Lifetime prediction system

Abstract

The purpose of the present invention is to provide a life prediction system capable of performing highly reliable life prediction even if the recording of product events is not accurate. A service life prediction system (1) is provided with a server device (5) for predicting the service life of a device such as a work machine (2) or a product as a component of the device. The server device (5) includes: an operation information database (52) that stores operation information for each of the plurality of products; a history information database (53) that stores history information indicating a history of event data, which is a record of events occurring in each of the plurality of products; and an arithmetic processing device (54) that predicts the life of the product. On the basis of the operation information and the history information, the arithmetic processing device (54) calculates a calculation event data acquisition rate indicating the proportion of a product having event data in which the occurrence of an event is accurately recorded among the plurality of products, and predicts a corrected product life on the basis of the calculated event data acquisition rate.

Description

Technical Field

[0001] The present invention relates to a life prediction system for predicting the life of a product. Background Art

[0002] As a method for predicting the life of a product such as equipment such as a working machine or equipment parts, a survival time analysis (survival analysis) method for calculating a survival rate curve of a product is known (for example, Patent Document 1).

[0003] Patent Document 1 discloses the use of the Kaplan-Meier method, which is one of the survival time analysis methods, to evaluate the device life of plant equipment.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent No. 6944586 Summary of the invention

[0007] Problems to be solved by the invention

[0008] The Kaplan-Meier method is an effective method for predicting product life based on "events" such as product repairs (including parts replacement). However, the Kaplan-Meier method is based on the premise that the events that occur in the product are accurately recorded.

[0009] For example, when factory equipment is the object of observation as in Patent Document 1, it is possible to accurately record events occurring in the factory equipment (repair or abnormality of the factory equipment) because professional managers, suppliers, etc. regularly perform maintenance.

[0010] On the other hand, in the case of working machines, users may perform repairs or replace parts with parts other than genuine products (hereinafter also referred to as "counterfeit products"). Since the replacement history of counterfeit products cannot be accurately detected, when the working machine is the object of observation, the events that occurred may not be accurately recorded. As a result, when the Kaplan-Meier method is used to predict the life of a product such as a working machine, even if an event such as a repair including a part replacement occurs, a situation that is treated as no event occurs (no repair occurs) may occur, so a prediction of a life that is much longer than the original life may be made.

[0011] In view of the above circumstances, an object of the present invention is to provide a life prediction system capable of performing highly reliable life prediction even when the record of an event occurring in a product is inaccurate.

[0012] Means for solving problems

[0013] In order to solve the above-mentioned problems, the life prediction system of the present invention has a server device for predicting the life of a device or a product serving as a component of the device, and is characterized in that the server device includes: an operation information database, which stores the operation information of each of a plurality of products; a history information database, which stores history information representing the history of event data, wherein the event data is a record of events occurring to each of the plurality of products; and an operation processing device, which predicts the life, wherein the operation processing device calculates an event data acquisition rate based on the stored operation information and the history information, and predicts the corrected life based on the calculated event data acquisition rate, wherein the event data acquisition rate represents the proportion of products among the plurality of products that have the event data that accurately records the occurrence of the event.

[0014] Effects of the Invention

[0015] According to the present invention, it is possible to provide a life prediction system capable of performing highly reliable life prediction even when the record of an event that occurred in a product is inaccurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a diagram showing the configuration of the life prediction system according to the present embodiment.

[0017] Figure 2 Yes Description Figure 1 FIG. 2 is a diagram illustrating the information stored in the operation information database shown.

[0018] Figure 3 Yes Description Figure 1 FIG. 1 is a diagram illustrating information stored in the historical information database shown.

[0019] Figure 4 It is a figure which shows the lifespan prediction result.

[0020] Figure 5 It is shown by Figure 1 Flowchart of life prediction related processing performed by the arithmetic processing device shown.

[0021] Figure 6 It is shown Figure 5 A detailed flow chart of step s2 is shown.

[0022] Figure 7 It is shown Figure 5 A detailed flow chart of step s3 is shown.

[0023] Figure 8 It is a diagram showing another display example of the life prediction result.

[0024] Fig. 9 It is a diagram showing another display example of the life prediction result.

[0025] Fig.10 It is a diagram showing another display example of the life prediction result. DETAILED DESCRIPTION

[0026] Hereinafter, embodiments of the present invention will be described using the drawings. Unless otherwise specified, components denoted by the same reference numerals in each embodiment have the same functions in each embodiment, and description thereof will be omitted.

[0027] Figure 1 It is a diagram showing the configuration of the life prediction system 1 according to the present embodiment. Figure 2 Yes Description Figure 1 A diagram of information stored in the operating information database 52 is shown. Figure 3 Yes Description Figure 1 FIG. 5 is a diagram of information stored in the history information database 53 shown. Figure 4 It is a figure which shows the lifespan prediction result.

[0028] The life prediction system 1 is a life prediction system for predicting the life of a device or a product that is a device component. The product that is the object of observation of the life prediction system 1 may be a product such as a work machine 2 that may not accurately create a record of the events that have occurred. Alternatively, the product that is the object of observation of the life prediction system 1 may be a product that accurately creates a record of the events that have occurred. In this embodiment, a case where the product that is the object of observation of the life prediction system 1 is the work machine 2 is described.

[0029] The life prediction system 1 includes a working machine 2, a terminal device 3, a wireless base station 4, and a server device 5. The working machine 2, the terminal device 3, and the server device 5 are connected via a communication network N so as to be able to communicate with each other.

[0030] The working machine 2 includes an operation information sensor 21, a control device 22, a communication device 23, and a display 24. The operation information sensor 21 is installed on the working machine 2, and collects operation information including operation time, travel distance, and pressure of a hydraulic actuator. The control device 22 has an operation information acquisition unit 221 that acquires operation information from the operation information sensor 21 and an operation information output unit 222 that outputs the acquired operation information to the communication device 23 or the display 24. The communication device 23 communicates with the server device 5 via the wireless base station 4 constituting the communication network N. The communication device 23 transmits the acquired operation information of the working machine 2 to at least the server device 5 via the communication network N. The display 24 can display the operation information output from the control device 22 or the life prediction result sent from the server device 5.

[0031] The terminal device 3 is a portable terminal device such as a smart phone having a wireless communication function, a camera function, an input function, a display function, etc. The terminal device 3 may also be a terminal device held by a maintenance personnel of the working machine 2. The terminal device 3 includes a display 31 composed of a touch screen, etc. as an input function and a display function. The display 31 displays the maintenance items of the working machine 2, and accepts the input of the maintenance results performed by the maintenance personnel, and stores them in a specified storage area of ​​the terminal device 3. The maintenance results include a record of repairs (including parts replacement) performed on the working machine 2, that is, a repair record. The terminal device 3 sends the maintenance results of the working machine 2 to at least the server device 5 via the communication network N. In addition, the display 31 displays the operation information sent from the working machine 2 or the life prediction results sent from the server device 5, etc.

[0032] The server device 5 predicts the life of the working machine 2 as an observation object. The server device 5 includes a communication device 51 , an operation information database 52 , a history information database 53 , a calculation processing device 54 , and a prediction result database 55 .

[0033] The communication device 51 communicates with the working machine 2 or the terminal device 3 via the communication network N.

[0034] The operation information database 52 is a database that stores the operation information of each of the plurality of working machines 2. Figure 2 As shown, information identifying the "machine number" of each working machine 2, information on the "operating time" at the time obtained in each working machine 2, and information on the "operating date and time" indicating the date and time when each working machine 2 obtained the operating information are associated with each other and stored in the operating information database 52. The operating information is sent from the working machine 2 to the server device 5 periodically (for example, once a day).

[0035] The history information database 53 is a database that stores the history information of each of the plurality of working machines 2. The history information indicates the history of event data, which is a record of events that occurred in each of the plurality of working machines 2. An event indicates a change from a normal state of the product of the observation object. In the case where the product of the observation object is the working machine 2, an event occurs when the working machine 2 changes from a normal operating state to a state where a component failure or repair occurs. That is, the event in this embodiment includes repairing the working machine 2. The event data in this embodiment includes a repair record of the working machine 2.

[0036] like Figure 3As shown in FIG. 5 , information identifying the “machine number” of each working machine 2, information on the “repair date” as one of the repair records of each working machine 2, and information on the “repaired parts” as one of the repair records of each working machine 2 are associated with each other and stored in the history information database 53. In addition to the repair records of the working machines 2, the inspection results of the working machines 2 are also associated with the “machine number” information and stored in the history information database 53.

[0037] The operation processing device 54 predicts the life of the working machine 2 based on the operation information stored in the operation information database 52 and the history information stored in the history information database 53. Specifically, the operation processing device 54 calculates the event data acquisition rate, and predicts the corrected life prediction based on the calculated event data acquisition rate, and the event data acquisition rate indicates the ratio of products in the plurality of working machines 2 that have event data that accurately records the occurrence of events.

[0038] The calculation processing device 54 is configured to include a processor such as a CPU and memories such as a ROM and a RAM. The various functions of the calculation processing device 54 are realized by the processor executing a program stored in the ROM.

[0039] like Figure 1 As shown, the arithmetic processing device 54 includes an information acquisition unit 541 , an extraction unit 542 , a first calculation unit 543 , a second calculation unit 544 , a life prediction unit 545 , a maintenance period determination unit 546 , and an information output unit 547 .

[0040] The information acquisition unit 541 acquires the operation information and history information of the operating machine 2 to be observed from the operation information database 52 and the history information database 53, respectively. The operating machine 2 to be observed is specified by a user such as a maintenance personnel inputting into the terminal device 3. At this time, one operating machine 2 can be specified, or a plurality of operating machines 2 having common attributes such as machine type, vehicle specification (weight), etc. can be specified in groups. When specified in groups, it is preferred because the reliability of life prediction is improved. It should be noted that, in the case where there are a plurality of operating machines 2 around the maintenance personnel, an overview of the operating machines 2 that are candidates for observation can also be automatically displayed on the terminal device 3 based on the position information of the operating machines 2 and the terminal device 3, and an arbitrary operating machine 2 can be specified therefrom.

[0041] The extraction unit 542 extracts long-term operating products that operate longer than the design life from a plurality of operating machines 2 based on the operating information of the operating machines 2. The operating information includes the operating time of the operating machines 2. The operating machines 2 to be observed include operating machines 2 that are equivalent to long-term operating products and operating machines 2 that are not equivalent to long-term operating products. The extraction unit 542 calculates the total operating time based on the operating time included in the operating information, and the total operating time is the sum (accumulation) of the operating time from shipment to the present. The extraction unit 542 extracts the operating machines 2 whose calculated total operating time reaches, for example, 1.5 times or more of the design life as long-term operating products. The operating machines 2 that are equivalent to long-term operating products also include operating machines 2 with history information of the existence of repair records and operating machines 2 with history information of the absence of repair records. It should be noted that the above-mentioned "1.5 times" is an example of a design life based on a standard, and the judgment criteria for long-term operating products can also be changed arbitrarily.

[0042] The first calculation unit 543 calculates the event data acquisition rate based on the history information, and the event data acquisition rate indicates the ratio of the working machines 2 having the event data accurately recording the occurrence of the event among the plurality of working machines 2 as the observation object. For example, when the number of the working machines 2 as the observation object is 10,000 and the number of the working machines 2 for which the occurrence of the event is accurately observed is 500, the event data acquisition rate is calculated as (500 / 10,000)=5%.

[0043] The operating machines 2 to be observed include operating machines 2 with repair records as event data and operating machines 2 without repair records. The operating machines 2 with repair records as event data refer to the operating machines 2 with repair records in the history information, and are hereinafter referred to as "operating machines 2 with accurately recorded repair records". The operating machines 2 without repair records also include, for example, operating machines 2 that were actually repaired but have no repair records, and operating machines 2 that were repaired using imitations, and such operating machines 2 are hereinafter referred to as "operating machines 2 with inaccurately recorded repair records". Even if the Kaplan-Meier method is used to perform life prediction, it is difficult to perform a highly reliable life prediction for operating machines 2 with inaccurately recorded repair records. Therefore, in order to exclude operating machines 2 with inaccurately recorded repair records from performing life prediction, the first calculation unit 543 calculates the event data acquisition rate representing the proportion of operating machines 2 with accurately recorded repair records among the operating machines 2 to be observed.

[0044] Here, long-term operating products with a total operating time of more than 1.5 times the design life should originally have repair records in all long-term operating products. However, in reality, long-term operating products include long-term operating products with no repair records and long-term operating products with repair records due to the influence of repairs using imitation products. It is very likely that the proportion of long-term operating products with repair records among long-term operating products is equal to the proportion of operating machines 2 with accurately recorded repair records among the operating machines 2 of the observation object. Therefore, the first calculation unit 543 of the present embodiment calculates the proportion of long-term operating products with repair records among the long-term operating products extracted by the extraction unit 542 as the event data acquisition rate. As a result, the first calculation unit 543 of the present embodiment can accurately calculate the event data acquisition rate.

[0045] Specifically, the first calculation unit 543 of the present embodiment calculates the event data acquisition rate Fr using the mathematical formula 1. In the mathematical formula 1, Fr represents the event data acquisition rate, A represents the number of long-term operation products with repair records, and B represents the number of long-term operation products without repair records. The denominator (A+B) on the right side of the mathematical formula 1 represents the total number of long-term operation products extracted by the extraction unit 542.

[0046]

Mathematical formula 1

[0047]

[0048] It should be noted that the first calculation unit 543 of the present embodiment counts the first repair record (specified as a calculation object) and calculates the event data acquisition rate when the history information of each long-term operating product includes multiple repair records. The first repair record of the operating machine 2 indicates the first repair of any component in the operating machine 2 after the operating machine 2 leaves the factory. Depending on the operating machine 2, multiple repairs are sometimes performed. For example, in the case where a certain component of the operating machine 2 is repaired after a failure (first repair date) and the component fails again after the repair and is repaired again, the repair record of the operating machine 2 includes multiple repair records. The first calculation unit 543 of the present embodiment calculates the number of long-term operating products with repair records by determining the first repair record with the earliest repair date among the multiple repair records of each long-term operating product and counting the determined first repair records. As a result, the first calculation unit 543 does not need to repeatedly count the repair records, and can therefore accurately calculate the event data acquisition rate.

[0049] The second calculation unit 544 calculates the operation time of the working machine 2 from the factory shipment to the first repair record, that is, the inter-failure operation time, based on the operation information and the history information. Specifically, the second calculation unit 544 determines the first repair record of each of the plurality of working machines 2 as the observation object, and determines the first repair date based on the determined first repair record. Then, the second calculation unit 544 calculates the operation time from the factory shipment date to the first repair date of each of the plurality of working machines 2 as the observation object as the inter-failure operation time.

[0050] The life prediction unit 545 uses the Kaplan-Meier method to predict the life of the working machine 2. At this time, the life prediction unit 545 predicts the corrected life based on the event data acquisition rate calculated by the first calculation unit 543, and calculates the corrected life prediction result. The corrected life prediction result refers to the life prediction result corrected according to the life prediction result calculated by directly applying the Kaplan-Meier method (hereinafter also referred to as "the life prediction result before correction"). Specifically, the life prediction unit 545 corrects the survival number of multiple working machines 2 as the observation object according to the event data acquisition rate calculated by the first calculation unit 543. Then, the life prediction unit 545 estimates the survival rate of multiple working machines 2 using the Kaplan-Meier method based on the corrected survival number of the working machines 2 and the inter-failure operation time calculated by the second calculation unit 544. In this way, the life prediction unit 545 can calculate the corrected life prediction result. It should be noted that the survival number of the working machine 2 represents the number of working machines 2 that have not failed among the multiple working machines 2 as the observation object. The survival rate of the working machines 2 refers to the ratio of the number of working machines 2 that have not failed to the total number of the plurality of working machines 2 that are the observation targets.

[0051] The life prediction result before correction calculated by directly applying the Kaplan-Meier method is expressed by Mathematical Formula 2. In Mathematical Formula 2, S^ KM (t) represents the estimated value of the survival rate of the working machine 2 according to the total operating time t. In the mathematical formula 2, n on the right side is i represents the number of working machines 2 that survived before the total operating time t (survival number). i Indicates the number of working machines 2 that failed during the total operating time t.

[0052]

Mathematical formula 2

[0053]

[0054] On the other hand, as shown in Mathematical Formula 3, the life prediction unit 545 of the present embodiment calculates the survival number n of the working machine 2 at the event data acquisition rate Fr. iThe correction is performed to estimate the survival rate of the working machine 2. In the equation 3, S^ AKM (t) represents the estimated value of the survival rate of the working machine 2 after correction.

[0055]

Mathematical formula 3

[0056]

[0057] The life prediction unit 545 can calculate Figure 4 The lifespan prediction results are shown. Figure 4 The vertical axis represents the estimated value of the survival rate of the working machine 2. Figure 4 The horizontal axis represents the total operating time of the working machine 2. Figure 4 In the figure, the corrected life prediction results are represented by a solid line, and the corrected life prediction results are represented by a dotted line. Figure 4 As shown in the figure, for the life prediction results before correction, due to the survival number (n i ) includes the number of working machines 2 that were repaired using imitations and the number of working machines 2 that were actually repaired but have no repair records. Therefore, even after the total operating time has passed, the estimated value of the survival rate may not become zero. In contrast, the corrected life prediction result is due to the fact that the number of working machines 2 that survive (n) is increased by the event data acquisition rate (Fr). i ) is corrected, and when the total operating time passes a long period, the estimated value of the survival rate reaches zero. Comparing the corrected life prediction result with the life prediction result before correction, it can be seen that the corrected life prediction result of mathematical formula 3 of the service life prediction unit 545 is appropriate.

[0058] The maintenance period determination unit 546 determines the period when the operating machine 2 should be maintained based on the corrected life prediction result calculated by the life prediction unit 545. For example, the maintenance period determination unit 546 can regard the period T1 when the estimated value of the survival rate of the operating machine 2 as the corrected life prediction result reaches the first threshold value S1 (for example, 75%) as the period when the remaining life of a certain operating machine 2 reaches the first threshold value S1. The maintenance period determination unit 546 determines the period T1 when the estimated value of the survival rate of the operating machine 2 reaches the first threshold value S1 as the recommended maintenance period of the operating machine 2. The maintenance period determination unit 546 determines the period T2 when the estimated value of the survival rate of the operating machine 2 as the corrected life prediction result reaches the second threshold value S2 (for example, 50%) as the strongly recommended maintenance period of the operating machine 2. In addition, the maintenance period determination unit 546 determines the period T3 when the estimated value of the survival rate of the working machine 2 as the corrected life prediction result reaches the third threshold value S3 (for example, 25%) as the period when the working machine 2 should be urgently maintained. In this way, the maintenance period determination unit 546 determines the period when the working machine 2 should be maintained based on the corrected life prediction result calculated by the life prediction unit 545.

[0059] It should be noted that in this embodiment, the service life of the working machine 2 at the time of shipment is set to 100%, and the standard of the service life of the working machine 2 at present is expressed as a percentage corresponding to the total operating time after shipment, and the index of this percentage is also called "life progress". In this embodiment, the estimated value of the survival rate of the working machine 2 estimated by the life prediction unit 545 is regarded as the life progress of the working machine 2.

[0060] The information output unit 547 registers the life prediction result calculated by the life prediction unit 545 and the inspection period determination result obtained by the inspection period determination unit 546 in the prediction result database 55 and outputs them to the communication device 51. The communication device 51 transmits the life prediction result and the inspection period determination result to at least the terminal device 3 via the communication network N. The terminal device 3 transmits the life prediction result and the inspection period determination result in the form of, for example, Figure 4 The graph shown is displayed on the display 31. It should be noted that other display examples such as life prediction results are displayed using Figure 8 to Figure 10 As described later.

[0061] The prediction result database 55 stores the life prediction results calculated by the life prediction unit 545 and the maintenance period determination results obtained by the maintenance period determination unit 546. The life prediction results stored in the prediction result database 55 are provided to the calculation processing device 54, which helps to improve the reliability of the life prediction of the calculation processing device 54.

[0062] It should be noted that the life prediction unit 545 can calculate not only the corrected life prediction result represented by Mathematical Formula 3, but also the life prediction result before correction represented by Mathematical Formula 2. In this case, the information output unit 547 outputs the corrected life prediction result and the life prediction result before correction to the communication device 51. The communication device 51 sends the corrected life prediction result and the life prediction result before correction to the terminal device 3. The terminal device 3 displays the corrected life prediction result and the life prediction result before correction on the display 31.

[0063] In addition, the life prediction unit 545 may perform life prediction of the working machine 2 using a method other than the Kaplan-Meier method, such as the Nelson-Aalen method.

[0064] Figure 5 is and by Figure 1 Flowchart of the process related to life prediction performed by the arithmetic processing unit 54 shown.

[0065] In step s1 , the information acquisition unit 541 of the processing device 54 acquires the operation information and the history information of the plurality of working machines 2 to be observed from the operation information database 52 and the history information database 53 .

[0066] In step s2, the extraction unit 542 of the processing device 54 extracts long-term operation products from the plurality of working machines 2. Figure 6 As described later.

[0067] In step s3, the first calculation unit 543 of the processing device 54 calculates the number of first repairs obtained by counting the first repair records of each extracted long-running product. In mathematical formula 1, the number of first repairs is equivalent to the number A of long-running products with repair records. Figure 7 As described later.

[0068] In step s4 , the first calculation unit 543 of the arithmetic processing device 54 calculates the event data acquisition rate.

[0069] In step s5 , the second calculation unit 544 of the processing device 54 specifies the first repair date of each of the plurality of working machines 2 to be observed.

[0070] In step s6 , the second calculation unit 544 of the arithmetic processing device 54 calculates the inter-failure operation time of each of the plurality of working machines 2 .

[0071] In step s7, the life prediction unit 545 of the processing device 54 performs life prediction of the working machine 2. Then, the processing device 54 ends Figure 5 Processing shown.

[0072] It should be noted that the processing device 54 may also perform the processing of steps s2 to s4 and the processing of steps s5 to s6 in parallel after step s1, and then perform step s7. In addition, the processing device 54 may also perform the processing of steps s5 to s6 first after step s1, then perform the processing of steps s2 to s4, and then perform step s7.

[0073] Figure 6 It is shown Figure 5 A detailed flow chart of step s2 is shown.

[0074] In step s21 , the extraction unit 542 of the processing device 54 calculates the total operating time of each of the plurality of working machines 2 to be observed.

[0075] In step s22, the extraction unit 542 of the processing device 54 determines whether the total operation time of the working machine 2 is 1.5 times or more of the design life. If the total operation time is 1.5 times or more of the design life of the working machine 2, the extraction unit 542 proceeds to step s23. If the total operation time is not 1.5 times or more of the design life of the working machine 2, the extraction unit 542 proceeds to step s24.

[0076] In step s23, the extraction unit 542 of the calculation processing device 54 extracts the working machine 2 that is the object of the determination process in step s22 as a long-term operation product.

[0077] In step s24, the extraction unit 542 of the processing device 54 determines whether the determination process of step s22 has been performed on all the working machines 2 that are the observation objects. If the determination process of step s22 has been performed on all the working machines 2, the extraction unit 542 ends. Figure 6 This process shown is transferred to Figure 5 If the determination process of step s22 has not been performed on all the working machines 2, the extraction unit 542 proceeds to step s3. Figure 5 Step s22.

[0078] Figure 7 It is shown Figure 5 A detailed flow chart of step s3 is shown.

[0079] In step s31, the first calculation unit 543 of the arithmetic processing device 54 specifies the first repair record of each extracted long-term operation product.

[0080] In step s32, the first calculation unit 543 of the processing device 54 counts the determined first repair records to calculate the number of first repairs. Then, the first calculation unit 543 ends. Figure 7This process shown is transferred to Figure 5 Step s4.

[0081] Figure 8 It is a diagram showing other display examples of life prediction results and the like. Fig. 9 It is a diagram showing other display examples of life prediction results and the like. Fig.10 It is a diagram showing other display examples of life prediction results and the like.

[0082] The terminal device 3 can transmit the life prediction result and the judgment result of the maintenance period to Figure 8 The display form shown is displayed on the display 31. For example, Figure 8 As shown on the left side of the figure, the terminal device 3 can display an alarm 311 indicating the urgency of the maintenance period corresponding to the remaining life of the current working machine 2, a label 312 for displaying detailed life prediction results, etc., and a label 313 for setting the display format of the display 31. When the user clicks the label 312, as shown in Figure 8 As shown on the right side of FIG. 3 , the terminal device 3 displays detailed life prediction results, etc. in the form of a display such as a chart 314. Figure 8 As shown on the right side of the figure, the terminal device 3 can display a label 315 for switching the display of the life prediction result before correction and the display of the life prediction result after correction. When the user clicks the label 315, the terminal device 3 can switch the display of the life prediction result before correction and the life prediction result after correction. The terminal device 3 can also display the life prediction result before correction and the life prediction result after correction on the same screen at the same time for comparison display.

[0083] As described above, the life progress of the working machine 2 is an index that represents the standard of the current life of the working machine 2 as a percentage based on the total operating time after the working machine 2 is shipped, with the service life of the working machine 2 at the time of shipment being set as 100%. When the life progress is assumed to be 50%, it is estimated that about half of the working machines 2 are faulty based on past repair records, etc. Since the urgency of maintenance is high when the life progress is low, the server device 5 (maintenance period determination unit 546 of the operation processing device 54) can set a plurality of thresholds for the remaining life according to the urgency of maintenance, and display a different alarm 311 on the terminal device 3 each time the remaining life reaches each threshold. For example, when the life progress reaches 75%, the terminal device 3 can display an alarm 311 of "maintenance recommendation" because the service life is relatively sufficient. For example, when the life progress reaches 50%, the terminal device 3 can display an alarm 311 of "important" that strongly recommends maintenance. For example, when the life progress reaches 25%, the terminal device 3 can display an alarm 311 of "urgent" that indicates that maintenance should be urgently performed.

[0084] In addition, in order to make it easier for users to understand the current life progress stage, such as Figure 8 As shown on the right side of , the terminal device 3 displays the current life progress as a horizontal line S in the chart 314 or displays the maintenance period corresponding to the current life progress as a vertical line T in the chart 314. At this time, the terminal device 3 can display the horizontal line S and the vertical line T in different colors according to the urgency of the maintenance. For example, before the life progress reaches "maintenance recommendation", the terminal device 3 can display the horizontal line S and the vertical line T in green. For example, the terminal device 3 can display the horizontal line S and the vertical line T in blue after the life progress reaches "maintenance recommendation" and before reaching "important". For example, the terminal device 3 can display the horizontal line S and the vertical line T in yellow after the life progress reaches "important" and before reaching "urgent". For example, the terminal device 3 can display the horizontal line S and the vertical line T in red after the life progress reaches "urgent".

[0085] In addition, the terminal device 3 can display the service life of the working machine 2 corresponding to the current life progress or the date when the service life ends or the date of the maintenance period of the working machine 2 corresponding to the current life progress in a pop-up window that appears near the horizontal line S or the vertical line T. For example, as the service life of the working machine 2 corresponding to the current life progress (the date when the service life ends), the terminal device 3 can display the message "within half a year (× month × day)" in the pop-up window. It should be noted that the text expression is an example, and it can also be a calendar display, and the display format is not limited to messages and pop-ups. For example, during the period when there is room for the service life, it stays at a simple expression of "3 months ago". If the date when the service life ends is close, it can also be a display that marks a specific date standard.

[0086] In addition, the terminal device 3 may also transmit the life prediction result and the judgment result of the maintenance period to Fig. 9 Form 316 shown instead of Figure 8 A display form of a graph 314 shown on the right side of is displayed on the display 31 .

[0087] In addition, the terminal device 3 can transmit the life prediction result and the judgment result of the maintenance period to Figure 8 and Fig. 9 The display forms other than the display forms shown are displayed on the display 31. For example, the terminal device 3 Fig.10As shown, the life prediction result and the result of the inspection period determination are displayed in the form of message 317. Message 317 can also be sent to the terminal device 3 via an email or SMS from the server device 5, and pop-up display in the terminal device 3. Message 317 includes the life progress of the working machine 2 and the standard of the inspection period. The sending frequency and display frequency of message 317 can be the frequency of sending and displaying every time the life progress reaches a threshold, or the frequency of sending and displaying every certain period such as every week. In order to improve the convenience of the user, the sending frequency and display frequency of message 317 can be arbitrarily changed according to the user's settings.

[0088] It should be noted that the terminal device 3 can Figure 8 to Figure 10 The terminal device 3 can arbitrarily change the display examples according to the user's settings. Figure 8 to Figure 10 The combination of each display example displays the life prediction result and the judgment result of the maintenance period. Figure 8 to Figure 10 In the example in which the life prediction result and the determination result of the maintenance period are displayed on the display 31 of the terminal device 3, the life prediction result and the determination result of the maintenance period can be displayed not only on the terminal device 3, but also on the display 24 of the working machine 2.

[0089] As described above, the life prediction system 1 of the present embodiment is a life prediction system including a server device 5 for predicting the life of a device such as a working machine 2 or a product as a device component. The server device 5 includes an operation information database 52 storing operation information of each of a plurality of products to be observed and a history information database 53 storing history information representing records of events occurring in each of the plurality of products, i.e., history of event data. The server device 5 includes a calculation processing device 54 for predicting the life of the product. The calculation processing device 54 calculates an event data acquisition rate representing the proportion of products having event data accurately recording the occurrence of events among a plurality of products based on the operation information stored in the operation information database 52 and the history information stored in the history information database 53. The calculation processing device 54 predicts the corrected product life based on the calculated event data acquisition rate.

[0090] Thus, even when the event that occurred to the product is not accurately recorded, the life prediction system 1 of this embodiment can perform life prediction. Therefore, even when the record of the event that occurred is inaccurate, the life prediction system 1 of this embodiment can perform life prediction with high reliability.

[0091] Furthermore, in the life prediction system 1 of the present embodiment, the calculation processing device 54 predicts the product life using the Kaplan-Meier method.

[0092] Thus, the life prediction system 1 of the present embodiment can use the Kaplan-Meier method which is a conventional life prediction method. Therefore, even when the record of the occurrence event is inaccurate, the life prediction system 1 of the present embodiment can easily perform a highly reliable life prediction.

[0093] In the life prediction system 1 of the present embodiment, the event data includes a repair record of the product. The operation processing device 54 includes an extraction unit 542 that extracts a long-term operating product that has been operating longer than the designed life from among a plurality of products based on the operation information stored in the operation information database 52. The operation processing device 54 includes a first calculation unit 543 that calculates the ratio of long-term operating products with repair records among the long-term operating products extracted by the extraction unit 542 as an event data acquisition rate based on the history information stored in the history information database 53. The operation processing device 54 includes a life prediction unit 545 that predicts the corrected product life based on the event data acquisition rate calculated by the first calculation unit 543.

[0094] Thus, the life prediction system 1 of the present embodiment can accurately calculate the event data acquisition rate indicating the proportion of products that accurately record the occurrence of events. Furthermore, since the life prediction system 1 of the present embodiment can perform life prediction based on the accurate event data acquisition rate, it is possible to improve the accuracy of life prediction. Therefore, even when the record of the occurrence event is inaccurate, the life prediction system 1 of the present embodiment can further perform a life prediction with high reliability.

[0095] In addition, in the life prediction system 1 of the present embodiment, the operation processing device 54 further includes a second calculation unit 544 that calculates the operating time of the product from the shipment of the product to the first above-mentioned repair record, that is, the inter-failure operation time, based on the operation information and the history information. The life prediction unit 545 corrects the survival number of the plurality of products according to the event data acquisition rate calculated by the first calculation unit 543. The life prediction unit 545 calculates the corrected life prediction result by estimating the survival rate of the plurality of products using the Kaplan-Meier method based on the corrected survival number and the calculated inter-failure operation time.

[0096] Thus, as a life prediction method, the life prediction system 1 of the present embodiment can expand the application range of the effective Kaplan-Meier method to perform life prediction even when the record of the product event is inaccurate. Therefore, even when the record of the event is inaccurate, the life prediction system 1 of the present embodiment can easily and reliably perform a life prediction with high reliability.

[0097] Furthermore, in the life prediction system 1 of the present embodiment, when the history information of each long-term operating product includes a plurality of repair records, the first calculation unit 543 counts the first repair records and calculates the event data acquisition rate.

[0098] Thus, the life prediction system 1 of this embodiment can accurately calculate the event data acquisition rate without repeatedly counting the repair records. Therefore, even if the records of the occurrence events are inaccurate, the life prediction system 1 of this embodiment can perform a more reliable life prediction.

[0099] Furthermore, in the life prediction system 1 of the present embodiment, the calculation processing device 54 includes a maintenance time determination unit 546 that determines the time when the product should be maintained based on the corrected life prediction result.

[0100] Therefore, even when the record of the occurrence event is inaccurate, the life prediction system 1 of the present embodiment can accurately determine the time when the product should be inspected while performing highly reliable life prediction.

[0101] In addition, the life prediction system 1 of the present embodiment further includes a terminal device 3 connected to the server device 5 via a communication network N. The server device 5 further includes a communication device 51 for communicating with the terminal device 3. The operation processing device 54 calculates the corrected life prediction result and calculates the life prediction result before correction indicating the life before correction. The communication device 51 sends the corrected life prediction result and the life prediction result before correction to the terminal device 3. The terminal device 3 displays the corrected life prediction result and the life prediction result before correction.

[0102] Thus, the life prediction system 1 of this embodiment can provide the user with the life prediction results before and after the correction, thereby improving the convenience of the user who wants to compare the life prediction results before and after the correction. Thus, even if the record of the event is inaccurate, the life prediction system 1 of this embodiment can improve the convenience of the user while performing a highly reliable life prediction.

[0103] The embodiments of the present invention are described in detail above, but the present invention is not limited to the above embodiments, and various changes can be made within the scope of the spirit of the present invention recorded in the claims. The present invention can add the structure of a certain embodiment to the structure of other embodiments, or replace the structure of a certain embodiment with other embodiments, or delete part of the structure of a certain embodiment.

[0104] Description of Reference Numerals

[0105] 1 ... life prediction system, 2 ... working machine, 3 ... terminal device, 5 ... server device, 51 ... communication device, 52 ... operation information database, 53 ... history information database, 54 ... operation processing device, 541 ... information acquisition unit, 542 ... extraction unit, 543 ... first calculation unit, 544 ... second calculation unit, 545 ... life prediction unit, 546 ... maintenance period determination unit

Claims

1. A life prediction system comprising a server device for predicting the life of a device or a product as a device component, wherein the life prediction system is characterized in that: The server device comprises: An operation information database storing operation information of each of the plurality of products; a history information database storing history information indicating a history of event data, the event data being a record of events occurring in each of the plurality of products; and an operation processing device, which predicts said life span, The processing unit calculates an event data acquisition rate based on the stored operating information and the historical information, and predicts the corrected life based on the calculated event data acquisition rate, wherein the event data acquisition rate represents the proportion of products among the multiple products that have the event data that accurately records the occurrence of the event.

2. The life prediction system according to claim 1, characterized in that: The arithmetic processing device predicts the life span using the Kaplan-Meier method.

3. The life prediction system according to claim 1, characterized in that: The event data includes a repair record of the product, The processing device has: an extracting unit that extracts, based on the stored operation information, long-operation products that operate longer than a designed life from among the plurality of products; a first calculation unit that calculates, as the event data acquisition rate, a ratio of the long-term operating products having the repair record among the extracted long-term operating products based on the stored history information; and A lifespan prediction unit predicts the corrected lifespan based on the calculated event data acquisition rate.

4. The life prediction system according to claim 3, characterized in that: The processing device further includes a second calculation unit, which calculates an inter-failure operation time based on the stored operation information and the history information, wherein the inter-failure operation time is an operation time of the product from the time when the product is shipped to the time when the first repair record is made. The life prediction unit corrects the survival number of the plurality of products according to the calculated event data acquisition rate, The corrected life prediction result is calculated by estimating the survival rate of the plurality of products using the Kaplan-Meier method based on the corrected survival number and the calculated operating time between failures.

5. The life prediction system according to claim 3, characterized in that: When the history information of each of the long-term operating products includes a plurality of the repair records, the first calculation unit counts the first repair records to calculate the event data acquisition rate.

6. The life prediction system according to claim 1, characterized in that: The arithmetic processing device includes a maintenance period determination unit configured to determine a period when the product should be maintained based on a corrected life prediction result.

7. The life prediction system according to claim 1, characterized in that: further comprising a terminal device connected to the server device via a communication network, The server device further includes a communication device for communicating with the terminal device. The operation processing device calculates the corrected life prediction result and the corrected life prediction result. The communication device transmits the corrected life prediction result and the uncorrected life prediction result to the terminal device, The terminal device displays the corrected life prediction result and the uncorrected life prediction result.

Images