Equipment maintenance system and equipment maintenance method

Abstract

An equipment maintenance system 1 includes an asset knowledge database 41 in which failure information relating to an equipment failure, the cause of the failure, whether or not a site visit to the equipment by a maintenance person is required when the failure occurs, and the cost of dealing with the failure are managed in association with each other. The equipment maintenance system 1 includes a failure cause presentation unit 33 that calculates, for a failure cause requiring a site visit, among a plurality of failure causes estimated by a failure cause estimating unit 31, a ratio between a first cost for a case in which a maintenance person deals with the failure on the current site visit, and a second cost for a case in which the maintenance person makes another site visit to deal with the failure in subsequent occurrence. On the basis of the ratio calculated by the failure cause presentation unit 33, the equipment maintenance system 1 outputs failure causes that should be addressed preferentially, among the failure causes requiring a site visit.

Description

Equipment maintenance system and equipment maintenance method

[0001] The present invention relates to an equipment maintenance system and an equipment maintenance method.

[0002] When equipment breaks down (for example, a multifunction printer cannot print), the maintenance department listens to the symptoms from the user, visits the user's site, deduces several possible causes of the failure, and maintenance staff then repairs the cause of the failure and takes action such as replacing parts.

[0003] For example, Patent Document 1 discloses a technology that generates repair items based on the risk of failure, the frequency of failure, and the impact of failure, and supports the creation of a repair plan for equipment.

[0004] Japanese Patent Application Laid-Open No. 2005-11327

[0005] However, the maintenance support system disclosed in the above-mentioned Patent Document 1 does not take into consideration the number of business trips, travel time required for business trips, and transportation costs when providing maintenance services.

[0006] When maintenance personnel respond to a failure at a user site, there are multiple failure modes that can occur. Due to response time limitations, it is common to only respond to failure modes that are likely to occur. However, in this case, other failure modes may occur at a later date, which creates the problem of increased travel costs due to additional visits.

[0007] The present invention has been made in consideration of the above circumstances, and aims to suppress an increase in travel costs due to another travel to respond to a failure when another failure mode occurs after responding to the failure.

[0008] In one aspect of the present invention, an equipment maintenance system for maintaining and managing equipment includes an asset knowledge database in which failure information regarding failures of the equipment, the causes of the failures, whether a maintenance worker will be dispatched to the equipment in the event of the failure, and the cost of responding to the failure are managed in a corresponding manner; an input unit that accepts input of the failure information related to the equipment; a failure cause estimation unit that refers to the asset knowledge database based on the failure information accepted by the input unit and estimates multiple causes of the failure; a failure cause presentation unit that calculates, for a failure cause that requires a dispatch, of the multiple failure causes estimated by the failure cause estimation unit, a ratio between a first cost in the case where the maintenance worker will respond to the failure on the current dispatch and a second cost in the case where the failure cause recurs in the next or subsequent cases and the maintenance worker will make another dispatch to respond; and an output unit that outputs to a terminal, the failure causes that require a dispatch, based on the ratio calculated by the failure cause presentation unit.

[0009] According to the present invention, it is possible to suppress an increase in travel costs due to another travel to deal with a failure after another failure mode occurs after the failure has been dealt with.

[0010] 1 is a diagram showing the configuration of an equipment maintenance system according to a first embodiment. FIG. 1 is a diagram showing the configuration of maintenance knowledge data according to a first embodiment. FIG. 2 is a diagram showing failure probability data according to a first embodiment. FIG. 3 is a diagram showing the probability of a child node abnormality occurring when a parent node abnormality occurs according to a first embodiment. FIG. 4 is a diagram showing the probability of a child node abnormality occurring when a parent node is normal according to a first embodiment. FIG. 5 is a diagram showing the configuration of a failure cause response method table according to a first embodiment. FIG. 6 is a diagram showing the configuration of a trip expense table according to a first embodiment. FIG. 7 is a diagram showing a maintenance knowledge Bayesian network according to a first embodiment. FIG. 8 is a flowchart showing a maintenance knowledge Bayesian network generation process according to a first embodiment. FIG. 9 is a flowchart showing a failure cause estimation process according to a first embodiment. FIG. 10 is a diagram showing a failure cause estimation result according to a first embodiment. FIG. 11 is a diagram showing an overview of a check item search process according to a first embodiment. FIG. 12 is a flowchart showing a check item search process for an investigation target according to a first embodiment. FIG. 13 is a diagram showing investigation target check items according to a first embodiment. FIG. 14 is a flowchart showing a failure cause presentation process according to a first embodiment. FIG. 15 is a diagram showing a cost ratio calculation result according to a first embodiment. FIG. 16 is an input screen according to a first embodiment. FIG. 17 is a diagram showing a result screen according to a first embodiment. FIG. 18 is a diagram showing the configuration of an equipment maintenance system according to a second embodiment. FIG. 19 is a diagram showing a failure cause response history according to a second embodiment. 10 is a flowchart showing a fault cause presentation update process according to the second embodiment.

[0011] In the following description, a "CPU (Central Processing Unit)" is an example of one or more processor devices. The at least one processor device is typically not limited to a CPU, and may be another type of processor device such as a GPU (Graphics Processing Unit). The at least one processor device may be a single-core or multi-core. The at least one processor device may be a processor core.

[0012] At least one processor device may be a circuit that is a collection of gate arrays written in a hardware description language that performs some or all of the processing, such as a broad processor device such as an FPGA (Field-Programmable Gate Array), a CPLD (Complex Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit).

[0013] In the following explanation, the processing may be mainly explained by the "yyy program." In this case, the program is executed by the CPU to realize a processing function called the "yyy function unit," and is the main executing entity of the processing. The processing function may be realized by one or more computer programs being executed by a processor, or may be realized by one or more hardware circuits (e.g., FPGA or ASIC), or may be realized by a combination thereof.

[0014] When a function is realized by executing a program by a processor, the specified processing is performed using a storage device and / or an interface device, etc., as appropriate, and therefore the function may be considered to be at least a part of the processor. Processing described using a functional unit as the subject may also be processing performed by a processor or a device having that processor.

[0015] The program may be installed from a program source. The program source may be, for example, a program distribution computer or a computer-readable recording medium (e.g., a non-transitory recording medium). The description of each function is an example, and multiple functions may be combined into one function, or one function may be divided into multiple functions. A "yyy function unit" may be called a "yyy unit."

[0016] In the following description, various pieces of information may be described in table format, but the data format of the information may be a format other than a table format (for example, CSV (Comma Separated Values) format, etc.) Furthermore, various pieces of information may be stored in a storage unit as a table, or may be embedded as logic in a program.

[0017] In addition, in the following description, when describing elements of the same type without distinguishing between them, common reference symbols will be used, and when describing elements of the same type with distinction between them, reference symbols will be used.

[0018] [Embodiment 1] (Configuration of equipment maintenance system 1 according to embodiment 1) Fig. 1 is a diagram showing the configuration of equipment maintenance system 1 according to embodiment 1. Equipment maintenance system 1 has a CPU (Central Processing Unit) 2, memory 3, external storage device 4, and input / output unit 5. Input / output unit 5 includes an input unit that accepts information input via a maintenance worker terminal 6, and an output unit that outputs information to the display screen of maintenance worker terminal 6. In this embodiment, equipment maintenance system 1 is described as being configured using a standalone computer, but it may also be realized by a server on the cloud, or each function may be divided and realized by multiple computers.

[0019] The memory 3 includes a fault cause estimation unit 31 , a check item search unit 32 , and a fault cause presentation unit 33 .

[0020] The failure cause estimation unit 31 estimates the cause of a failure based on the input equipment status and alarms. When an abnormality occurs in the equipment, the failure cause estimation unit 31 generates a maintenance knowledge Bayesian network BN ( FIG. 8 ) described below for cause estimation. The failure cause estimation unit 31 inputs check items into the maintenance knowledge Bayesian network BN according to information on the symptoms of the failure input by a maintenance worker or a user, and calculates the occurrence probability of each failure mode.

[0021] The check item search unit 32 identifies and outputs items that maintenance personnel should check from the failure causes estimated by the failure cause estimation unit 31. The failure cause presentation unit 33 calculates and outputs a ranking of failure causes that should be addressed with priority during on-site maintenance from the check items identified by the check item search unit 32, in accordance with a designation of whether financial cost or time cost is to be prioritized.

[0022] The external storage device 4 stores an asset knowledge database 41 and a failure mode response method database 42 .

[0023] The asset knowledge database 41 is a database that accumulates maintenance knowledge. The maintenance knowledge is information extracted from, for example, a maintenance manual, FMEA (Failure Mode and Effects Analysis), etc. Specifically, the asset knowledge database 41 stores maintenance knowledge data T1, failure probability data T2, a child node abnormality occurrence probability T3 when a parent node is abnormal, and a child node abnormality occurrence probability T4 when the parent node is normal.

[0024] The maintenance knowledge data T1 is information that associates functional failures and alarms with the failure modes that are the causes of the failures. The failure probability data T2 is data that stores the probability that the cause of the failure will occur.

[0025] The failure mode response method database 42 stores a failure cause response method table T5 that determines whether a failure can be handled remotely or whether an on-site response is required, a travel expense table T6 that lists the time and expense required for a travel, and failure cause estimation results T7. The failure cause estimation results T7 are a list of failure causes (failure modes) estimated based on the input failure information and the maintenance knowledge Bayesian network BN, associated with their respective occurrence probabilities, and listed in order of occurrence probability.

[0026] (Maintenance Knowledge Data T1 According to Embodiment 1) FIG. 2 is a diagram showing the configuration of the maintenance knowledge data T1 according to embodiment 1. Functional failure T11 indicates a failure phenomenon such as cooling water not circulating or component 2 not being able to be stored. Failure mode T12 indicates the cause of the failure. Check item T13 is a column for storing items to be checked, such as equipment sensor data, environment, equipment, components, etc. Check item T13 is used to identify phenomena that can occur when a failure mode occurs and to check whether the phenomenon has occurred. There is not necessarily one check item corresponding to one failure mode. Alarm T14 includes information about alarms that frequently occur when a failure corresponding to functional failure T11 occurs.

[0027] 3 is a diagram showing failure probability data T2 according to embodiment 1. The failure probability data T2 is data for setting a probability that is set when creating a Bayesian network. The failure probability data T2 includes a failure mode T21 in which the name of the failure mode is stored, a state T22, and a probability T23. The occurrence probability of each failure mode can be known based on the failure probability data T2.

[0028] The state T22 stores Y or N. "Y" in the state T22 means that the failure mode has occurred, and "N" means that the failure mode has not occurred. These states are set so that the sum of the respective probabilities T23 of Y and N related to the same failure mode is 1.

[0029] The probability T23 stores the probability of the failure mode state. Here, the failure mode probability may be set to a fixed value such as 50% as the probability corresponding to the failure mode states Y and N. However, this is not limited to this, and calculations can also be made from past history, etc. The probability of the failure mode occurrence state is, for example, a value obtained by dividing the number of occurrences of the failure mode in the failure history by the total number of occurrences.

[0030] 4 is a diagram showing the child node abnormality occurrence probability T3 when the parent node is abnormal according to embodiment 1. The child node abnormality occurrence probability T3 when the parent node is abnormal includes a parent node T31, a state T32, a child node T33, a child node state T34, and a probability T35. The child node abnormality occurrence probability T3 when the parent node is abnormal indicates the probability of abnormality or normality of a check item of a child node when a failure mode in the parent node occurs.

[0031] The parent node T31 stores the name of the failure mode, and the state T32 stores Y, which indicates the occurrence state of the failure mode.

[0032] A check item is stored in child node T33. A child node status T34 stores the status of the check item, and stores the status of the part corresponding to child node T33. A probability T35 stores the probability of the status of the check item stored in child node T33. The sum of the abnormal and normal probabilities related to the status of the same check item is set to 1.0 (100%). Note that fixed values ​​such as abnormal: 100% and normal: 0% may also be set. The fixed values ​​can be calculated from past failure history.

[0033] (Child Node Abnormality Occurrence Probability T4 When Parent Node is Normal According to Embodiment 1) Fig. 5 is a diagram showing the child node abnormality occurrence probability T4 when the parent node is normal according to embodiment 1. The child node abnormality occurrence probability T4 when the parent node is normal includes a child node T41, a child node state T42, and a probability T43.

[0034] A check item is stored in child node T41. A child node status T42 stores the status of the check item, and stores the status of the part corresponding to child node T41. A probability T43 stores the probability of the status of the check item stored in child node status T42. The sum of the abnormal and normal probabilities related to the status of the same check item is set to 1.0 (100%). Note that fixed values ​​such as abnormal: 100% and normal: 0% may also be set. The fixed values ​​can be calculated from past failure history.

[0035] (Fault cause response method table T5 according to embodiment 1) Fig. 6 is a diagram showing the configuration of the fault cause response method table T5 according to embodiment 1. The fault cause response method table T5 contains information used in the check item search process (Fig. 14) described later. The fault cause response method table T5 includes a failure mode T51, a response method T52, and a response time T53.

[0036] The response method T52 is information indicating whether on-site or remote repair is possible for the failure mode T51. The response time T53 is the work time required to take action for each failure mode T51.

[0037] (Travel Expense Table T6 According to First Embodiment) Fig. 7 is a diagram showing the configuration of the travel expense table T6 according to the first embodiment. The travel expense table T6 contains information used in the failure cause presentation process (Fig. 16) described below. The travel expense table T6 includes customer names T61, travel time to the customer's site T62, transportation costs to the customer's site T63, and labor costs T64. The labor costs T64 are the labor costs per hour per worker for each customer name T61.

[0038] (Maintenance knowledge Bayesian network BN according to embodiment 1) Fig. 8 is a diagram showing the maintenance knowledge Bayesian network BN according to embodiment 1. The maintenance knowledge Bayesian network BN is a Bayesian network with a four-layer structure of alarms BN1, functional failures BN2, failure modes BN3, and check items BN4. In the maintenance knowledge Bayesian network BN, arrows represent causal relationships, with the base of the arrow being a parent node and representing a cause, and the tip of the arrow being a child node and representing a result.

[0039] (Maintenance Knowledge Bayesian Network Generation Process According to Embodiment 1) Fig. 9 is a flowchart showing maintenance knowledge Bayesian network generation process according to embodiment 1. The maintenance knowledge Bayesian network generation process is executed by the failure cause estimation unit 31 using the maintenance knowledge data T1 stored in the asset knowledge database 41.

[0040] First, in step S11, the failure cause estimation unit 31 acquires maintenance knowledge data T1 from the asset knowledge database 41. Next, in step S12, the failure cause estimation unit 31 creates nodes of the maintenance knowledge Bayesian network BN from each cell of the maintenance knowledge data T1. At this time, the failure cause estimation unit 31 generates cells with the same content as one node.

[0041] Next, in step S13, the failure cause estimation unit 31 establishes parent-child relationship links for each node of the maintenance knowledge Bayesian network BN generated in step S12, in accordance with the causal relationships described in the maintenance knowledge data 18. Specifically, the causal relationships (parent-child relationships) are such as "failure mode" → "functional failure," "failure mode" → "check item," and "functional failure" → "alarm," and arrows representing the causal relationships are set for each node.

[0042] Next, in step S14, the failure cause estimation unit 31 sets the probability of the failure probability data T2 as a priori probability for the arrow set in step S13. Next, in step S15, the failure cause estimation unit 31 sets a posterior probability for the arrow indicating the parent-child relationship using the child node abnormality occurrence probability T3 when the parent node is abnormal and the child node abnormality occurrence probability T4 when the parent node is normal.

[0043] (Flowchart showing fault cause estimation processing according to the first embodiment) FIG. 10 is a flowchart showing fault cause estimation processing according to the first embodiment.

[0044] First, in step S21, the failure cause estimation unit 31 receives information on the symptoms of the failure from a maintenance worker operating the maintenance worker terminal 6 via the input / output unit 5. The maintenance worker operates the input screen D1 (FIG. 17) to input "abnormal" or "normal" in the check items on the screen, thereby inputting information on the equipment failure. "Abnormal" means that an abnormality has occurred, and "normal" means that an abnormality has not occurred. If it is unclear whether an abnormality has occurred, do not check "Abnormal". For example, if there is information that "oil is leaking", the maintenance worker checks "Abnormal" in the check item "oil is leaking".

[0045] Next, in step S22, the failure cause estimation unit 31 generates a maintenance knowledge Bayesian network BN based on the information input in step S21. Next, in step S23, the failure cause estimation unit 31 calculates the occurrence probability of each failure mode using the maintenance knowledge Bayesian network BN, and outputs the failure cause estimation result T7 to the failure cause presentation unit 33.

[0046] (Fault Cause Estimation Result According to Embodiment 1) Fig. 11 is a diagram showing a fault cause estimation result T7 according to embodiment 1. In step S23, an occurrence probability T72 corresponding to each failure mode T71 is calculated, as in the fault cause estimation result T7 shown in Fig. 11 .

[0047] 12 is a diagram showing an outline of the check item search process according to the first embodiment. The check item search process extracts failure modes for which the response method is "on-site visit" from the failure cause estimation results, and extracts check items that should be checked by maintenance personnel who have visited the facility to investigate whether or not these failure modes have occurred.

[0048] As an example of the check item search process, as shown in FIG. 12, assume that a symptom of "oil leaking" occurs and a maintenance worker checks "oil leaking." In this case, since the failure mode "sticking of part 3" is linked only to "oil leaking" (the link shown by the dashed line), there is no need to check other check items to investigate the occurrence of the failure mode "sticking of part 3." For other failure modes, it would be better to check the check item "abnormal noise" (shown in hatching) for the failure mode "poor installation of part 1," and "oil dripping" and "high internal temperature" (shown in hatching) for the failure mode "damage to part 4." Therefore, such check items are extracted. The check item search process can obtain information on the check items that the maintenance worker should check, and this information is used in the failure cause presentation process (FIG. 15).

[0049] 13 is a diagram showing a flowchart of the check item search process for the investigation target according to embodiment 1. The check item search process processes the failure cause estimation results T7 one by one using the maintenance knowledge data T1 and the failure cause response method table T5.

[0050] First, in step S31, the check item search unit 32 reads the maintenance knowledge data T1, the failure cause response method table T5, and one unprocessed record of the failure cause estimation result T7. The check item search unit 32 then searches the failure cause response method table T5 based on the failure mode included in the read failure cause estimation result T7, and obtains the response method that corresponds to either on-site or remote work for the failure mode.

[0051] Next, in step S32, the check item search unit 32 determines whether the response method for the failure mode read in step S31 is a business trip. If it is a business trip (step S32: Yes), the check item search unit 32 proceeds to step S33, and if it is remote (step S32: No), the check item search unit 32 returns to step S31.

[0052] Next, in step S33, the check item search unit 32 refers to the maintenance knowledge data T1 and acquires check items corresponding to the failure mode determined to be a business trip in step S32. However, in step S33, check items that have already been acquired are excluded so as not to be acquired redundantly.

[0053] Next, in step S34, the check item search unit 32 outputs the failure mode determined to be a business trip in step S32, the check items to be investigated acquired in step S33, and the response method and response time acquired from the failure cause response method table T5. This information is output to the check items to be investigated T8 (FIG. 14).

[0054] Next, in step S35, the check item search unit 32 determines whether all records of the failure cause estimation results T7 have been processed. If all records of the failure cause estimation results T7 have been processed (Yes in step S35), the check item search unit 32 ends the check item search process. On the other hand, if there are any records of the failure cause estimation results T7 that have not been processed (No in step S35), the check item search unit 32 returns the process to step S31.

[0055] (Investigation Target Check Items T8 According to Embodiment 1) FIG. 14 is a diagram showing investigation target check items T8 according to embodiment 1. The investigation target check items T8 include a failure mode T81, a check item T82, a response method T83, and a probability T84. The check item T82 is a check item that should be checked to investigate whether or not a failure mode in the failure cause estimation result T7 has occurred, in addition to the check items already entered by the maintenance personnel. The response method T83 is a response method of on-site or remote work corresponding to the failure mode T71 in the failure cause estimation result T7, and only the on-site failure mode is included after the check item search process ( FIG. 13 ). The probability T84 is the occurrence probability of the failure mode in the failure cause estimation result T7.

[0056] (Failure Cause Presentation Process According to First Embodiment) Fig. 15 is a flowchart showing the failure cause presentation process according to the first embodiment. The failure cause presentation process is a process for generating a priority ranking for which failure modes the maintenance personnel will prioritize in dealing with during the current visit. The failure cause presentation process reads the investigation target check items T8 generated in the check item search process (Fig. 13).

[0057] First, in step S41, the failure cause presentation unit 33 searches for K (a predetermined number) check items from the check items to be investigated T8.

[0058] Next, in step S42, the failure cause presentation unit 33 obtains the travel time T62, transportation costs T63, and labor costs T64 corresponding to the customer name entered by the maintenance worker via the input / output unit 5 from the travel expense table T6.

[0059] Next, in step S43, the failure cause presentation unit 33 calculates the expected cost C of a business trip at a later date according to formula (1-1) if the priority cost is a financial cost, or according to formula (1-2) if the priority cost is a time cost. 後日 P (failure cause) trip is the occurrence probability of the failure mode during the processing of the failure cause estimation result T7. The priority cost is set from the input screen (FIG. 7) described later. Cost C 後日 is an example of the second cost. The cost C based on the formula (1-1) 後日 is the expected value of the cost based on the amount. 後日 is the expected value of the time-based cost. 後日 = P (failure cause) business trip × (T 交通 ×M 人件費 / Ｈ +M 交通費 +T 対応時間 ×M 人件費 / Ｈ ) ... (1-1) C 後日 = P (Cause of failure) Business trip × (T 交通 +T 対応時間 ) ... (1-2)

[0060] Next, in step S44, the failure cause presentation unit 33 calculates the cost C of the investigation time expected for this business trip according to formula (2-1) if the priority cost is a financial cost, or according to formula (2-2) if the priority cost is a time cost. 本日 Calculate the cost C 本日 is an example of the first cost. The cost C based on the formula (2-1) 本日 is the expected value of the cost based on the amount. 本日 is the expected value of the time-based cost. 本日 = [Number of check items K x T 点検時間 +P (cause of failure) x T 対応時間 ]×M 人件費 / Ｈ ... (2-1) C 本日 = [Number of check items K x T 点検時間 +P (cause of failure) x T 対応時間] ... (2-2)

[0061] Next, in step S45, the fault cause presentation unit 33 後日 and C 本日 Ratio C 後日 / C 本日 =C 優先度 Calculate the ratio C 優先度 Based on this, the failure modes are ranked in the form of cost ratio calculation result T9 (FIG. 16).

[0062] (Cost Ratio Calculation Result T9 According to First Embodiment) Fig. 16 is a diagram showing a cost ratio calculation result T9 according to the first embodiment. The cost ratio calculation result T9 includes a failure mode T91, a cost ratio T92, a check item T93, and a probability T94. The cost ratio T92 is calculated by multiplying the ratio C calculated in the failure cause presentation process (Fig. 15). 優先度 is.

[0063] (Input Screen D1 According to First Embodiment) Fig. 17 is a diagram showing the input screen D1 according to the first embodiment. The input screen D1 has an input form for inputting a customer name, an alarm, check boxes for inputting check items, and a check box for inputting a priority cost (a type of cost that is prioritized between financial cost and time cost). After inputting the input information, when the failure cause presentation button is clicked, the failure cause estimation process (Fig. 10), the investigation target check item search process (Fig. 13), and the failure cause presentation process (Fig. 15) are executed in sequence, and the screen transitions to the display of the result screen D2 shown in Fig. 18.

[0064] (Results screen D2 according to the first embodiment) FIG. 18 is a diagram showing the results screen D2 according to the first embodiment. The results screen D2 displays the cost ratio calculation results T9. By referring to the results screen D2, the maintenance personnel can understand the failure modes with high priority and the check items that should be checked. The cost ratio calculation results T9 are not limited to the results screen D2, and may be presented in an interactive format using an interface such as artificial intelligence (AI).

[0065] (Effects of First Embodiment) In the first embodiment, a failure cause that should be dealt with as a priority among failure causes that require a visit is output to a terminal based on the ratio of a first cost in the case where a maintenance worker responds on a current visit to the terminal and a second cost in the case where the failure cause recurs in the next or subsequent visits and the maintenance worker is required to deal with the failure again. Therefore, when a maintenance worker is on a visit, the maintenance worker can recognize a failure mode that should be dealt with as a priority within a limited time and can deal with the failure efficiently.

[0066] In the first embodiment, the cause of a failure that should be addressed with priority is output to the terminal based on the expected value of the monetary cost based on the response time of the maintenance personnel to respond to the failure, the travel time of the maintenance personnel to travel to the customer's site, and the labor costs of the maintenance personnel as the first cost and the second cost. This makes it possible to efficiently respond to failures while reducing wasteful monetary costs.

[0067] In the first embodiment, the cause of the failure to be dealt with on a priority basis is output to the terminal based on the expected value of the time-based cost based on the response time of the maintenance personnel to deal with the failure and the travel time of the maintenance personnel to travel to the customer's site as the first cost and the second cost, thereby making it possible to efficiently deal with the failure while reducing time wastage.

[0068] Furthermore, in the first embodiment, a maintenance knowledge Bayesian network is generated based on the input failure information and maintenance knowledge data, including parent-child relationships that represent the causal relationships from failure modes to functional failures and alarms, and parent-child relationships that represent the causal relationships from failure modes to check items. Furthermore, the occurrence probability of each failure mode in the maintenance knowledge Bayesian network is calculated using the child node abnormality occurrence probability data when the parent node is abnormal and the child node abnormality occurrence probability data when the parent node is normal. Therefore, it is possible to determine the occurrence probability of a failure mode with a high likelihood based on the Bayesian network, and to present failure modes that should be addressed with priority based on this occurrence probability.

[0069] [Embodiment 2] In embodiment 2, a history of past fault cause presentations is used to improve the accuracy of the fault cause presentation unit 33. In the following, in each drawing for explaining this embodiment, differences from embodiment 1 will be mainly explained, and the same components as those in embodiment 1 will be given the same names and symbols, and repeated explanations will be omitted.

[0070] (Equipment maintenance system 1B according to embodiment 2) Fig. 19 is a diagram showing the configuration of an equipment maintenance system 1B according to embodiment 2. Compared to the equipment maintenance system 1 according to embodiment 1, the equipment maintenance system 1B further includes a failure cause presentation and update unit 34 and a failure cause response history database 43.

[0071] (Failure cause response history T10 according to embodiment 2) Fig. 20 is a diagram showing the failure cause response history T10 according to embodiment 2. The failure cause response history T10 is stored in the failure cause response history database 43. The failure cause response history T10 is configured to include information on the work date T101 on which the work was performed, the failure mode T102 which is the investigation result of the work target, and the check item T103 which was entered when the failure mode T102 was identified. The failure cause response history T10 manages the failure mode and check item that were actually handled by the maintenance personnel by referring to the failure cause estimation result T7.

[0072] (Fault Cause Presentation Update Process According to Second Embodiment) FIG. 21 is a flowchart showing fault cause presentation update process according to the second embodiment.

[0073] The failure cause presentation update process is executed after the failure cause presentation process (FIG. 15) of the first embodiment is executed. In the failure cause presentation update process, the past failure cause response history T10 is searched for the same history as the check item input by the maintenance worker this time, and the ratio C of the corresponding failure mode included in the processing result by the failure cause presentation unit 33 is calculated. 後日 / C 本日 is multiplied by a positive parameter α. As a result, the ranking of the corresponding failure mode is increased in the current failure cause presentation result, and the priority of the corresponding failure mode is increased for maintenance personnel. The parameter α is a fixed or variable parameter greater than 1, for example.

[0074] First, in step S51, the failure cause presentation unit 33 reads each record of the cost ratio calculation result T9 (FIG. 16) one by one. Next, in step S52, the failure cause presentation unit 33 refers to the failure cause response history T10 (FIG. 20) and searches for a check item T103 that is the same as the check item T93 included in the record read in step S51.

[0075] Next, in step S53, the failure cause presentation unit 33 determines whether or not the failure cause response history T10 ( FIG. 20 ) contains a check item T103 that is the same as the check item T93 included in the record of the cost ratio calculation result T9 read in step S51. If the same check item T103 is present (step S53: Yes), the failure cause presentation unit 33 proceeds to step S54. On the other hand, if the same check item T103 is not present (step S53: No), the failure cause presentation unit 33 returns the process to step S51.

[0076] In step S54, the failure cause presentation unit 33 determines whether the failure mode T102 corresponding to the check item T103 determined to be identical to the check item T93 in the failure cause response history T10 is identical to the failure mode T91 corresponding to the cost ratio calculation result T9. If the failure mode T102 and the failure mode T91 are identical (step S54: Yes), the failure cause presentation unit 33 proceeds to step S55. On the other hand, if the failure mode T102 and the failure mode T91 are different (step S54: No), the failure cause presentation unit 33 returns the process to step S51.

[0077] In step S55, the failure cause presentation unit 33 multiplies the cost ratio T92 of the cost ratio calculation result T9 for which it was determined in steps S53 and S54 that a record with the same check items and failure mode exists in the failure cause response history T10 by the parameter α.

[0078] Effect of Second Embodiment In the second embodiment, when the failure cause response history contains data with the same failure mode and check item as the cost ratio calculation result, the ratio corresponding to the failure mode and check item in the cost ratio calculation result is multiplied by a predetermined parameter to update the ratio. Therefore, by multiplying by a parameter the ratio of the cost ratio calculation result having the same failure mode and check item as the past failure response history and emphasizing it, the accuracy of the failure cause presentation function can be improved using the history of past failure cause presentations.

[0079] Although the embodiments of the present disclosure have been described in detail above, the present disclosure is not limited to the above-described embodiments and can be modified in various ways without departing from the spirit of the present disclosure. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present disclosure is not necessarily limited to those including all of the described configurations. Furthermore, some of the configurations of the above-described embodiments can be added to, deleted from, or replaced with other configurations.

[0080] Furthermore, the above-described configurations, functional units, processing units, etc. may be partially or entirely implemented in hardware, for example, by designing them as integrated circuits. The above-described configurations, functions, etc. may also be implemented in software by a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function can be stored in a memory, a storage device such as an HDD or SSD, or a recording medium such as an IC card, an SD card, or a DVD.

[0081] In addition, in the above-mentioned drawings, the control lines and information lines shown are those that are considered necessary for explanation, and do not necessarily show all the control lines and information lines that are actually implemented. For example, it may be considered that almost all components are actually connected to each other.

[0082] The above-described arrangement of the processing functions and data is merely an example, and the arrangement of the processing functions and data can be changed to an optimal arrangement in terms of the performance of the hardware and software, processing efficiency, communication efficiency, etc.

[0083] 1, 1B: Equipment maintenance system, 5: Input / output unit, 6: Maintenance worker terminal, 18: Maintenance knowledge data, 31: Failure cause estimation unit, 32: Check item search unit, 33: Failure cause presentation unit, 34: Failure cause presentation update unit, 41: Asset knowledge database, 42: Failure mode response method database, 43: Failure cause response history database, BN: Maintenance knowledge Bayesian network, T1: Maintenance knowledge data, T2: Failure probability data, T3: Child node abnormality occurrence probability when parent node is abnormal, T4: Child node abnormality occurrence probability when parent node is normal, T5: Failure cause response method table, T6: Travel expense table, T7: Failure cause estimation result, T8: Check items to be investigated, T9: Cost ratio calculation result, T10: Failure cause response history.

Claims

1. An equipment maintenance system for maintaining and managing equipment, comprising: An asset knowledge database in which failure information regarding the failure of the equipment, the cause of the failure, the necessity of a business trip by a maintenance staff member to the equipment in the event of the failure, and the cost associated with responding to the failure are managed in association with each other; An input unit that receives input of the failure information regarding the equipment; A failure cause estimation unit that refers to the asset knowledge database based on the failure information received by the input unit and estimates a plurality of the failure causes of the failure; A failure cause presentation unit that calculates a ratio between a first cost when the maintenance staff member responds to the failure cause that requires the business trip among the plurality of failure causes estimated by the failure cause estimation unit during the current business trip, and a second cost when the failure cause occurs again after the next time and the maintenance staff member responds by making another business trip; An output unit that outputs, to a terminal, the failure cause that should be preferentially addressed among the failure causes that require the business trip based on the ratio calculated by the failure cause presentation unit. An equipment maintenance system characterized by having the above components.

2. The equipment maintenance system according to claim 1, further comprising: A failure mode response method database that manages the response time required for the response for each failure cause, the travel time required for the business trip for each customer who installs the equipment, the transportation cost required for the travel of the business trip, and the labor cost of the maintenance staff member; The failure cause presentation unit calculates, as the first cost and the second cost, an expected value of a cost based on the amount based on the response time, the travel time, and the labor cost. An equipment maintenance system characterized by the above features.

3. The equipment maintenance system according to claim 1, further comprising: A failure mode response method database that manages the response time required for the response for each failure cause and the travel time required for the business trip for each customer who installs the equipment; The failure cause presentation unit calculates, as the first cost and the second cost, an expected value of a cost based on the time based on the response time and the travel time. An equipment maintenance system characterized by the above features.

4. The facility maintenance system according to claim 1, wherein the asset knowledge database manages, in association with each other, maintenance knowledge data including a functional failure representing the failure by a phenomenon, a failure mode representing a classification of the functional failure, a check item to be checked by the maintenance staff in the event of the failure mode, and a type of alarm reported in the event of the functional failure, failure probability data for managing the occurrence probability of each of the failure modes, child node abnormal occurrence probability data for managing the probability that the check item corresponding to the failure mode is abnormal and normal in the maintenance knowledge data when the parent node is abnormal, and child node abnormal occurrence probability data for managing the probability that the check item corresponding to the failure mode is abnormal and normal in the maintenance knowledge data when the parent node is normal. The failure cause estimation unit generates a maintenance knowledge Bayesian network including a parent-child relationship representing a causal relationship from the failure mode to the functional failure and the alarm, and a parent-child relationship representing a causal relationship from the failure mode to the check item, based on the failure information received by the input unit and the maintenance knowledge data, and calculates and outputs the occurrence probability of each of the failure modes in the maintenance knowledge Bayesian network using the child node abnormal occurrence probability data when the parent node is abnormal and the child node abnormal occurrence probability data when the parent node is normal. A facility maintenance system characterized by the above.

5. The facility maintenance system according to claim 4, wherein the failure cause presentation unit generates a cost ratio calculation result by associating the failure mode, the ratio, the check item, and the occurrence probability, and the output unit outputs the cost ratio calculation result to the terminal. A facility maintenance system characterized by the above.

6. The facility maintenance system according to claim 5, further comprising a failure cause prompt update unit that stores the failure mode and the check items handled by the maintenance staff as a failure cause response history, wherein when there is data in the failure cause response history that is the same as the cost ratio calculation result, the failure mode, and the check items, the failure cause prompt update unit multiplies a predetermined parameter by the ratio corresponding to the failure mode and the check items in the cost ratio calculation result to update the ratio. A facility maintenance system characterized by the above.

7. A facility maintenance method executed by a facility maintenance system for maintaining and managing facilities, wherein the facility maintenance system includes a processor and an asset knowledge database in which failure information related to a failure of the facility, the cause of the failure, the necessity of a business trip of a maintenance staff to the facility in the event of the failure, and the cost required for dealing with the failure are managed in association with each other. The processor receives an input of the failure information related to the facility, refers to the asset knowledge database based on the received failure information, estimates a plurality of the causes of the failure, calculates a ratio between a first cost when the maintenance staff deals with the cause of the failure that requires a business trip among the plurality of estimated causes of the failure during this business trip and a second cost when the cause of the failure occurs again and the maintenance staff makes another business trip to deal with it, and outputs the cause of the failure that should be preferentially addressed among the causes of the failure that require a business trip based on the ratio. A facility maintenance method characterized by having each of the above processes.

8. The facility maintenance method according to claim 7, wherein the facility maintenance system includes a failure mode response method database that manages the response time required for dealing with each cause of the failure, the travel time required for the business trip for each customer where the facility is installed, the transportation cost required for the travel of the business trip, and the labor cost of the maintenance staff. The processor calculates an expected value of a cost based on an amount based on the response time, the travel time, and the labor cost as the first cost and the second cost. A facility maintenance method characterized by the above.

9. The facility maintenance method according to claim 7, further comprising a failure mode response method database that manages the response time for each cause of failure and the travel time for business trips for each customer where the facility is installed, wherein the processor calculates, as the first cost and the second cost, an expected value of a time-based cost based on the response time and the travel time. A facility maintenance method characterized by the above.

10. The facility maintenance method according to claim 7, wherein the asset knowledge database includes maintenance knowledge data that associates and manages a functional failure representing the failure by a phenomenon, a failure mode representing the classification of the functional failure, check items to be checked by the maintenance personnel in the event of the failure mode, and the type of alarm reported in the event of the functional failure, failure probability data that manages the occurrence probability of each failure mode, child node abnormal occurrence probability data that manages the probability that the check items corresponding to the failure mode are abnormal and normal in the maintenance knowledge data when the failure mode occurs, and child node abnormal occurrence probability data when the parent node is normal that manages the probability that the check items corresponding to the failure mode are abnormal and normal in the maintenance knowledge data when the failure mode has not occurred, and the processor generates a maintenance knowledge Bayesian network including a parent-child relationship representing a causal relationship from the failure mode to the functional failure and the alarm and a parent-child relationship representing a causal relationship from the failure mode to the check items based on the received failure information and the maintenance knowledge data, and calculates and outputs the occurrence probability of each failure mode in the maintenance knowledge Bayesian network using the child node abnormal occurrence probability data when the parent node is abnormal and the child node abnormal occurrence probability data when the parent node is normal. A facility maintenance method characterized by the above.

11. The facility maintenance method according to claim 10, wherein the processor generates a cost ratio calculation result by associating the failure mode, the ratio, the check item, and the occurrence probability, and outputs the cost ratio calculation result. A facility maintenance method characterized by the above.

12. The facility maintenance method according to claim 11, wherein the processor stores the failure mode and the check item addressed by the maintenance staff as a failure cause response history, and when there is data in the failure cause response history where the cost ratio calculation result is the same as the failure mode and the check item, a predetermined parameter is multiplied by the ratio corresponding to the failure mode and the check item in the cost ratio calculation result to update the ratio. A facility maintenance method characterized by this.

Images