily stores a program stored in the auxiliary storage device and data used when the program is executed.

The auxiliary storage device 120 is a storage device that holds data and programs. The auxiliary storage device is a large-capacity non-volatile storage device such as a magnetic storage device (HDD) or a flash memory (SSD). The CPU 114 reads data and programs held in the auxiliary storage device 120 into the memory 112 as necessary. The auxiliary storage device 120 includes data such as a failure history 121, customer information 123, maintenance information 125, and output information 127, and programs such as an abnormality estimation unit 122, a cost loss calculation unit 124, an information acquisition unit 126, and an output unit 128. .

The program executed by the CPU 114 is provided to the maintenance server 100 via a removable medium (CD-ROM, flash memory, etc.) or a network, and stored in the auxiliary storage device 120 which is a non-temporary storage medium. For this reason, the maintenance server 100 may have an interface for reading data from a removable medium.

The maintenance server 100 may be physically one computer, or may be a computer system composed of a plurality of computers logically or physically. Further, the above-described program may be processed by a plurality of threads in one computer, or may be processed by a virtual computer constructed on a plurality of physical computer resources.

The failure history 121 includes information acquired from the parts by the vehicle-mounted device 150 and information indicating whether or not a failure has occurred. The customer information 123 includes information related to the target device, and particularly includes information acquired from a customer who uses the target device.

The maintenance information 125 includes information regarding the target device, and particularly includes information acquired as a result of maintaining the parts. The output information 127 includes a processing result in the maintenance server 100 that is output using the user terminal 170.

The information acquisition unit 126 acquires information generated in the target device from the vehicle-mounted device 150 and stores the acquired information in the failure history 121. The information acquisition unit 126 receives information from the input device connected to the user terminal 170 or the maintenance server 100, and stores the received information in the customer information 123, the maintenance information 125, and the like.

The information acquisition unit 126 may acquire information via any device, and may acquire information generated in the target device from the information server connected to the vehicle-mounted device 150 or the user terminal 170.

The abnormality estimation unit 122 obtains the failure probability according to the elapsed time of the parts based on the information acquired from the vehicle-mounted device 150. The cost loss calculation unit 124 calculates a predicted value of a user's disadvantage that occurs due to maintenance. The output unit 128 outputs the calculated predicted value of the user's disadvantage to the user terminal 170 or the like.

Although the abnormality estimation unit 122, the cost loss calculation unit 124, the information acquisition unit 126, and the output unit 128 illustrated in FIG. 1 are programs, the maintenance server 100 according to the first embodiment includes physical devices having the functions of these programs. May be. In addition, these programs may include subprograms divided for each process executed by the program, or a plurality of programs may be integrated into one program.

In addition, a person who is involved in the target device, such as a component owner and a component user, is simply referred to as a user in the following. The first embodiment provides information for determining an appropriate maintenance timing by calculating a disadvantage that a user suffers when a part is maintained.

FIG. 2 is a functional block diagram of the maintenance server 100 according to the first embodiment.

The information acquisition unit 126 collects information on abnormality from the vehicle-mounted device 150 and stores it in the failure history 121. In addition, the information acquisition unit 126 collects information about parts from the vehicle-mounted device 150 and stores the information in the customer information 123. The abnormality estimation unit 122 refers to the failure history 121 and calculates failure probabilities at a plurality of predetermined times. Then, the abnormality estimation unit 122 stores the calculated failure probability in the failure history 121.

Here, the predetermined time is an elapsed time after occurrence of an abnormality in Example 1 below, but may be an elapsed time after the use of a part is started.

Based on the failure probability calculated by the abnormality estimation unit 122, the customer information 123, and the maintenance information 125, the cost loss calculation unit 124 calculates a predicted value of a disadvantage that occurs when parts are replaced at a plurality of predetermined times. To do. Then, the cost loss calculation unit 124 stores information on the calculated predicted value of disadvantage in the output information 127 as a processing result.

In addition, in the case of “maintenance” in the following, it indicates a replacement.

FIG. 3 is an explanatory diagram showing the failure history 121 of the first embodiment.

The failure history 121 includes an abnormality degree transition table 300 and a failure probability table 310. The failure history 121 may hold a plurality of abnormality degree transition tables 300 for each user or maintenance person of the target device.

The information acquisition unit 126 stores information generated in the vehicle-mounted device 150 in a part of the abnormality degree transition table 300 and the failure probability table 310. The abnormality degree transition table 300 includes an abnormality name 301, an abnormality ID 302, a part name 303, a part ID 304, a time 305, and an abnormality degree 306.

The abnormality name 301 indicates the type of abnormality that has occurred in the part of the target device. The abnormality ID 302 is an identifier that uniquely indicates a combination of the type of abnormality and the type of component in which the abnormality has occurred.

The part name 303 indicates the name of the type of the part in which an abnormality has occurred. The component ID 304 is an identifier that uniquely indicates a component in all target devices of the maintenance system according to the first embodiment. Time 305 is an elapsed time since the occurrence of the abnormality. The time 305 indicates the elapsed time by, for example, seconds or minutes. The degree of abnormality 306 indicates the degree of abnormality calculated by the vehicle-mounted device 150.

The failure probability table 310 includes an abnormality name 311, an abnormality degree 312, the number of failures 313, and a failure probability 314. The abnormality name 311, the abnormality degree 312, and the number of failures 313 are information collected from the vehicle-mounted device 150 by the information acquisition unit 126.

The abnormality name 311 corresponds to the abnormality name 301 and indicates the type of abnormality. The abnormality degree 312 corresponds to the abnormality degree 306. The number of failures 313 indicates the number of failures detected in the combination of the type of abnormality indicated by the abnormality name 311 and the degree of abnormality 312. The failure number 313 indicates the number of failures detected for the first time.

The failure probability 314 indicates the probability that a failure will occur in a component at each abnormality level 312 of one abnormality type. The failure probability 314 indicates an instantaneous failure probability and indicates the probability of the number of cases where a failure is detected for the first time.

When information collected from the vehicle-mounted device 150 is stored in the failure probability table 310, the information acquisition unit 126 includes the failure name 313 of the entry corresponding to the abnormality name and abnormality included in the collected information, and is included in the collected information. Add the number of failures.

In addition, when updating the failure probability table 310, the information acquisition unit 126 calculates, as a failure probability, a ratio of the number of failure cases 313 of each entry to the total value of the failure cases 313 in one abnormality name 311. Here, the following information acquisition unit 126 calculates failure probabilities so that the sum of failure probabilities becomes 100 for one abnormality name 301. Then, the abnormality estimation unit 122 stores the calculated failure probability in the failure probability 314.

The failure probability 314 may be calculated by a device other than the maintenance server 100, and the information acquisition unit 126 may acquire a failure probability calculated by a device other than the maintenance server 100.

FIG. 4 is an explanatory diagram showing the customer information 123 of the first embodiment.

Customer information 123 includes a production record table 400 and an operation record table 410. The production result table 400 is information input in advance from the user terminal 170.

The production performance table 400 indicates the production amount of the target device on which the part is installed. The production result table 400 includes a customer ID 401, a site ID 402, a machine ID 403, a production amount 404 per unit time, and a product unit price 405.

Customer ID 401 uniquely indicates the user or maintenance person of the target device. The site ID 402 uniquely indicates a place where the target device is installed. The machine ID 403 uniquely indicates a target device on which a part is installed. The machine ID 403 of the first embodiment is a unique identifier among all target devices included in the maintenance system.

The production amount 404 per unit time is the production amount per unit time of the target device indicated by the machine ID 403. The production amount 404 per unit time may be any value as long as it is a value that quantitatively indicates the effect produced by the operation of the target device. For example, the production amount 404 per unit time may be the weight of an object that is produced or transported by the target device, or may be the amount of packet traffic that passes when the target device is a network device.

The product unit price 405 indicates the price of a product produced by the target device indicated by the machine ID 403. The production amount 404 per unit time and the product unit price 405 may be any items as long as they indicate profits generated per unit time when the target device indicated by the machine ID 403 operates.

The operation result table 410 includes information acquired from the vehicle-mounted device 150. The operation result table 410 includes a machine ID 411, a component ID 412, and a component hour meter 413.

The machine ID 411 corresponds to the machine ID 403 in the production result table 400 and uniquely indicates the target device. The component ID 412 corresponds to the component ID 304 in the abnormality degree transition table 300 and uniquely indicates each component in the maintenance system.

The part hour meter 413 indicates the value of the hour meter provided for the part indicated by the part ID 412. That is, the component hour meter 413 indicates the elapsed time that has elapsed since the component was installed in the target device. In particular, the component hour meter 413 according to the first embodiment indicates an elapsed time after the component is installed in the target device when the vehicle-mounted device 150 starts to detect the abnormality of the component.

For example, when the information corresponding to the degree of abnormality 306 included in the information collected from the vehicle-mounted device 150 changes from normal to a value indicating abnormality, the information acquisition unit 126 collects the value of the hour meter and uses the collected value as a component. It may be stored in the hour meter 413.

FIG. 5 is an explanatory diagram showing the maintenance information 125 of the first embodiment.

The maintenance information 125 includes a parts price table 500, a maintenance countermeasure table 510, and a maintenance result table 520.

The part price table 500 indicates the unit price for each type of part. The component price table 500 is information input in advance from the user terminal 170.

The part price table 500 includes a part name 501 and a part unit price 502. The part name 501 corresponds to the part name 303 in the abnormality degree transition table 300 and uniquely indicates the type of the part. The component unit price 502 indicates the unit price of the type of component indicated by the component name 501.

The maintenance measure table 510 indicates the cost required for maintenance. The maintenance countermeasure table 510 may be input in advance from the user terminal 170, or the cost loss calculation unit 124 may generate the maintenance countermeasure table 510 based on the maintenance result table 520. The maintenance countermeasure table 510 includes an abnormality name 511, a part name 512, a maintenance cost before failure 513, a maintenance cost after failure 514, a maintenance required time 515 before failure, and a maintenance required time 516 after failure.

The abnormality name 511 corresponds to the abnormality name 301 of the abnormality degree transition table 300 and uniquely indicates the type of abnormality. The part name 512 corresponds to the part name 303 in the abnormality degree transition table 300 and uniquely indicates the type of the part.

The pre-failure maintenance cost 513 indicates a statistical value of the cost required for replacement when the component is replaced before the component fails. The maintenance cost 514 after failure indicates a statistical value of cost required for replacement when the component is replaced after the component has failed.

The maintenance cost of the first embodiment is a cost necessary for maintaining a part, and is a disadvantage that a user incurs in maintaining the part. In general, it is less expensive to replace parts before failure occurs than to replace parts after failure occurs.

Here, the maintenance cost includes at least the price of the parts that have become abnormal and need to be maintained. Further, the maintenance cost may include the price of a part in which an abnormality has occurred and the price of a part that needs to be replaced together with the part in which the abnormality has occurred. Further, the maintenance cost may include labor cost necessary for maintenance in particular.

The maintenance required time 515 before failure indicates a statistical value of time required for maintenance when a part is replaced before failure occurs. The maintenance required time 516 after failure indicates a statistical value of time required for maintenance when a part is replaced after a failure occurs.

When a part is maintained after the part has failed, there is a high possibility that other parts have also failed, so the post-failure maintenance cost 514 is generally higher than the pre-failure maintenance cost 513. In addition, since it takes less time for maintenance to replace parts before failure occurs than to replace parts after failure occurs, maintenance time before failure 515 generally requires maintenance after failure. Less than time 516.

The statistical value in Example 1 is an average value or an intermediate value calculated from past performance values. The pre-failure maintenance cost 513, the post-failure maintenance cost 514, the pre-failure maintenance required time 515, and the post-failure maintenance required time 516 in FIG. 5A are statistical values calculated using the actual maintenance value table 520. However, the values of the maintenance cost before failure 513, the maintenance cost after failure 514, the maintenance cost before failure 513, and the maintenance cost after failure 514 may be directly input by the maintenance person.

The maintenance result table 520 holds the actual values of the maintenance cost and the required time that were required when the parts were replaced. The maintenance result table 520 is information input in advance from the user terminal 170. The maintenance result table 520 includes an abnormality name 521, an abnormality ID 522, before and after a failure 523, a maintenance cost 524, and a required time 525.

The abnormality name 521 corresponds to the abnormality name 301 in the abnormality degree transition table 300 and uniquely indicates the type of abnormality. The abnormality ID 522 corresponds to the abnormality ID 302 in the abnormality degree transition table 300 and uniquely indicates an abnormality that has occurred in the maintenance system.

Pre-failure 523 indicates whether the part has been replaced before or after the failure. When “before and after failure” 523 shown in FIG. 5 is “after failure”, it indicates that the component has been replaced after failure, and when “before failure”, it indicates that the component has been replaced before failure.

The maintenance cost 524 indicates a maintenance cost, and indicates a cost necessary for maintaining the part. The required time 525 indicates the time required to maintain the part. The required time 525 may include a work time for exchanging parts, a time for stopping the target device for replacement, and a time for starting the target device and starting use.

The information acquisition unit 126 updates the maintenance measure table 510 when updating the maintenance result table 520. Specifically, the information acquisition unit 126 divides the entries in the maintenance result table 520 into entry groups having the same abnormality name 521 and the same before and after failure 523. Then, the information acquisition unit 126 calculates the statistical value of the maintenance cost 524 and the statistical value of the required time 525 using the maintenance cost 524 and the required time 525 of the divided entry group.

When the entry group before and after failure 523 indicates “after failure”, the information acquisition unit 126 extracts an entry in the maintenance countermeasure table 510 including the abnormality name 511 and the part name 512 corresponding to the abnormality name 521 of the entry group. The maintenance cost 514 after failure of the extracted entry is updated with the calculated statistical value of the maintenance cost 524. Further, the information acquisition unit 126 updates the post-failure maintenance required time 516 of the extracted entry with the calculated statistical value of the required time 525.

If the entry group before and after failure 523 indicates “before failure”, the information acquisition unit 126 updates the pre-failure maintenance cost 513 of the extracted entry with the calculated statistical value of the maintenance cost 524 and calculates the calculated required time. The maintenance required time 515 before failure of the extracted entry is updated with the statistical value of 525.

Note that the above-mentioned before and after the failure 523 includes only two values of “before failure” and “after failure”, but before and after the failure 523 in this embodiment is “1 hour before failure” or “3 hours after failure”. You may have the maintenance cost and the time required for maintenance in a plurality of time zones after the failure, such as “after 6 hours and before 6 hours”. The maintenance countermeasure table 510 may have maintenance costs and required maintenance times in a plurality of time zones after a failure.

FIG. 6 is an explanatory diagram showing the output information 127 of the first embodiment.

The output information 127 includes at least a maintenance timing 531, a maintenance cost 532, an operation loss 533, a life loss 534, and a total cost loss 535. Further, the output information 127 may hold a processing result by the cost loss calculation unit 124 other than the information shown in FIG.

The maintenance timing 531 indicates a maintenance day. The maintenance timing 531 shown in FIG. 6 indicates the number of days that have elapsed since the occurrence of the abnormality, but may indicate the date and time of maintenance using a general date and time.

The maintenance cost 532 indicates a cost required to replace a part when the part is replaced at the maintenance timing 531. The operation loss 533 indicates an operation loss that occurs when parts are replaced at the maintenance timing 531. The life loss 534 indicates a life loss that occurs when a part is replaced at the maintenance timing 531. The total cost loss 535 indicates the total value of the maintenance cost 532, the operation loss 533, and the life loss 534.

The cost loss calculation unit 124 updates the maintenance countermeasure table 510 using the updated maintenance record table 520 when the maintenance record table 520 is updated.

The operation loss and life loss of Example 1 are described below. Operation loss and life loss are disadvantages experienced by users in maintaining parts.

The operational loss of the first embodiment is a profit lost by not operating the target device when maintaining the part, and a profit generated by the target device if the part is operating. The operation loss is calculated by the time required for maintenance and the profit per unit time generated by the target device.

The time required for maintenance before failure is generally shorter than the time required for maintenance after failure. For this reason, the operation loss is generally higher after the failure than before the failure.

The life loss in Example 1 is the value remaining in the part that is lost when the part is replaced. The life loss is zero after the part fails. On the other hand, it is a value that decreases according to the elapsed time since the component was installed before the component failed.

The maintenance cost and operation loss in Example 1 are values given by an index indicating that the disadvantage before failure is lower than that after failure. Moreover, the life loss in Example 1 is an index indicating that the disadvantage after failure is lower than that before failure. Note that the maintenance server 100 according to the first embodiment uses any index as long as it is an index indicating a disadvantage of different values before and after the failure, such as an index that gives maintenance cost, operation loss, and life loss. Also good.

FIG. 7 is a flowchart showing a process for calculating a maintenance cost, an operation loss, and a life loss according to the maintenance timing of the first embodiment.

At the start of the process shown in FIG. 7, the failure history 121, customer information 123, and maintenance information 125 are updated to the latest state. The processing illustrated in FIG. 7 may be started when the maintenance server 100 detects that an abnormality with a predetermined abnormality degree has occurred in the component of the target device.

Specifically, when the information acquisition unit 126 acquires information indicating the degree of abnormality equal to or greater than a predetermined threshold and the component for which the degree of abnormality is measured from the vehicle-mounted device 150, it is necessary to maintain the component of the target device. It may be determined that the error has occurred, and the abnormality estimation unit 122 may be instructed to start the process illustrated in FIG. Here, the information acquisition unit 126 inputs at least information that identifies a part that needs to be maintained and an abnormality that has occurred in the part to be maintained to the abnormality estimation unit 122.

Further, the process illustrated in FIG. 7 may be started when the information acquisition unit 126 is instructed from the user terminal 170. Here, the instruction to start the process shown in FIG. 7 includes information for identifying a part that needs to be maintained and an abnormality that has occurred in the part to be maintained.

Here, the information for identifying the part that needs to be maintained and the abnormality that has occurred in the part to be maintained may be an abnormality ID (corresponding to the abnormality ID 302), or an abnormality name and a component ID (abnormal name 301). And a component ID 304).

Furthermore, the processing shown in FIG. 7 may be started when the maintenance server 100 is activated. In this case, the information acquisition unit 126 holds in advance information for identifying a part that needs to be maintained and an abnormality that has occurred in the part to be maintained.

7 starts, the abnormality estimation unit 122 obtains a failure probability according to the time 305 by using the abnormality degree transition table 300 and the failure probability table 310 (S1001).

Specifically, in step S1001, the abnormality estimation unit 122 first identifies the abnormality name 301 from the information (information identifying the type of component and the type of abnormality) acquired when starting the processing illustrated in FIG. To do.

The abnormality estimation unit 122 uses the information identifying the abnormality acquired at the start of the process illustrated in FIG. 7, the abnormality name 301 and the abnormality ID 302 of the abnormality degree transition table 300, and the value of the abnormality name 301 (hereinafter referred to as abnormality). A) is specified. Further, the abnormality estimation unit 122 uses the information for identifying the component acquired at the start of the process illustrated in FIG. 7, the abnormality ID 302, the component name 303, and the component ID 304 of the abnormality degree transition table 300, and the value of the component name 303. (Hereinafter, component A) and the value of component ID 304 (hereinafter, component IDa) are specified.

Then, the abnormality estimation unit 122 extracts an entry in the abnormality degree transition table 300 having the abnormality A in the abnormality name 301. Then, the abnormality estimation unit 122 extracts the degree of abnormality 306 at the same time 305 from the extracted entries. Then, the abnormality estimation unit 122 calculates a statistical value such as an average value or an intermediate value of the extracted degree of abnormality 306.

Thus, by calculating the statistical value of the degree of abnormality 306 extracted every time 305, the abnormality estimation unit 122 can obtain the transition of the degree of abnormality 306 according to the time 305 in abnormality A. In the following, the abnormality estimation unit 122 obtains the failure probability according to the time 305 according to the transition of the abnormality degree 306 according to the time 305 and the failure probability table 310.

FIG. 8A is an explanatory diagram showing a transition 601 of the failure probability according to the degree of abnormality of the first embodiment.

FIG. 8A shows the abnormality degree 312 and failure probability 314 of the entry in the failure probability table 310 in which the abnormality name 311 indicates abnormality A. That is, FIG. 8A shows a failure probability transition 601 according to the degree of abnormality.

The horizontal axis of FIG. 8A is the degree of abnormality, and corresponds to the degree of abnormality 312. The vertical axis in FIG. 8A represents the failure probability and corresponds to the failure probability 314.

FIG. 8B is an explanatory diagram showing the transition 602 of the degree of abnormality according to the elapsed time of the first embodiment.

The horizontal axis of FIG. 8B is the elapsed time after the occurrence of abnormality, and corresponds to time 305. The vertical axis in FIG. 8B indicates the degree of abnormality. The vertical axis in FIG. 8B corresponds to the statistical value of the degree of abnormality 306 according to the time 305 calculated by the abnormality estimation unit 122 in step S1001. Therefore, FIG. 8B shows a transition 602 of the degree of abnormality 306 according to time 305 in abnormality A.

After obtaining the transition 602 of the degree of abnormality 306 as shown in FIG. 8B, the abnormality estimation unit 122 associates the horizontal axis of the transition 601 with the vertical axis of the transition 602 to change the failure probability according to the elapsed time. 603 is obtained.

FIG. 8C is an explanatory diagram showing a transition 603 of the failure probability according to the elapsed time of the first embodiment.

The horizontal axis of the transition 603 shown in FIG. 8C corresponds to the time 305. Further, the vertical axis of the transition 603 illustrated in FIG. 8C corresponds to the failure probability 314. In step S <b> 1001, the abnormality estimation unit 122 inputs the obtained transition 601, transition 602, and transition 603 to the cost loss calculation unit 124.

After step S1001, the Kosutorosu calculation unit 124 reads the argument _{T M} is a candidate of the maintenance timing in the memory 112, stores 0 as the initial value in the argument _{T M} (S1002). After step S1002, Kosutorosu calculating unit 124 adds 1 to the parameter _{T M} (S1003).

Here, the argument _TM is a maintenance timing candidate. In particular, the argument T _M of Example 1, in maintenance is required parts, the elapsed time from the abnormality occurs. In step S1002, the cost loss calculation unit 124 may store the transition 601, the transition 602, and the transition 603 in the output information 127. Accordingly, the output unit 128 can output any one of the transition 601, the transition 602, and the transition 603 toward the maintenance person or the like.

After step S1003, the cost loss calculation unit 124 calculates the maintenance cost in step S1004, calculates the operation loss in step S1006, and calculates the life loss in step S1008. Details of the processing are shown below.

In step S1004, the cost loss calculation unit 124 extracts an entry including the abnormality name 511 indicating the abnormality A and the component name 512 indicating the part A from the maintenance countermeasure table 510. Then, the cost loss calculation unit 124 extracts the pre-failure maintenance cost 513 and the post-failure maintenance cost 514 from the extracted entries.

Further Kosutorosu calculation unit 124, the elapsed time corresponding to the argument T _M at least one failure probability calculated in the previous elapsed time (horizontal axis transition 603) (ordinate transition 603), extracted post-fault maintenance costs Multiply each by 514. Further, Kosutorosu calculation unit 124, the argument T _M is the elapsed time indicated failure probability calculated in the time elapsed after the (horizontal axis transition 603) (ordinate transition 603), the pre-fault maintenance costs 513 extracted Multiply each.

This will replace parts maintenance timing argument T _M, if maintenance timing argument T _M is thereafter failed, maintenance costs required for replacement is for a post-fault maintenance costs 514. Further, to replace parts maintenance timing argument T _M, if before the failure the maintenance timing of the argument T _M, maintenance costs required for replacement is for a pre-fault maintenance costs 513.

As a result of multiplying the failure probability by the maintenance cost before failure 513 or the maintenance cost after failure 514, the cost loss calculation unit 124 obtains a distribution 701 described later. The Kosutorosu calculation unit 124 updates the parameter _{T M} in step S1012 (Fig. 10) and step S1003 to be described later, repeat step S1004. Thereby, the cost loss calculation unit 124 calculates the distribution 701 at a plurality of maintenance timings (a plurality of arguments T _M ). The distribution 701 can then be used to calculate a disadvantage predicted value.

FIG. 9A is an explanatory diagram showing a distribution 701 as a result of multiplying the failure probability and the maintenance cost of the first embodiment.

Distribution shown in FIG. 9A 701 is an example of an output result in step S1004 if the argument _{T M} is 33. According to Figure 9A, the results with parts multiplies and maintenance costs and failure probability when the previously failed maintenance timing argument T _M is maintenance costs when part has failed just after the maintenance timing of the argument T _M Higher than the result of multiplying the failure probability.

This failure probability of the previous maintenance timing argument T _M is maintenance cost of less than the failure probability of the previous maintenance timing argument T _M, while the the failure immediately before the maintenance timing of the argument T _M generated (post-fault maintenance costs 514) is larger than the maintenance costs when a failure immediately after the maintenance timing of the argument T _M occurs (failure before maintenance costs 513).

Furthermore, the cost loss calculation unit 124 calculates an operation loss in step S1006. Specifically, the cost loss calculation unit 124 extracts an entry including the abnormality name 511 indicating the abnormality A and including the component name 512 indicating the part A from the maintenance countermeasure table 510. Then, the cost loss calculation unit 124 extracts the pre-failure maintenance required time 515 and the post-failure maintenance required time 516 from the extracted entries.

In step S1006, the cost loss calculation unit 124 extracts an entry in the operation result table 410 including the component ID 412 indicating the component IDa, and specifies the machine ID 411 of the extracted entry. Then, the cost loss calculation unit 124 extracts an entry of the production performance table 400 including the machine ID 403 corresponding to the specified machine ID 411. Then, the cost loss calculation unit 124 multiplies the extracted output 404 per unit time by the product unit price 405, and calculates the profit per unit time of the target device in which the part indicated by the part IDa is installed.

Furthermore, Kosutorosu calculation unit 124, in step S1006, (horizontal axis transition 603) the elapsed time indicated by the argument _{T M} to at least one of the failure probability in the previous elapsed time (vertical axis transition 603), after extracted fault maintenance The required time 516 is multiplied by the calculated profit per unit time. Further, Kosutorosu calculation unit 124, the argument T _M is the elapsed time indicated failure probability calculated in the time elapsed after the (horizontal axis transition 603) (ordinate transition 603), extracted failure before maintenance time required 515 And the calculated profit per unit time.

The target device stops its operation for the required maintenance time or fails to operate normally if its own parts break down. For this reason, during the time required for maintenance, a profit that the target device should generate is not generated, which causes a disadvantage for the user.

For this reason, the operation loss in the first embodiment is a profit to be generated by the target device, which is damaged by the maintenance. The operation loss in the first embodiment is a result of multiplying the maintenance required time 515 before the failure or the maintenance required time 516 after the failure by the profit per unit time.

As a result of step S1006, the cost loss calculation unit 124 calculates a distribution 702, which will be described later, as a distribution related to operation loss. The Kosutorosu calculation unit 124 updates the parameter T _M by processing described later, by repeating the steps S1006, it calculates a distribution 702 of the operational losses in the plurality of maintenance timing.

FIG. 9B is an explanatory diagram illustrating a distribution 702 obtained by multiplying the failure probability and the operation loss according to the first embodiment.

Distribution shown in FIG. 9B 702 is an example of an output result in step S1006 if the argument _{T M} is 33. According to Figure 9B, the result obtained by multiplying the operational loss and failure probability when a part immediately before the elapsed time argument T _M fails, similarly to the distribution 701, the parts immediately after the elapsed time parameter T _M It is higher than the result of multiplying the operation loss and the failure probability when a failure occurs.

The cost loss calculation unit 124 can add the profit that should have been obtained by the user according to the maintenance timing to the predicted value of the disadvantage by obtaining the distribution 702.

In the case where maintenance measures table 510 has a maintenance costs and maintenance time required in a plurality of time zone after a failure, in S1004 and S1006, Kosutorosu calculation unit 124, a time zone after the failure corresponding to the elapsed time parameter T _M Using the maintenance cost, the required maintenance time, and the failure probability, the expected value of the maintenance cost and the operation loss after the failure may be calculated.

Cost loss calculation unit 124 calculates a life loss in step S1008. Here, the cost loss calculation unit 124 sets 0 as a life loss that occurs after a failure. Then, the cost loss calculation unit 124 extracts the entry of the part price table 500 including the part name 501 indicating the part A, and extracts the part unit price 502 of the part A.

Further, the cost loss calculation unit 124 extracts the operation result table 410 including the component ID 412 indicating the component IDa, and extracts the component hour meter 413 of the component when an abnormality occurs. The Kosutorosu calculation unit 124 calculates the life loss before the failure using the extracted value and the argument T _M and the following equation (1).

Life Loss Before Failure = Parts Unit Price 502 / (Part Hour Meter 413 at Time of Abnormality + Time Elapsed After Abnormality (: Argument T _M )) (1)

Formula (1) is a formula that divides the component unit price 502 by the time elapsed since the component was installed in the target device. However, the time elapsed since the part was installed in the target device is a positive number that is not zero. The equation for calculating the life loss before failure may be any equation as long as it calculates a value that decreases according to the elapsed time.

Furthermore, Kosutorosu calculation unit 124, in step S1008, the argument _{T M} is the elapsed time indicated at least one failure probability calculated in the previous elapsed time (horizontal axis transition 603) (ordinate transition 603), after a failure Multiply by 0, which is the lifetime loss. Further, Kosutorosu calculation unit 124, the argument T _M is the elapsed time indicated failure probability calculated in the time elapsed after the (horizontal axis transition 603) (ordinate transition 603), multiplying each life loss before failure .

As a result of step S1008, the cost loss calculation unit 124 calculates a distribution 703, which will be described later, as a distribution related to the life loss. The Kosutorosu calculation unit 124 updates the parameter T _M by processing described later, by repeating the steps S1008, it calculates a distribution 703 of the lifetime loss in multiple maintenance timing.

FIG. 9C is an explanatory diagram illustrating a distribution 703 obtained by multiplying the failure probability and the life loss in the first embodiment.

Distribution shown in FIG. 9C 703 is an example of an output result in step S1008 if the argument _{T M} is 33. According to FIG. 9C, if the part has failed life loss immediately before the elapsed time parameter T _M 0. By calculating the distribution 703, the cost loss calculation unit 124 can add the value that still remains to the parts during maintenance to the predicted value of disadvantage.

When the distribution 701, distribution 702, and distribution 703 are obtained in step S1004, step S1006, and step S1008, the cost loss calculation unit 124 may store the distribution 701, distribution 702, and distribution 703 in the output information 127. Then, the output unit 128 may output the distribution 701, the distribution 702, and the distribution 703 to the user terminal 170.

Cost loss calculation unit 124 executes step S1005 after step S1004, executes step S1007 after step S1006, and executes step S1009 after step S1008.

FIG. 10 is a flowchart illustrating a process for calculating a predicted value of a disadvantage in the maintenance of the first embodiment.

Before starting any of Step S1005, Step S1007, and Step S1008, the cost loss calculation unit 124 stores the expected value of maintenance cost, the expected value of operation loss, and the expected value of life loss. A new entry A is generated. The Kosutorosu calculation unit 124, the maintenance timing 531 of the new entry, and stores the value of the argument _{T M.}

Kosutorosu calculation unit 124, in step S1005, by determining the total value of the distribution 701 in the argument _{T M,} calculates the expected value of the maintenance costs. In addition, the cost loss calculation unit 124 stores the expected value of the calculated maintenance cost in the maintenance cost 532 of the entry A.

Further, Kosutorosu calculation unit 124, in step S1007, by determining the total value of the distribution 702 in the argument _{T M,} calculates the expected value of the operational loss. Then, the cost loss calculation unit 124 stores the calculated expected value of the operation loss in the operation loss 533 of the entry A.

Further, Kosutorosu calculation unit 124, in step S1009, by determining the total value of the distribution 702 in the argument _{T M,} calculates the expected value of life loss. Then, the cost loss calculation unit 124 stores the calculated expected value of the life loss in the life loss 534 of the entry A.

Step S1005, after the step S1007 and step S1009, Kosutorosu calculation unit 124, maintenance costs 532 entries A maintenance timing 531 is argument _{T M} output information 127, the total value of the operational losses 533 and lifetime loss 534, not Calculated as the predicted value of profit. Then, the cost loss calculating unit 124 stores the calculated predicted value in the cost loss total 535 of the entry A (S1010).

Through the processing up to step S1010, the maintenance server 100 reflects the disadvantages that may occur when the maintenance timing is before the failure and the disadvantages that may occur when the maintenance timing is after the failure based on the failure probability. Therefore, a more accurate predicted value can be calculated for each maintenance timing.

Note that the cost loss calculation unit 124 may calculate the predicted value of the disadvantage after multiplying the maintenance cost 532, the operation loss 533, and the life loss 534 by weights specified in advance in step S1010. As a result, the cost loss calculation unit 124 can calculate a predicted disadvantage value that can be evaluated with emphasis on the expected value that the user attaches importance to.

After step S1010, Kosutorosu calculation unit 124, the total value of the failure probability from the past to the argument _{T M} in transition 603 is calculated as the cumulative probability (S1011). After step S1011, for example, when the failure probability is expressed as a percentage (percent), the cost loss calculation unit 124 determines whether the cumulative probability is 100 (S1012).

In the determination processing in step S1012, the cost loss calculation unit 124 executes step S1013 when the expected values of the maintenance cost, the operation loss, and the life loss are calculated based on all failure probabilities other than 0 in the transition 603, and the maintenance cost If the expected values of operation loss and life loss are not calculated, any determination process may be used as long as the process returns to step S1003.

If the cumulative probability in the above example of step S1012 is not 100, Kosutorosu calculating unit 124 adds 1 to the parameter _{T M} returns to step S1003. When the cumulative probability is 100, the cost loss calculation unit 124 extracts the minimum value from the predicted values calculated in step S1010. The Kosutorosu calculation unit 124, the argument _{T M} of the extracted minimum value, determined recommended maintenance timing (S1013), stores the recommended maintenance timing output information 127.

After step S1013, the cost loss calculation unit 124 ends the process shown in FIG.

FIG. 11A is an explanatory diagram illustrating processing results 801 of the maintenance cost 802, the operation loss 803, and the life loss 804 at the maintenance timing of the first embodiment.

The processing result shown in FIG. 11A shows the results of step S1005, step S1007, and step S1009 shown in FIG. Maintenance costs 802 shown in FIG. 11A is an output example of displaying the expected value of the resulting maintenance costs by step S1005 shown in FIG. 10 according to the argument _{T M.} Operation Ross 803 shown in FIG. 11A is an output example of displaying the expected value of the obtained operational loss by step S1007 shown in FIG. 10 according to the argument _{T M.} Life loss 804 shown in FIG. 11A is an output example of displaying the expected value of the obtained lifetime loss in step S1009 shown in FIG. 10 according to the argument _{T M.}

The maintenance cost 802 and the operation loss 803 are larger values as the argument T _M (the elapsed time from occurrence of an abnormality and the maintenance timing) increases. Moreover, life loss 804 larger argument _{T M} is a small value.

The output unit 129 displays the processing result 801 shown in FIG. 11A on the user terminal 170 based on the information stored in the output information 127, so that the maintenance person and the user can maintain the maintenance cost 802, the operation loss 803, and the life loss. After referring to the change 804, the maintenance timing of the part can be determined. In addition, the maintenance person and the user can recognize the breakdown of the disadvantage due to the maintenance timing.

FIG. 11B is an explanatory diagram showing a predicted value of disadvantage at the maintenance timing of the first embodiment.

The disadvantage 806 of the processing result 805 shown in FIG. 11B is the result of step S1010 shown in FIG. Disadvantage 806, maintenance costs 802 at the same arguments _{T M,} the total value of the operational losses 803 and lifetime loss 804.

The disadvantage 806 shown in FIG. 11B includes a minimum value. While this maintenance costs 802 and operational loss 803 increases as the argument T _M is increased, because the life loss 804 is reduced. Therefore, the argument T _M at the minimum value of the penalty 806, operational loss 803, when considering all operational losses 803 and lifetime loss 804 shows the most disadvantaged less maintenance timing by maintenance.

The output unit 129 displays the processing result 801 illustrated in FIG. 11A on the user terminal 170, so that the maintenance person and the user can maintain the expected maintenance cost value, the expected operation loss value, and the expected life loss value for each maintenance timing. By recognizing the total, it is possible to recognize a predicted value of a disadvantage that may occur due to the maintenance timing. The maintenance person and the user can determine an appropriate maintenance timing based on the recognized degree of disadvantage.

FIG. 12A is an explanatory diagram illustrating a screen 810 that displays a disadvantage due to maintenance after a failure according to the first embodiment.

The output unit 128 generates and outputs data on the screen 810 in order to express to the user the disadvantage that occurs when maintenance is performed. A screen 810 is an example of a screen output to the user terminal 170 by the output unit 128.

The screen 810 includes a post-maintenance screen 811 and a predictive maintenance screen 812. When a maintenance person or the like selects one of the two tabs displayed on the screen 810, the post-maintenance screen 811 or the predictive maintenance screen 812 corresponding to the selected tab may be displayed with priority. Further, both the post-maintenance screen 811 and the predictive maintenance screen 812 may be displayed on the screen 810.

The post-maintenance screen 811 shows a user's disadvantage that occurs when a part is maintained after the part has failed. The post-maintenance screen 811 includes a failure probability 813 and a disadvantage list 814.

The failure probability 813 is a result of step S1001 shown in FIG. 7, and shows a transition 603 shown in FIG. 8C. The disadvantage list 814 includes the post-failure maintenance cost used in step S1004 shown in FIG. 7 (corresponding to the post-failure maintenance cost 514), the post-failure operation loss used in step S1006, and the fault used in step S1008. Indicates post-life loss. The disadvantage list 814 indicates the total value of the maintenance cost after failure, the operation loss after failure, and the life loss after failure.

FIG. 12B is an explanatory diagram illustrating a screen 810 that displays expected values of a plurality of indexes according to the maintenance timing of the first embodiment and a predicted value of disadvantage.

Prediction maintenance screen 812 shows the expected value of the maintenance costs in accordance with the parameter T _M, the expected value of the operational loss, expected value of life loss, and the predicted value of disadvantages. The predictive maintenance screen 812 includes a cost loss screen 821, a disadvantage screen 822, and a details screen 823.

The cost loss screen 821 shows the results of step S1005, step S1007, and step S1009 shown in FIG. 10, and is the same as the processing result 801 shown in FIG. 11A. The disadvantage screen 822 shows the result of step S1010 shown in FIG. 10 and is the same as the processing result 805. Detail screen 823 shows the expected value of the maintenance costs of the argument T _M, the expected value of the operating loss, the sum of the expected values and their expected values of life loss.

The output unit 128, the detail screen 823, the maintenance cost of the argument T _M which disadvantages 806 becomes the minimum value may be displayed operational loss and life loss. Thereby, the output unit 128 can provide the user with the maintenance timing with the smallest predicted value of the disadvantage due to maintenance as the recommended maintenance timing.

The output unit 128 may display information other than the contents of the screen 810 shown in FIGS. 12A and 12B as long as the information is stored in the maintenance server 100. In particular, the output unit 128 may display the distributions 701 to 703 and the like shown in FIGS. 9A to 9C on the user terminal 170.

FIG. 13A is an explanatory diagram illustrating an application example of the maintenance system according to the first embodiment.

FIG. 13A shows an example of roles of a user and a maintenance person who use the maintenance system of the first embodiment. The maintenance system according to the first embodiment is used by a service provider 901, a user 902, a manufacturer 903, and a maintenance company 904.

The user 902 is a user who uses target devices such as the excavator 140 and the truck 130. The user 902 may be a mining company or a construction company, for example. Further, the user 902 may be a production company having a factory as long as the target device is a generator or the like. Further, if the target device is a computer or storage, it may be a company that provides a service by a computer system.

User 902 determines the maintenance timing of the target device. The user 902 provides the content of the customer information 123 to the service provider 901.

Maker 903 is a person who manufactured the target device. The manufacturer 903 provides the service provider 901 with the content of the failure history 121 (particularly, the abnormality degree transition table 300) and the content of the maintenance information 125 (particularly the parts price table 500).

The maintenance company 904 is a maintainer of the target device. The maintenance business operator 904 provides the service provider 901 with the content of the maintenance information 125 (particularly, the maintenance result table 520).

The service provider 901 holds the maintenance server 100. Based on the information received from the user 902, the manufacturer 903, and the maintenance company 904, the service provider 901 determines the maintenance cost, the expected value of the operation loss and the life loss, the predicted value of the disadvantage, the recommended maintenance timing, etc. The maintenance server 100 is made to calculate. Then, the service provider 901 provides the calculated result to the user 902.

FIG. 13B is a sequence diagram illustrating the operation of the first embodiment.

User 902 provides the contents of customer information 123 to service provider 901 (911). The manufacturer 903 provides the content of the failure history 121 and the content of the maintenance information 125 to the service provider 901 (912). The maintenance company 904 provides the content of the maintenance information 125 to the service provider 901 (913).

After the

sequences

911, 912, and 913, the maintenance server 100 of the service provider 901 calculates the expected value of maintenance cost, operation loss and life loss, and predicted value of disadvantage by the processing shown in FIGS. The recommended maintenance timing is determined (914). After the sequence 914, the service provider 901 provides the calculated expected value and the determined recommended maintenance timing to the user 902 (915).

The service provider 901 may provide the user 902 with the calculated expected value and predicted value, the determined recommended maintenance timing, and the user 902 by causing the output device such as the user terminal 170 to output the processing result. Further, the maintenance server 100 may output the processing result to a portable storage medium, and the service provider 901 may send the storage medium to the user 902.

The user 902 determines the maintenance timing of the parts based on the provided expected value and the recommended maintenance timing (916).

According to the first embodiment, the maintenance server 100 acquires the failure probability in the elapsed time after the occurrence of the abnormality as the actual value related to the lifetime, and performs maintenance according to the maintenance timing using the acquired failure probability. Calculate the predicted value of the disadvantage that will occur. By using the failure probability in this way, it is possible to calculate a more accurate predicted value of the disadvantage in accordance with the actual operation status of the parts. Then, by providing the predicted value calculated by the maintenance server 100 to the user, information for making an appropriate maintenance plan can be provided to the user.

Also, the maintenance server 100 determines a recommended maintenance timing based on the calculated predicted value and provides it to the user. This allows the user to develop an appropriate maintenance plan that can minimize the disadvantages they suffer.

Furthermore, the maintenance server 100 calculates a failure probability based on the elapsed time and the degree of abnormality since the abnormality occurred. Thereby, the maintenance server 100 can calculate a failure probability using the result measured by the sensor of the vehicle-mounted device 150, and can calculate a more accurate failure probability.

In Example 1, the failure probability was calculated according to the elapsed time after the occurrence of the abnormality. In the second embodiment, unlike the first embodiment, the failure probability is calculated according to the elapsed time after the component is installed in the target device.

The maintenance system of the second embodiment has the same configuration as the maintenance system shown in FIG. The maintenance server 100 according to the second embodiment includes the same physical devices and functional units as the maintenance server 100 according to the first embodiment. However, the failure history 121 of the second embodiment is different from the failure history 121 of the first embodiment. Moreover, the process of the abnormality estimation part 122 of Example 2 and the process of the abnormality estimation part 122 of Example 1 are different.

FIG. 14 is an explanatory diagram showing a failure history 121 according to the second embodiment.

The failure history 121 of the second embodiment includes an hour meter failure probability table 540 instead of the abnormality degree transition table 300 and the failure probability table 310. The hour meter failure probability table 540 includes a component name 541, a component hour meter 542, the number of failures 543, and a failure probability 544.

The part name 541 indicates the type of part. The component hour meter 542 indicates an elapsed time since the component was installed in the target device.

The number of failures 543 indicates the number of failures of the type of component indicated by the component name 541 in the elapsed time indicated by the component hour meter 542. The failure probability 544 is a probability that a component of the type indicated by the component name 541 will fail in the elapsed time indicated by the component hour meter 542.

The information acquisition unit 126 according to the second embodiment displays the component name, the component hour meter, and the number of failures held by the vehicle-mounted device 150 from the vehicle-mounted device 150 or the user terminal 170 in a predetermined cycle or when a user instruction is given. To get to. Then, the information acquisition unit 126 extracts an entry in the hour name failure probability table 540 of the component name 541 and the component hour meter 542 corresponding to the acquired information, and the number of failures included in the collected information is included in the failure number 543 of the extracted entry. Is added.

FIG. 15 is a flowchart illustrating a process for calculating a predicted value of a disadvantage in the maintenance of the second embodiment.

The process shown in FIG. 15 is started under the same conditions as the process shown in FIG. That is, the process shown in FIG. 15 is started when the maintenance server 100 detects that an abnormality with a predetermined abnormality level has occurred in the component of the target device or when instructed by the user terminal 170.

However, the information acquisition unit 126 according to the second embodiment includes information for identifying a component that needs to be maintained at the start of the process illustrated in FIG. 15 and the component hour meter at the time when the process illustrated in FIG. 15 is started. Get the value. Specifically, the information acquisition unit 126 may acquire the value of the hour meter from the vehicle-mounted device 150 connected to the component that needs to be maintained at the start of the process illustrated in FIG.

The information for identifying the part that needs to be maintained is information that can identify the type of the part that needs to be maintained (hereinafter, part A) and the part that needs to be maintained (hereinafter, part IDa). including.

And the information acquisition part 126 of Example 2 inputs the information acquired at the time of the start of the process shown in FIG. When the process illustrated in FIG. 15 is started and information is input from the information acquisition unit 126, the abnormality estimation unit 122 calculates the failure probability 544 of the part A in the hour meter failure probability table 540. Then, the abnormality estimation unit 122 extracts the failure probability distribution of the part A in the period indicating the future from the input hourmeter value from the calculated failure probability 544.

Specifically, the abnormality estimation unit 122 calculates the total value of the failure number 543 of the entry whose part name 541 indicates the part A, and calculates the ratio of the number of the failure number 543 to the calculated total value for each entry. To do. Thereby, the abnormality estimation unit 122 calculates the failure probability of each entry, and updates the failure probability 544 with the calculated failure probability.

FIG. 16A is an explanatory diagram showing a failure probability distribution 613a according to the component hour meter 542 of the second embodiment.

In the distribution 613a shown in FIG. 16A, the horizontal axis corresponds to the component hour meter 542 of the hour meter failure probability table 540. The vertical axis corresponds to the failure probability 544 of the hour meter failure probability table 540.

Distribution 613a indicates that the component hour meter is low and the failure probability is high immediately after the component is installed. This indicates that initial failure is likely to occur in the part. The distribution 613a indicates that the probability of failure is high when the component hour meter is high and a long time has elapsed since the component was installed.

After generating the hour meter failure probability table 540, the abnormality estimation unit 122 extracts a component hour meter 542 that is equal to or greater than the value of the hour meter acquired at the start of the processing shown in FIG. 15 and the failure probability 544 in step S1101. Accordingly, the abnormality estimation unit 122 extracts the distribution 613b illustrated in FIG. 16B.

FIG. 16B is an explanatory diagram illustrating a future failure probability distribution 613b according to the second embodiment.

In the distribution 613b, the horizontal axis corresponds to the component hour meter 542 and the vertical axis corresponds to the failure probability 544, as in the distribution 613a. The horizontal axis shown in FIG. 16B may indicate a relative time based on the hour meter value acquired at the start of the process shown in FIG. The distribution 613b shows a distribution obtained by extracting the failure probability 544 of the component hour meter 542 higher than the value of the hour meter acquired at the start of the process shown in FIG. 15 from the distribution 613a.

The abnormality estimation unit 122 inputs the distribution 613b to the cost loss calculation unit 124. The cost loss calculation unit 124 executes Steps S1002 to S1013 shown in FIGS. 7 and 10 using the distribution 613b (Step S1102). After step S1102, the cost loss calculation unit 124 ends the process illustrated in FIG.

According to the second embodiment, the failure probability is calculated from the value of the part hour meter, and the expected value is calculated. Thereby, it is not necessary to obtain the relationship between the elapsed time and the number of failures based on the combination of the elapsed time and the degree of abnormality and the combination of the degree of abnormality and the number of failures, and the processing can be reduced.

The operational loss calculated in Example 1 occurred when the target device stopped when the component was replaced because the target device on which the component was installed was not made redundant. However, if the target device is made redundant and the target device does not stop when the parts are maintained, the operation loss is reduced. The maintenance server 100 according to the third embodiment accurately calculates an operation loss by acquiring information related to other target devices that make the target device redundant. The maintenance system of the third embodiment has the same configuration as the maintenance system shown in FIG. The maintenance server 100 according to the third embodiment includes the same physical devices and functional units as the maintenance server 100 according to the first embodiment.

However, the contents of the customer information 123 of the third embodiment and the customer information 123 of the first and second embodiments are different. Further, the processing of the cost loss calculation unit 124 of the third embodiment is different from the processing of the cost loss calculation unit 124 of the first and second embodiments.

FIG. 17 is an explanatory diagram illustrating a surplus table 420 included in the customer information 123 according to the third embodiment.

Customer information 123 of the third embodiment includes a surplus table 420 in addition to the production performance table 400 and the operation performance table 410 of the first embodiment. The surplus table 420 indicates the number of other target devices that make the target device redundant.

The surplus table 420 may be transmitted from the in-vehicle device 150 in a predetermined cycle, or may be input via the user terminal 170 by the maintenance person. The surplus table 420 includes a machine ID 421, a date and time 422, and a surplus number 423.

The maintenance server 100 of the third embodiment may have the customer information 123 of the first embodiment or the customer information 123 of the second embodiment.

The machine ID 421 corresponds to the machine ID 403 of the production result table 400 and uniquely indicates the target device. The date and time 422 may indicate a day, or may indicate a date and time. The date and time 422 may indicate a day every day, a day every week, or a time every hour.

The surplus number 423 indicates the number of other target devices that make the target device indicated by the machine ID 421 redundant. When the surplus number 423 is 0, the target device indicated by the machine ID 421 is not made redundant.

FIG. 18 is a flowchart showing a process for calculating a predicted value of disadvantage in the maintenance of the third embodiment.

18 is started under the same conditions as the conditions for starting the process shown in FIG. 7 of the first embodiment and the process shown in FIG. 15 of the second embodiment. The information acquired by the information acquisition unit 126 of the third embodiment is also acquired at the start of the process illustrated in FIG. 7 of the first embodiment or at the start of the process illustrated in FIG. 15 of the second embodiment. Same as information.

Note that, at the start of the processing shown in FIG. 18, the abnormality estimation unit 122 acquires at least a component ID (component IDa) of a component that needs to be maintained.

When the process shown in FIG. 18 starts, step S1001 shown in FIG. 7 of the first embodiment or step S1101 shown in FIG. 15 of the second embodiment is executed (S1401).

After step S1401, the cost loss calculation unit 124 executes step S1002 shown in FIG. 7 (S1402), and executes step S1003 shown in FIG. 7 (S1403). Further, after step S1403, the cost loss calculation unit 124 executes steps S1004 and S1005 (S1404), while executing steps S1008 and S1009 (S1409).

Further, the cost loss calculation unit 124 executes step S1006 after step S1403 (S1405). After step S1405, the cost loss calculation unit 124 extracts an entry of the component ID 412 indicating the component IDa acquired at the start of the process illustrated in FIG. 18 from the operation result table 410, and identifies the machine ID 411 of the extracted entry.

Then, the cost loss calculation unit 124 extracts at least one entry of the machine ID 421 indicating the specified machine ID 411 from the surplus table 420. The cost loss calculation unit 124 specifies the surplus number 423 of the entry with the newest date and time 422 from the entries extracted from the surplus table 420. This is because the number of target devices that make the target devices redundant may change from day to day.

Then, the cost loss calculation unit 124 determines whether or not the specified surplus number 423 is other than 0 (S1405). When the specified surplus number 423 is 0, the cost loss calculation unit 124 executes Step S1007 because the target device is not made redundant and there is no need to reduce the operation loss (S1408).

If the specified surplus number 423 is other than 0, the cost loss calculation unit 124 determines that the target device is made redundant, and multiplies the operation loss calculated in step S1006 by 0, thereby setting 0 as the operation loss. Calculate (step S1407). Here, if the target device a redundant target device is given a possible operation time, Kosutorosu calculation unit 124, the time argument T _M Only target device redundancy corresponding with operable period, operational You may calculate 0 as a loss.

After step S1407, the cost loss calculation unit 124 executes step S1007 (S1408). After step S1404, step S1408, and step S1409, the cost loss calculation unit 124 executes step S1010 and step S1011 (S1410), executes step S1012 (S1411), and executes step S1013 (S1412). After step S1412, the cost loss calculation unit 124 ends the process illustrated in FIG.

In addition, although the case where the target device is made redundant is described above, when the component is made redundant in the target device, the surplus table 420 may have a component ID instead of the machine ID 421. In step S1406, the cost loss calculation unit 124 may determine whether or not the component IDa is made redundant using the surplus table 420. If the component IDa is made redundant, step S1407 may be executed.

Further, the above-described cost loss calculation unit 124 calculates the operation loss as 0 when the target device (main device) is made redundant. However, if the alternative target device (sub device) that operates during maintenance has a smaller production amount per unit time than the main device, the cost loss calculation unit 124 decreases by operating the sub device in step S1407. The obtained profit may be calculated as an operational loss.

Specifically, in step S1407, the cost loss calculation unit 124 calculates the difference in profit per unit time between the main device and the sub device, and divides the calculated difference by the profit per unit time of the main device. The cost loss calculation unit 124 multiplies the result distribution 702 in step S1006 by the divided result. Then, the cost loss calculation unit 124 adds the total value of the distribution 702 after the multiplication as an expected value of operation loss in step S1007. In this case, the operation loss and customer information 123 may hold the profit per unit time of the secondary device.

According to the third embodiment, by calculating the operation loss according to whether or not the target device is made redundant, it is possible to calculate an accurate expected value of the operation loss corresponding to the state of the target device or part. Then, the cost loss calculating unit 124 can accurately calculate the predicted value of disadvantage.

In addition, this invention is not limited to the above-mentioned Example, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described.

Further, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

Also, each of the above-described configurations, functions, processing units, processing procedures, etc. may be realized in hardware by designing a part or all of them, for example, with an integrated circuit. Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as a program, a table, or a file that realizes each function can be stored in a recording device such as a memory, a hard disk, or an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

Also, the control lines or information lines indicate what is considered necessary for the explanation, and not all control lines or information lines on the product are necessarily shown. In practice, almost all the components are connected to each other.

Claims

A maintenance support system,
A processor and memory;
The memory includes a first disadvantage index indicating a first disadvantage degree and a second disadvantage degree indicating a second disadvantage degree among disadvantages experienced by a user due to maintenance of components mounted on the device. Stores disadvantageous indicators and
The first disadvantage index is such that a disadvantage that occurs before a failure of the component is lower than a disadvantage that occurs after the failure of the component;
The second disadvantage index is such that a disadvantage that occurs before a failure of the component is higher than a disadvantage that occurs after the failure of the component;
The processor is
Obtaining a plurality of failure probabilities of the part at a plurality of points in time during which the part is used;
The component uses a predicted value of a disadvantage that occurs when the part is maintained using the plurality of acquired failure probabilities, the first disadvantage index, and the second disadvantage index. Calculated at a plurality of candidate timings during which maintenance is performed,
A maintenance support system, wherein the calculated plurality of predicted values are output.
The maintenance support system according to claim 1,
The processor is
Selecting one candidate timing from the plurality of candidate timings;
At each time point before the selected candidate timing, obtain a value obtained by multiplying the failure probability by the first disadvantage index after failure, and at each time point after the selected candidate timing, Obtaining a value obtained by multiplying the failure probability by the first disadvantage index before failure, and calculating a sum of the multiplied values as a first expected value at the candidate timing;
At each time point before the selected candidate timing, obtain a value obtained by multiplying the failure probability by the second disadvantage index after the failure, and at each time point after the selected candidate timing, Obtaining a value obtained by multiplying the failure probability by the second disadvantage index before failure, and calculating a sum of the multiplied values as a second expected value at the candidate timing;
The sum of the value based on the calculated first expected value and the value based on the calculated second expected value is used as a predicted value of a disadvantage that occurs when the part is maintained at the selected candidate timing. A maintenance support system characterized by calculating.
The maintenance support system according to claim 1,
The second disadvantage index is
In the period before the occurrence of the failure, the unit price of the component and the candidate timing are used to give a value that gradually decreases with time,
A maintenance support system characterized in that 0 is given in a period after the occurrence of the failure.
The maintenance support system according to claim 1,
The first disadvantageous index gives a value obtained by multiplying a maintenance time necessary for maintaining the part where the failure has occurred and a profit per unit time generated by a device on which the part is mounted. ,
The maintenance support system is characterized in that the maintenance time is short in a period before the occurrence of the failure and long in a period after the occurrence of the failure.
The maintenance support system according to claim 4,
The processor is
Selecting one candidate timing from the plurality of candidate timings;
Obtaining information indicating whether the component is redundant;
At each time point before the selected candidate timing, obtain a value obtained by multiplying the failure probability by the first disadvantage index after failure, and at each time point after the selected candidate timing, Obtaining a value obtained by multiplying the failure probability by the first disadvantage index before failure, and calculating a sum of the multiplied values;
When the component is made redundant, a value lower than the calculated total is set as the first expected value,
At each time point before the selected candidate timing, obtain a value obtained by multiplying the failure probability by the second disadvantage index after the failure, and at each time point after the selected candidate timing, Obtaining a value obtained by multiplying the failure probability by the second disadvantage index before the failure, and calculating a sum of the multiplied values as a second expected value at the candidate timing;
The sum of the set first expected value and the calculated second expected value is calculated as a predicted value of a disadvantage that occurs when the part is maintained at the selected candidate timing. Maintenance support system.
The maintenance support system according to claim 1,
The processor is
The actual value of the cost incurred in the past by performing maintenance before the failure of the part is set to the disadvantage before the failure indicated by the first disadvantage index,
A maintenance support system, wherein an actual value of expenses generated in the past by performing maintenance after failure of the part is set to a disadvantage after failure indicated by the first disadvantage index.
The maintenance support system according to claim 1,
The processor is
Obtaining a failure record indicating the occurrence of a past failure of the component at the plurality of time points;
A maintenance support system, wherein the plurality of failure probabilities are acquired by calculating failure probabilities at the plurality of points in time using the acquired failure records.
The maintenance support system according to claim 1,
The processor is
Get the sensor value that measured the part,
By calculating failure probabilities at the plurality of time points based on abnormality degree information including a sensor value after the abnormality has occurred in the component and a failure occurrence state of the component at the sensor value. A maintenance support system characterized by acquiring a failure probability.
The maintenance support system according to claim 1,
The processor is
Select the point in time when the predicted value of the disadvantage is minimum,
A maintenance support system that outputs the selected time point as a recommended maintenance timing.
A maintenance support method by a maintenance support system,
The maintenance support system includes a processor and a memory,
The memory includes a first disadvantage index indicating a first disadvantage degree and a second disadvantage degree indicating a second disadvantage degree among disadvantages experienced by a user due to maintenance of components mounted on the device. Stores disadvantageous indicators and
The first disadvantage index is such that a disadvantage that occurs before a failure of the component is lower than a disadvantage that occurs after the failure of the component;
The second disadvantage index is such that a disadvantage that occurs before a failure of the component is higher than a disadvantage that occurs after the failure of the component;
The maintenance support method includes:
The processor obtains a plurality of failure probabilities of the part at a plurality of points in time during which the part is used;
The processor uses the acquired plurality of failure probabilities, the first disadvantage index, and the second disadvantage index to predict predicted values of disadvantages that occur when the component is maintained, Calculate at a plurality of candidate timings during which maintenance is performed during the period in which the part is used,
A maintenance support method, wherein the processor outputs the calculated predicted values.
The maintenance support method according to claim 10,
The processor selects one candidate timing from the plurality of candidate timings;
The processor obtains a value obtained by multiplying the failure probability by the first disadvantage index after a failure at each time point before the selected candidate timing, and each processor after the selected candidate timing. At the time point, a value obtained by multiplying the failure probability by the first disadvantage index before the failure is obtained, and a sum of the multiplied values is calculated as a first expected value at the candidate timing,
The processor obtains a value obtained by multiplying the failure probability by the second disadvantage index after failure at each time point before the selected candidate timing, and each after the selected candidate timing. At the time point, a value obtained by multiplying the failure probability by the second disadvantage index before the failure is obtained, and a sum of the multiplied values is calculated as a second expected value at the candidate timing,
Disadvantages that occur when the processor maintains the part at the selected candidate timing, with the processor summing the value based on the calculated first expected value and the value based on the calculated second expected value A maintenance support method characterized in that it is calculated as a predicted value.
The maintenance support method according to claim 10,
The second disadvantage index is
In the period before the occurrence of the failure, the unit price of the component and the candidate timing are used to give a value that gradually decreases with time,
A maintenance support method, wherein 0 is given in a period after the occurrence of the failure.
A maintenance support program for executing a computer having a memory,
The memory includes a first disadvantage index indicating a first disadvantage degree and a second disadvantage degree indicating a second disadvantage degree among disadvantages experienced by a user due to maintenance of components mounted on the device. Stores disadvantageous indicators and
The first disadvantage index is such that a disadvantage that occurs before a failure of the component is lower than a disadvantage that occurs after the failure of the component;
The second disadvantage index is such that a disadvantage that occurs before a failure of the component is higher than a disadvantage that occurs after the failure of the component;
The maintenance support program is stored in the computer.
Obtaining a plurality of failure probabilities of the part at a plurality of points in time during which the part is used;
The component uses a predicted value of a disadvantage that occurs when the part is maintained using the plurality of acquired failure probabilities, the first disadvantage index, and the second disadvantage index. A procedure for calculating at a plurality of candidate timings during which maintenance is performed;
A maintenance support program for executing the procedure of outputting the calculated plurality of predicted values.
The maintenance support program according to claim 13,
In the computer,
A procedure for selecting one candidate timing from the plurality of candidate timings;
At each time point before the selected candidate timing, obtain a value obtained by multiplying the failure probability by the first disadvantage index after failure, and at each time point after the selected candidate timing, Obtaining a value obtained by multiplying the failure probability by the first disadvantage index before failure, and calculating a sum of the multiplied values as a first expected value at the candidate timing;
At each time point before the selected candidate timing, obtain a value obtained by multiplying the failure probability by the second disadvantage index after the failure, and at each time point after the selected candidate timing, Obtaining a value obtained by multiplying the failure probability by the second disadvantage index before failure, and calculating a sum of the multiplied values as a second expected value at the candidate timing;
The sum of the value based on the calculated first expected value and the value based on the calculated second expected value is used as a predicted value of a disadvantage that occurs when the part is maintained at the selected candidate timing. And a maintenance support program for executing the calculation procedure.
The maintenance support program according to claim 13,
The second disadvantage index is
In the period before the occurrence of the failure, the unit price of the component and the candidate timing are used to give a value that gradually decreases with time,
A maintenance support program characterized in that 0 is given in a period after the occurrence of the failure.