JP2022032684A - Equipment maintenance support system and equipment maintenance support method - Google Patents

Abstract

To appropriately support maintenance planning even in the case of a small amount of stored failure history information about sections of facilities.SOLUTION: A facility maintenance support system 1S supporting maintenance of facilities including sections comprises a facility information processing unit which constructs respective lifetime distributions of facilities on the basis of identification information of the facilities and failure times at which failure occurred in the sections of the facilities in failure history information in which the identification information of the facilities, identification information of the sections, and the failure times are stored in association with each other. The facility maintenance support system 1S comprises: a section information processing unit which constructs respective lifetime distributions of the sections of each of the facilities on the basis of the identification information of the facilities, the identification information of the sections, and the failure times in the failure history information; and a failure probability and weight calculation unit which calculates a failure probability of each of the facilities on the basis of the lifetime distribution of each facility and calculates a failure probability of each section of each facility on the basis of the lifetime distribution of each section of each facility.SELECTED DRAWING: Figure 1

Description

The present invention relates to an equipment maintenance support system and an equipment maintenance support method.

In a line that manufactures products and parts, a failure in a certain manufacturing facility may force the production of products to be stopped. This leads to a delay or lengthening of the product manufacturing construction period, and causes a large defect cost. Here, the main measures for the failure that has occurred are the identification of the failure part in the manufacturing equipment and the repair of the corresponding part.

In order to identify the above-mentioned failure site, it is necessary to estimate the cause of the failure from the condition of the equipment. However, there are many cases where a plurality of factors can be considered as the cause of the failure, and it takes a long time to identify the failure site.

In order to avoid these, it is effective to predict equipment failure in advance and to maintain the part where the failure is predicted to occur in advance. However, if maintenance is performed blindly, maintenance costs will be incurred, so it is necessary to identify parts with a high failure probability in order to carry out pre-maintenance.

Patent Document 1 describes a method of calculating a life distribution using the elapsed time until a failure occurs for each part in a semiconductor manufacturing apparatus having a plurality of parts, and obtaining a failure probability and / or a failure rate for each part. It has been disclosed.

Japanese Patent Application Laid-Open No. 2003-257808

In the invention disclosed in Patent Document 1, it is necessary to collect sufficient failure history information for each part in order to calculate the life distribution in the semiconductor manufacturing apparatus. However, in the case of a large-scale manufacturing facility such as that on a production line, there are many parts where failures can occur and the frequency of failures is low, so it takes a lot of time to collect failure history information for each part of the equipment. It ends up.

Therefore, the present invention calculates the failure probability of each part suitable for each equipment even when the accumulation of failure history information of the part of the equipment is small, supports maintenance plan creation, maintenance by failure prediction, or parts. The purpose is to appropriately provide part inventory management support to prevent shortages and surplus inventory.

In order to solve the above problems, the equipment maintenance support system that supports the maintenance of equipment having a part stores the identification information of the equipment, the identification information of the part, and the failure time when the failure occurred in the part of the equipment in association with each other. The equipment information processing unit that builds the life distribution for each equipment based on the equipment identification information of the failure history information and the failure time, the equipment identification information of the failure history information, the identification information of the part, and the above. Based on the failure time, the part information processing unit that builds the life distribution for each part of each equipment, and the failure probability of each equipment is calculated based on the life distribution for each equipment, and the life distribution for each part of the equipment. It is characterized by including a failure probability / weight calculation unit for calculating the failure probability for each part of each equipment based on the above.

According to the present invention, even when the accumulation of failure data of a part of a facility is small, it is possible to calculate the failure probability of each part suitable for each facility and support pre-maintenance and part inventory management. can.

The block diagram which shows an example of the structure of the equipment maintenance support system of Example 1. An explanatory diagram showing an example of the configuration of the manufacturing equipment targeted by the first embodiment. The figure which shows an example of the failure history information stored in a reliability information storage part. The figure which shows an example of the elapsed time information from the failure of each facility to the next failure. The figure which shows an example of the parameter information of the life distribution for each equipment stored in the equipment information storage part. The figure which shows an example of the graph which showed the unreliability function for the failure probability calculation. The figure which shows an example of the elapsed time information from the failure of each facility of each part to the next failure. The figure which shows an example of the elapsed time information which integrated the elapsed time information for each equipment of each part for each part. The figure which shows an example of the parameter information of the life distribution for each part accumulated in a part information storage part. The figure which shows an example of the failure count information for each equipment of each part. The figure which shows an example of the weight information for each part of each equipment which is stored in the weight information storage part. The flowchart which shows an example of the processing flow of the failure probability / weight calculation unit of the equipment maintenance support system of Example 1. The figure which shows an example of the calculation result by Example 1. FIG. The figure which shows an example of the sensing data for each part of each equipment stored in the reliability information storage part when the sensing data is used for the calculation of the weight information in the modification of Example 1. FIG. The figure which shows an example of the tendency information of the sensing data at the time of a failure for each part of each equipment stored in a weight information storage part. The figure which shows an example of the prediction of the temperature measurement value on the maintenance implementation recommended day. The flowchart which shows an example of the processing flow of the failure probability / weight calculation unit in the equipment maintenance support system of Example 2. The figure which shows the output example of the alert screen by Example 2. FIG. The figure which shows an example of the calculation result by Example 2. FIG. The block diagram which shows an example of the structure of the equipment maintenance support system which performs maintenance parts inventory support of Example 3. FIG. The figure which shows an example of the present inventory quantity information which manages by associating the current inventory quantity of a part with the equipment name and a part name which uses a component. The figure which shows an example of the present inventory quantity information 1202 which manages the present inventory quantity for each part. FIG. 5 is a flowchart showing an example of a processing flow of a failure probability / weight calculation unit in an equipment maintenance support system that supports maintenance parts inventory according to the third embodiment. The figure which shows an example of the calculation result by Example 3. FIG. The figure which shows an example of the hardware of the computer which realizes the equipment maintenance support system.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. The examples are examples for explaining the present invention, and are appropriately omitted and simplified for the sake of clarification of the description. The present invention can also be implemented in various other forms. Unless otherwise specified, each component may be singular or plural.

As an example of various information, it may be described by expressions such as "table", "list", and "queue", but various information may be expressed by a data structure other than these. For example, various information such as "XX table", "XX list", and "XX queue" may be "XX information". When describing the identification information, expressions such as "identification information", "identifier", "name", "ID", and "number" are used, but these can be replaced with each other. The storage unit that stores or stores various types of information is a storage or memory having a predetermined storage area.

When there are a plurality of components having the same or similar functions, they may be described by adding different subscripts to the same reference numerals. Further, when it is not necessary to distinguish between these a plurality of components, the subscripts may be omitted in the description.

In the embodiment, a process performed by executing a program may be described. Here, the computer executes a program by a processor (for example, CPU (Central Processing Unit), GPU (Graphics Processing Unit)), and uses a storage resource (for example, memory), an interface device (for example, a communication port), and the like in the program. Perform the specified processing. Therefore, the main body of the processing performed by executing the program may be a processor.

Similarly, the main body of the process for executing the program may be a controller having a processor, a device, a system, a computer, a node, or an XX unit for executing the program for performing the XX process. The main body of the processing performed by executing the program may be any arithmetic unit, and may include a dedicated circuit for performing a specific processing. Here, the dedicated circuit is, for example, an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), a CPLD (Complex Programmable Logic Device), or the like.

The program may be installed on the computer from the program source. The program source may be, for example, a program distribution server or a computer-readable storage medium. When the program source is a program distribution server, the program distribution server includes a processor and a storage resource for storing the program to be distributed, and the processor of the program distribution server may distribute the program to be distributed to other computers. Further, in the embodiment, two or more programs may be realized as one program, or one program may be realized as two or more programs.

In the following examples, the case of estimating the failure probability for each manufacturing facility having a part and each part of each facility based on the past failure history will be described, but the present invention is not limited to the manufacturing equipment, but the part or part. It can be widely applied to various equipment for which maintenance plans are created.

FIG. 1 is a block diagram showing an example of the configuration of the equipment maintenance support system 1S of the first embodiment. The equipment maintenance support system 1S is an information processing system that supports the creation of a maintenance plan for the target equipment, and includes a reliability information storage unit 101, an equipment information processing unit 102, a site information processing unit 103, a weight information processing unit 104, and a failure probability. It has a weight calculation unit 105 and an input / output device 106. The reliability information storage unit 101 and / or the input / output device 106 may be a device included in the equipment maintenance support system 1S or an external device connected to the equipment maintenance support system 1S.

The reliability information storage unit 101 includes failure time indicating the date and time when a failure occurred in the manufacturing equipment (hereinafter referred to as equipment) of products and parts on the manufacturing line, identification information of the faulty equipment, and identification information of the faulty part of the faulty equipment. Failure history information 301 (FIG. 3) configured to include the above is accumulated. The record of the failure history information 301 is acquired from an external system by real-time processing or batch processing, or is acquired by operator input.

The equipment information processing unit 102 builds a life distribution for each equipment. The part information processing unit 103 builds a life distribution for each part of each facility. The part is, for example, a module or a unit constituting the equipment. The details of the life distribution construction process will be described later. The weight information processing unit 104 calculates weight information for each part of each facility. Details of weight information processing will be described later.

The failure probability / weight calculation unit 105 calculates the failure probability of each facility, the failure probability of each part of each facility, and the weight information of each part of each facility, and weights the failure probability of each part of each facility. Make a breakdown. Then, the failure probability / weight calculation unit 105 determines the priority of maintenance work implementation for each part of each facility based on the weighting result of the failure probability for each part of each facility. The input / output device 106 receives an input of an arbitrary failure probability and outputs a calculation result by the equipment maintenance support system 1S to a display unit or the like.

That is, in the equipment maintenance support system 1S, the life distribution for each equipment is constructed by the equipment information processing unit 102 based on the failure history information 301 stored in the reliability information storage unit 101, and each by the part information processing unit 103. The life distribution for each part of the equipment is constructed, and the weight information processing unit 104 calculates the weight information for each part of the equipment. Then, the equipment maintenance support system 1S has a failure input based on the life distribution for each equipment, the life distribution for each part of the equipment, the weight information, and an arbitrary failure probability input from the input / output device 106. The failure probability / weight calculation unit 105 calculates the date and time when each equipment reaches the probability. Further, the equipment maintenance support system 1S calculates the failure probability and weight information for each part of each equipment at the date and time when the failure probability is reached, weights the failure probability for each part of each equipment, and weights the failure probability for each part of the equipment. Determine the maintenance priority. These calculation results are output from the input / output device 106.

The contents processed by such an equipment maintenance support system 1S will be described with reference to FIG. 2, which is a specific example. FIG. 2 is an explanatory diagram showing an example of the configuration of the manufacturing equipment targeted by the first embodiment. In this embodiment, as shown in FIG. 2, there are three equipments (equipment A, equipment B, equipment C) as manufacturing equipment to be managed, and each equipment has five types of parts (part a, part). It is composed of four types of sites, any one of b, site c, site d, and site e). The equipment maintenance support system 1S creates a maintenance plan assuming such a configuration. In this embodiment, the equipment name is used as the information for identifying the faulty equipment, and the part name is used as the information for identifying the faulty part.

FIG. 3 is a diagram showing an example of failure history information 301 stored in the reliability information storage unit 101. The failure history information 301 stores records for each part of the equipment to be managed. As shown in FIG. 3, the failure history information 301 includes a record in which the id of the record, the failure time when the failure occurred, the name of the equipment where the failure occurred, and the name of the failed part are associated with each other. Will be done.

Hereinafter, the life of the equipment and the part is determined by the elapsed time from the failure of the equipment and the part to the occurrence of the next failure. However, by recording the use start time of each facility and each part in the reliability information storage unit 101 together with the failure time when the failure occurs, it may be obtained by the operating time actually used. In this case, since the total operating time actually used from the first start of use can be substituted into the unreliability function (Equation (2)) described later, a more accurate failure probability can be calculated. Hereinafter, it is assumed that the life of the equipment or part is determined by either the elapsed time from the failure of the equipment or part to the next failure or the total operating time of the equipment or part.

Next, the contents processed by the equipment information processing unit 102 will be described. Here, as shown in FIG. 1, it is assumed that the equipment information processing unit 102 is composed of the equipment information calculation unit 111 and the equipment information storage unit 114.

First, the equipment information calculation unit 111 calculates and accumulates the elapsed time from one equipment failure to the next failure from the failure history information 301 of the reliability information storage unit 101. The equipment information calculation unit 111 numerically processes the elapsed time from the failure of each accumulated equipment to the failure, constructs a life distribution for each equipment, and stores the parameters of the life distribution in the equipment information storage unit 114.

Next, a method of constructing a life distribution will be described. FIG. 4A is a diagram showing an example of elapsed time information 401 from one failure of each facility to the next failure. FIG. 4B is a diagram showing an example of parameter information 402 of the life distribution for each equipment stored in the equipment information storage unit 114.

As described above, as a result of calculating the elapsed time from one failure of each facility to the next failure, the elapsed time information 401 as shown in FIG. 4A is obtained. The equipment information calculation unit 111 calculates the parameters of the life distribution of each equipment from the elapsed time information 401.

In this embodiment, the Weibull distribution is applied to all equipment as the life distribution. Here, in this embodiment, the Weibull distribution is applied, but the lifetime distribution is not limited to the Weibull distribution, and a normal distribution, a lognormal distribution, an exponential distribution, or the like can also be applied.

The Weibull distribution is expressed by the following equation (1) using the shape parameter m, the scale parameter η, and the position parameter γ.

The unreliability function F (t), which is the cumulative density function (cumulative failure probability) of the Weibull distribution, is expressed by the following equation (2).

Here, the shape parameter m is a parameter that changes the shape of the wibble distribution. When 0 <m <1, the initial failure with a reduced failure rate, and when m = 1, the accidental failure with a constant failure rate, m> 1. In the case of, it represents a wear deterioration failure of the failure rate increasing type. The scale parameter η is a life scale based on t = η in which the unreliability function F (t) of the equation (2) is 1-e- ¹ (about 63.2%) regardless of the value of m. show. The position parameter γ represents the position on the distribution where the failure starts, and if γ> 0, it means that there is a failure-free period, but normally, it is assumed that there is no failure-free period and γ = 0.

FIG. 5 is a diagram showing an example of a graph showing an unreliability function F (t) for calculating a failure probability. The figure shows the relationship between the unreliability function F (t) obtained by the equation (2) and the failure probability P (t) calculated in this embodiment. In FIG. 5, the horizontal axis represents time and the vertical axis represents probability. In this embodiment, the probability of the vertical axis shown in FIG. 5 is defined as the failure probability P (t) = F (t) of failure within the elapsed time t after the latest failure occurrence.

Further, the equipment information calculation unit 111 extracts the latest final failure time of each equipment from the failure history information 301 of the reliability information storage unit 101. After that, as shown in FIG. 4B, the equipment information calculation unit 111 performs a life distribution associated with the life distribution applied to each equipment, the parameters of the life distribution, and the latest final failure time calculated by the above method. The parameter information 402 of the above is stored in the equipment information storage unit 114.

Next, the contents processed by the part information processing unit 103 will be described. As shown in FIG. 1, the site information processing unit 103 includes a site information calculation unit 112 and a site information storage unit 115.

First, the site information calculation unit 112 calculates and accumulates the elapsed time from one site failure to the next failure for each facility from the failure history information 301 of the reliability information storage unit 101. Then, the part information calculation unit 112 integrates the elapsed time from the failure of each part of the accumulated equipment to the next failure for each part. After that, the part information calculation unit 112 numerically processes the elapsed time integrated for each part, constructs a life distribution for each part, and stores the parameter in the part information storage unit 115.

The method of constructing the life distribution for each part will be described. FIG. 6A is a diagram showing an example of elapsed time information 601 from a failure of each facility of each part to the next failure. As described above, as a result of calculating the elapsed time from the failure of each part of each facility to the next failure, the elapsed time information 601 as shown in FIG. 6A can be obtained.

As described above, in the case of a large-scale equipment such as that on a production line, it is possible that the accumulation of failure data of each part is small depending on the equipment. Therefore, the part information calculation unit 112 creates the elapsed time information 602 that integrates the elapsed time information for each part in all the equipment. FIG. 6B is a diagram showing an example of elapsed time information 602 in which elapsed time information 601 for each facility of each part is integrated for each part. In this way, the life distribution can be constructed by increasing the number of failure data for each part.

Then, the part information calculation unit 112 calculates the parameters of the life distribution of each part from the elapsed time information 602. FIG. 6C is a diagram showing an example of parameter information 603 of the life distribution for each site accumulated in the site information storage unit 115. The method of calculating the parameter of the life distribution for each part is the same as the method of calculating the parameter of the life distribution for each equipment in the equipment information calculation unit 111. Then, as shown in FIG. 6C, the part information calculation unit 112 calculates the life distribution applied to each part, the parameters of the life distribution, and the latest final failure time of each part for each equipment, which is calculated by the above method. The parameter information 603 of the life distribution associated with the above is stored in the site information storage unit 115.

Next, the contents processed by the weight information processing unit 104 will be described. As shown in FIG. 1, the weight information processing unit 104 includes a weight information calculation unit 113 and a weight information storage unit 116. The weight information processing unit 104 calculates the weight according to the difference in the susceptibility to failure for each part appearing between the facilities based on the failure history information 301 of the reliability information storage unit 101. As a result, the weight information processing unit 104 can extract features and tendencies such as susceptibility to failure of different parts in each equipment, and reflect these characteristics and tendencies in the failure probability of each part of each equipment.

First, the weight information calculation unit 113 obtains failure number information 701, which is a total of the number of failures for each equipment of each part as shown in FIG. 7A, based on the failure history information 301 stored in the reliability information storage unit 101. obtain. FIG. 7A is a diagram showing an example of failure frequency information 701 for each equipment of each part.

The weight information calculation unit 113 calculates the weight (weight information) for each facility of each part by using the following equation (3) based on the failure count information 701.
Weight = 1 + (number of failures of the target part in the target equipment / number of failures of the target part in all equipment)
・ ・ ・ (3)

With the weight calculated using the equation (3), the number of failures of the target part can be compared between the equipments having the same part, and the difference in the susceptibility to failure for each part appearing between different equipments can be expressed. .. Here, the weight value may be directly specified by the user. As a result, the difference in the fragility of the equipment parts based on the user's experience can be reflected in the failure probability of each equipment part.

Then, the weight information calculation unit 113 stores the weight information 702 for each part of each equipment calculated by the above method in the weight information storage unit 116. FIG. 7B is a diagram showing an example of weight information 702 for each part of each equipment stored in the weight information storage unit 116.

It should be noted that the failure probability / weight calculation unit 105 uses the failure probability for each part of each facility after weighting it with the weight information 702 in each process, but the present invention is not limited to this, and the failure probability / weight calculation unit 105 is not limited to this and is used for each part of each facility that is not weighted. The failure probability may be used.

Next, the contents processed by the failure probability / weight calculation unit 105 will be described with reference to FIG. FIG. 8 is a flowchart showing an example of the processing flow of the failure probability / weight calculation unit of the equipment maintenance support system of the first embodiment.

First, the failure probability / weight calculation unit 105 receives the failure probability of the target equipment as the implementation level of the maintenance work input by the user via the input / output device 106 (S801). The failure probability of the target equipment, which is the implementation level of the maintenance work, is the failure probability that the maintenance work is carried out for the corresponding equipment if it is equal to or higher than this value.

Next, in the failure probability / weight calculation unit 105, each equipment is set to S801 based on the life distribution parameter and the final failure time included in the parameter information 402 of the life distribution for each equipment stored in the equipment information storage unit 114. The date and time that becomes the received failure probability is calculated and accumulated as the maintenance implementation recommended date (S802). At this time, the recommended maintenance implementation date th for each facility is expressed as the following equation (4) by _modifying the equation (2).

Here, _ts is the final failure time of the target equipment, and p is the failure probability accepted in S801. Further, in the formula (4), γ = 0 is set for simplicity. Further, in this embodiment, since the _Weibull distribution is applied as the life distribution, the calculation formula for calculating the maintenance implementation recommended date th for each facility is the formula (4), but the calculation of the _maintenance implementation recommended date th for each facility is used. The equation depends on the lifetime distribution applied.

Next, the failure probability / weight calculation unit 105 includes the maintenance implementation recommended date calculated in S802, the life distribution parameters included in the life distribution parameter information 603 of each part stored in the part information storage unit 115, and the life distribution parameters of each part. Based on the final failure time for each facility, the failure probability P (t) = F (t) for each part of each facility is calculated by the equation (2) (S803). Specifically, the failure probability / weight calculation unit 105 is the difference between the recommended maintenance implementation date for each facility calculated in S802 and the final failure time for each part of each facility included in the parameter information 603 of the life distribution. By substituting the time t and the parameters of the life distribution into the equation (2), the failure probability P (t) for each part of each equipment is calculated.

Next, the failure probability / weight calculation unit 105 weights the failure probability for each part of the equipment on the recommended maintenance date using the weight information calculated as described above (S804). Next, the failure probability / weight calculation unit 105 determines the maintenance priority of each part in each facility according to the magnitude of the failure probability after weighting in S804 (S805).

Next, the failure probability / weight calculation unit 105 determines the equipment failure probability received in S801, the maintenance implementation recommended date calculated in S802, the failure probability for each part of each equipment calculated in S803, and the maintenance implementation priority calculated in S805. Based on the order, the calculation result 901 that displays the maintenance implementation recommendation date of each equipment, the maintenance implementation priority, the failure probability, and the weight for each part of each equipment on each maintenance implementation recommended date as shown in FIG. 9 is displayed on the screen. Output to the input / output device 106 (S806).

FIG. 9 is a diagram showing an example of the calculation result 901 according to the first embodiment. In FIG. 9, the "equipment failure probability" is the value input in S801, the "maintenance implementation recommended date" is the maintenance implementation recommended date calculated in S802, the "failure probability" is the failure probability calculated in S803, and the "weight" is. The "weight" (Fig. 7B) and "maintenance implementation priority" of the weight information 702 for each part of each equipment are determined according to the magnitude of the value of "part failure probability" x "weight" in the same equipment. It is the implementation order of maintenance of each part.

In this way, even if the accumulation of failure history information for each part of each equipment is small, a more appropriate and accurate maintenance plan is based on the failure probability of each equipment and the maintenance priority order for each part of each equipment. Can be created.

It should be noted that a search function may be provided in which only the information of the corresponding equipment is output as the calculation result 901 by inputting the specific equipment. This has the effect of shortening the calculation time because the recommended maintenance implementation date for only specific equipment is calculated. Further, by determining a threshold value for the recommended maintenance implementation date, a function may be provided in which the recommended maintenance implementation date exceeding the threshold value is not displayed. This has the effect of improving the visibility of necessary equipment information when creating a maintenance plan for a specific period such as monthly.

(Variation example of Example 1)
In the modified example of the first embodiment, the reliability information storage unit 101 stores sensing data such as temperature and vibration representing the operating status of each part of the equipment. Then, the value of the weight information is determined according to whether or not the predicted value of the future sensing data predicted based on the accumulated sensing data corresponds to the condition of the predetermined tendency. In this modification, the weight determined according to whether or not the predicted value of the predicted future sensing data corresponds to the condition of the predetermined tendency is used instead of the weight calculated by the equation (3). ..

FIG. 10 is a diagram showing an example of sensing data 302 for each part of each equipment stored in the reliability information storage unit 101 when the sensing data is used for calculating the weight information in the modified example of the first embodiment. In this modification, the sensing data 302 stores the measured values of temperature and vibration for each part of the equipment.

In this modification, as described above, the reliability information storage unit 101 stores the sensing data 302 for each part of the equipment, and analyzes and analyzes the failure tendency of each part of each equipment based on the sensing data 302. Determine the weight according to the result.

Specifically, the weight information calculation unit 113 extracts the failure history information 301 and the sensing data 302 stored in the reliability information storage unit 101, and based on these, the sensing data when each part fails. Analyze trends. Then, the failure probability / weight calculation unit 105 predicts the value of the sensing data of each part on the future maintenance implementation recommended date by using a time series model or the like. Then, the failure probability / weight calculation unit 105 determines when the value of the sensing data of each part on the predicted future maintenance implementation recommended date corresponds to the tendency of the sensing data when each part fails in advance. Use a large weight value.

When the weight value as described above is used, the tendency information 703 of the sensing data when each part fails, as shown in FIG. 11, is stored in the weight information storage unit 116. FIG. 11 is a diagram showing an example of the tendency information 703 of the sensing data at the time of failure for each part of the equipment stored in the weight information storage unit 116. Further, as described above, the sensing data 302 extracted from the reliability information storage unit 101 is also stored in the weight information storage unit 116 in the same manner.

In this modification, in S804 of the processing flow of the failure probability / weight calculation unit 105 (FIG. 8), the failure probability / weight calculation unit 105 stores weight information for weighting the failure probability for each part of each facility. Trend information 703 (FIG. 11) of sensing data at the time of failure for each part of each equipment stored in unit 116, sensing data 302 (FIG. 10), and maintenance implementation recommended date calculated in S802 (FIG. 8). Calculate from.

Specifically, first, the failure probability / weight calculation unit 105 sets the value of the sensing data of each part on the recommended maintenance date to the ARIMA model (based on the sensing data or the periodicity of the sensing data for a certain period in the past). Predict by time-series analysis using a time-series model such as the autoregressive integrated moving average model). Then, the failure probability / weight calculation unit 105 calculates the weight according to the value of the sensing data of each part on the predicted maintenance implementation recommended date. For example, when the predicted value of the sensing data of each part matches the tendency of the tendency information 703 (FIG. 11) of the sensing data at the time of failure of each part of each equipment, the weight value does not match 2. In this case, the weight value is 1.

FIG. 12 is a diagram showing an example of prediction of temperature measured values on the recommended maintenance implementation date. FIG. 12 is assumed to show a case where, for example, the sensing data (temperature measured value) of the part a of the equipment A is time-series predicted in the prediction interval from the present to the recommended maintenance implementation date. In FIG. 12, it is predicted that the temperature prediction value on the maintenance implementation recommended date based on the past time-series data before the present of the part a of the equipment A deviates from the distribution average by 3σ. Since this prediction matches the tendency of the part a of the equipment A recorded in the tendency information 703 of FIG. 11, the weight value is calculated as 2.

In this modification, as described above, it is possible to calculate the weight information that reflects the time information and the actual situation of the equipment and the part. Therefore, it is possible to perform equipment maintenance more appropriately and accurately according to the actual conditions of the equipment and parts.

The weight information of this modification can also be used in other examples described later. For example, in Example 2 described later, the weight is determined based on the tendency of the predicted value of the sensing data at an arbitrary date in the future for which failure is to be predicted. Further, in Example 3 described later, the weight is determined based on the tendency of the predicted value of the sensing data at an arbitrary date in the future for which inventory information is desired to be confirmed.

In the second embodiment, the failure probability for each part of each equipment is calculated based on the combination of the failure probability for each equipment and the failure probability for each part, the failure is predicted based on the calculated failure probability, and the pre-maintenance is carried out. Assist. In this embodiment, the failure probability for each part of each facility on a future date set by the user is calculated based on the formula (5) described later. Then, by outputting an alert from the input / output device 106 for the part of the equipment whose failure probability exceeds the threshold value set by the user, it is possible to predict the failure and support the pre-maintenance.

The configuration of the equipment maintenance support system 2S of this embodiment is the same as that of FIG. However, in this embodiment, unlike the first embodiment, the failure probability / weight calculation unit 105 describes the failure probability of each equipment, the failure probability of each part, and the weight information for each part of each equipment, which will be described later. It applies to equation (5). That is, the failure probability for each part of each equipment is calculated by multiplying the result of weighting the failure probability of each part by the failure probability of each equipment. The failure probability / weight calculation unit 105 accepts the input of the failure probability threshold for which the maintenance recommendation alert of the equipment part is to be output at any future date for which failure is to be predicted, and is calculated by the equation (5) on this future date. A list of equipment parts that exceed the failure probability threshold is output as an alert.

Hereinafter, as in Example 1, three facilities composed of four types of parts as shown in FIG. 2 will be considered. First, by the same processing as in the first embodiment, the parameter information of the life distribution of each equipment and the parameter information 402 of the life distribution for each equipment including the last failure time are stored in the equipment information storage unit 114. Similarly, the parameter information of the life distribution of each part and the parameter information 603 of the life distribution for each part including the last failure time for each equipment of each part are stored in the part information storage unit 115. Similarly, the weight information 702 for each part of each equipment is stored in the weight information storage unit 116.

Next, the contents processed by the failure probability / weight calculation unit 105 will be described with reference to FIG. FIG. 13 is a flowchart showing an example of the processing flow of the failure probability / weight calculation unit 105 in the equipment maintenance support system 2S of the second embodiment.

First, the failure probability / weight calculation unit 105 is a part of each facility that is input via the input / output device 106 and wants to output an arbitrary future date (failure probability calculation date or scheduled safety implementation date) for which failure is to be predicted and an alert. The input of the threshold value of the failure probability for each is accepted (S1001). In this embodiment, for example, an arbitrary date in the future for which failure is to be predicted is entered 7 days after the input date (today: the date on which the alert screen 1101 described later is displayed (2019/07/01)) (2019/07/08). ), And the case where the threshold value of the failure probability is 60% is shown.

Next, the failure probability / weight calculation unit 105 uses the optional future date input in S1001 and the parameters and equipment of the life distribution included in the parameter information 402 of the life distribution of each equipment stored in the equipment information storage unit 114. Based on the final failure time for each, the failure probability of each facility is calculated by the equation (2) (S1002). Specifically, the failure probability / weight calculation unit 105 determines the elapsed time t, which is the difference between the arbitrary future date input in S1001 and the final failure time of each facility included in the parameter information 402 of the life distribution. By substituting the life distribution parameter into Eq. (2), the failure probability of each equipment is calculated.

Next, the failure probability / weight calculation unit 105 includes the future arbitrary date input in S1001, the life distribution parameters included in the life distribution parameter information 603 of each part stored in the part information storage unit 115, and each. Based on the final failure time for each part of the equipment, the failure probability for each part of each equipment is calculated by the equation (2) (S1003). Specifically, the failure probability / weight calculation unit 105 is the elapsed time, which is the difference between the arbitrary future date input in S1001 and the final failure time of each part included in the parameter information 402 of the life distribution. By substituting t and the parameter of the life distribution into the equation (2), the failure probability for each equipment of each part is calculated.

Next, the failure probability / weight calculation unit 105 calculates the failure probability of each equipment portion by the following equation (5) (S1004).
Equipment failure probability = Part failure probability x Weight x Equipment failure probability ... (5)

The "probability of failure of a part" in the formula (5) is the probability calculated by S1003. Further, the “equipment failure probability” in the equation (5) is the probability calculated in S1002. Further, the "weight" in the equation (5) is a weight value for each part of the equipment of the weight information 702 stored in the weight information storage unit 116, or each calculated as in the modification of the first embodiment. It is the value of the weight for each part of the equipment. Here, depending on the weight value, the value calculated by the equation (5) may be 100% or more, but in that case, the scale is reduced to 99%.

Next, the failure probability / weight calculation unit 105 extracts information on the part of the equipment whose failure probability of the equipment part calculated by the equation (5) is equal to or higher than the threshold value of the failure probability accepted in S1001 (S1005).

Next, the failure probability / weight calculation unit 105 is as shown in FIG. 14A based on the failure probability calculation date received in S1001, the failure probability of each equipment calculated in S1002, and the failure probability of each part calculated in S1003. The alert screen 1101 for displaying the parts of each equipment that are equal to or greater than the failure probability threshold on the failure calculation date is output to the input / output device 106 (S1006).

FIG. 14A is a diagram showing an output example of the alert screen 1101 according to the second embodiment. As shown in FIG. 14A, the alert screen 1101 includes a basic information screen 1102 and a maintenance recommended equipment site output screen 1103.

On the basic information screen 1102, a box 1104 that displays the date of the day on which the screen is displayed, a box 1105 that displays the date on which the failure probability is calculated, a box 1106 that displays the input failure probability threshold, and the like are displayed. .. Further, on the maintenance recommended equipment part output screen 1103, a box 1107 or the like for displaying the equipment name and the part name of the part of each equipment extracted in S1005 (FIG. 13) is displayed.

Further, the failure probability / weight calculation unit 105 records the calculated failure probability and failure probability calculation date, the failure probability of the equipment, and the failure probability and weight of the part together with each equipment name and each part name as shown in FIG. 14B. The result 1108 is output (S1007). FIG. 14B is a diagram showing an example of the calculation result 1108 according to the second embodiment.

In this embodiment, the information in the alert screen 1101 and the calculation result 1108 is used to predict the failure in each part of the equipment in real time and output it as an alert even when the number of failure data of the parts of the equipment is small. , Support for pre-maintenance can be provided more quickly, appropriately and accurately.

In the third embodiment, appropriate parts inventory management is supported by comparing the number of parts required for maintenance calculated based on the failure probability of equipment and parts with the current number of parts inventories. FIG. 15 is a block diagram showing an example of the configuration of the equipment maintenance support system 3S that supports the inventory of maintenance parts according to the third embodiment.

As shown in FIG. 12, the equipment maintenance support system 3S of this embodiment further has an inventory information processing unit 107 that aggregates and stores the current number of parts in stock for each part in the equipment maintenance support system 1S of FIG. The inventory information processing unit 107 has an inventory information calculation unit 117 and an inventory information storage unit 118. The inventory information calculation unit 117 compares the number of parts required for maintenance of each part with the current number of parts in stock, and stores the comparison result in the inventory information storage unit 118.

Hereinafter, as in Example 1, three facilities composed of four types of parts as shown in FIG. 2 will be considered. First, by the same processing as in the first embodiment, the parameter information of the life distribution of each equipment and the parameter information 402 of the life distribution for each equipment including the last failure time are stored in the equipment information storage unit 114. Similarly, the parameter information of the life distribution of each part and the parameter information 603 of the life distribution for each part including the last failure time for each equipment of each part are stored in the part information storage unit 115. Similarly, the weight information 702 for each part of each equipment is stored in the weight information storage unit 116.

Next, the contents processed by the inventory information processing unit 107 will be described. In this embodiment, the inventory information processing unit 107 includes an inventory information calculation unit 117 and an inventory information storage unit 118. FIG. 15 is a block diagram showing an example of the configuration of the equipment maintenance support system 3S that supports the inventory of maintenance parts according to the third embodiment.

In the inventory information processing unit 107, first, the inventory information calculation unit 117 first collects the current inventory of parts for each part based on the current inventory quantity information 1201 stored in the factory recorded in the reliability information storage unit 101. The current inventory quantity information 1202 that aggregates the numbers is created and stored in the inventory information storage unit 118.

The inventory information calculation unit 117 creates the current inventory quantity information 1202 at a predetermined cycle or when the current inventory quantity information 1201 is updated. FIG. 16A is a diagram showing an example of the current inventory quantity information 1201 that manages the current inventory quantity of parts and the equipment name and part name that use the parts in association with each other. FIG. 16B is a diagram showing an example of the current inventory quantity information 1202 that manages the current inventory quantity for each part.

Next, the contents processed by the failure probability / weight calculation unit 105 will be described with reference to FIG. FIG. 17 is a flowchart showing an example of the processing flow of the failure probability / weight calculation unit 105 in the equipment maintenance support system that supports the inventory of maintenance parts according to the third embodiment.

First, the failure probability / weight calculation unit 105 receives the date for which inventory information is desired to be confirmed, which is input via the input / output device 106 (S1301). Next, the failure probability / weight calculation unit 105, as in S1002 of the second embodiment, has the date on which the inventory information input in S1301 is desired to be confirmed and the parameter information of the life distribution of each equipment stored in the equipment information storage unit 114. Based on the life distribution parameter included in 402 and the final failure time for each facility, the failure probability of each facility is calculated by the equation (2) (S1302).

Next, the failure probability / weight calculation unit 105, as in S1003 of the second embodiment, has the date on which the inventory information input in S1001 is to be confirmed and the parameter information of the life distribution of each part stored in the part information storage unit 115. Based on the life distribution parameter included in 603 and the final failure time for each facility of each part, the failure probability for each target part of each facility is calculated by the equation (2) (S1303).

Next, the required inventory calculation unit 108 calculates the maintenance required inventory quantity of each part by the following formula (6) (S1304).
Maintenance required stock quantity of target part = Σ _{(all equipment)} [Failure probability of target part x weight _{(each equipment)} x failure probability of _{equipment (each equipment)} ]… (6)

Σ in the equation (6) means that the multiplication value of the failure probability and the weight of each equipment in the square brackets and the failure probability of each equipment is summed over all the equipments having the corresponding parts. .. Further, the “probability of failure of the target portion” in the equation (6) is the probability calculated in S1303. Further, the “equipment failure probability _{(each equipment)} ” in the equation (6) is the probability calculated by S1302. Further, the "weight _{(each equipment)} " in the equation (6) is the value of the weight for each part of each equipment of the weight information 702 for each part of each equipment stored in the weight information storage unit 116, or the first embodiment. It is a weight value for each part of each equipment calculated as in the modified example of. Here, if the number of equipment is small, the number of maintenance-required stocks of the part calculated by the formula (6) may be less than 1, but in that case, the maintenance-required stock number of the corresponding part is set to 1.

Next, the required inventory calculation unit 108 includes the current inventory quantity of the current inventory quantity information 1202 stored in the inventory information storage unit 118, and the maintenance required inventory quantity of the target part calculated by the failure probability / weight calculation unit 105 in S1304. , Is calculated for each part (S1305).

Next, as shown in FIG. 18, the required inventory calculation unit 108 includes the name of the equipment having the part using the corresponding part for each part of each part on the date when the inventory information is desired to be confirmed, and the target calculated in S1304. The calculation result 1401 for displaying the difference between the maintenance required inventory quantity of the part, the current inventory quantity of the current inventory quantity information 1202, and the current inventory quantity and the maintenance required inventory quantity calculated in S1305 is output to the input / output device 106 ( S1306).

FIG. 18 is a diagram showing an example of the calculation result 1401 according to the third embodiment. FIG. 18 shows the calculation result 1401 on the date when the inventory information input in S1301 (FIG. 17) is desired to be confirmed, and the “part name”, “equipment name”, “part name”, and “current inventory quantity” are the current inventory quantity information 1202. It is the same as "part name", "equipment name", "part name", and "current inventory quantity" in. Further, the "maintenance required inventory quantity" of the calculation result 1401 is the maintenance required inventory quantity for each part calculated in S1304, and the "(current inventory quantity)-(maintenance required inventory quantity)" is the current inventory for each part calculated in S1305. It is the difference between the number and the number of stocks required for maintenance.

In this embodiment, even if the number of failure data of the parts of the equipment is small based on the information of the calculation result 1401, the number of stocks required for maintenance is calculated and managed based on the failure probability of each part of the equipment. By comparing with the current number of parts in stock, it is possible to reduce the occurrence of surpluses and shortages.

FIG. 19 is a diagram showing an example of computer hardware that realizes the equipment maintenance support systems 1S, 2S, and 3S. In the computer 5000, the processor 5100, the memory 5200, the storage 5300, the network interface 5400, the input device 5500, and the output device 5600 are connected via the bus 5700. The processor 5100 is a CPU (Central Processing Unit) or the like. The memory 5200 is a RAM (Random Access Memory) or the like. The storage 5300 is an HDD (Hard Disk Drive), an SSD (Solid State Drive), a medium reader, or the like. The input device 5500 is a keyboard, a mouse, a touch panel, or the like. The output device 5600 is a display or the like.

In the computer 5000, each program for realizing the equipment maintenance support systems 1S, 2S, and 3S is read from the storage 5300 and executed in cooperation with the processor 5100 and the memory 5200, whereby the equipment maintenance support system 1S, 2S and 3S are realized respectively. Alternatively, each program for realizing the equipment maintenance support systems 1S, 2S, and 3S may be acquired from an external computer by communication via the network interface 5400. Alternatively, each program for realizing the equipment maintenance support systems 1S, 2S, and 3S is recorded on a portable recording medium (optical disk, semiconductor storage medium, etc.) and read by a medium reader to be read by the processor 5100 and the processor 5100. It may be executed by the cooperation of the memory 5200.

The equipment maintenance support systems 1S, 2S, and 3S may be realized on one computer 5000, and the functional units of the equipment maintenance support systems 1S, 2S, and 3S may be appropriately distributed and arranged on a plurality of computers 5000. May be done. Further, the equipment maintenance support systems 1S, 2S, and 3S may be constructed in the cloud or on-premises.

The present invention is not limited to the above-described embodiment, and includes various modifications. For example, the above-described embodiment has been described in detail in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to the one including all the described configurations. Further, as long as there is no contradiction, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment and add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to add, delete, replace, integrate, or disperse a part of the configuration of each embodiment. Further, the configurations and processes shown in the embodiments can be appropriately dispersed, integrated, or replaced based on the processing efficiency or the mounting efficiency.

1S, 2S, 3S: Equipment maintenance support system, 101: Reliability information storage unit, 102: Equipment information processing unit, 103: Part information processing unit, 104: Weight information processing unit, 105: Failure probability / weight calculation unit, 106 : Input / output device, 107: Inventory information processing unit, 108: Required inventory calculation unit, 111: Equipment information calculation unit, 112: Part information calculation unit, 113: Weight information calculation unit, 114: Equipment information storage unit, 115: Part Information storage unit, 116: Weight information storage unit, 117: Inventory information calculation unit, 118: Inventory information storage unit

Claims (13)

It is an equipment maintenance support system that supports the maintenance of equipment that has parts.
Life distribution for each equipment based on the equipment identification information and the failure time of the failure history information accumulated by associating the equipment identification information and the part identification information with the failure time when the failure occurred at the part of the equipment. With the equipment information processing department that builds
A part information processing unit that builds a life distribution for each part of each equipment based on the identification information of the equipment, the identification information of the part, and the failure time of the failure history information.
It is equipped with a failure probability / weight calculation unit that calculates the failure probability of each equipment based on the life distribution of each equipment and calculates the failure probability of each part of each equipment based on the life distribution of each part of the equipment. Equipment maintenance support system characterized by this. The equipment maintenance support system according to claim 1, further comprising a reliability information storage unit that stores at least the failure history information. The life distribution for each equipment and the life distribution for each part of the equipment are determined by either the elapsed time from the failure of the equipment or the part to the next failure, or the total operating time of the equipment or the part. The equipment maintenance support system according to claim 1, wherein the equipment is to be maintained. Further provided with a weight information processing unit that calculates weight information according to the difference in failure susceptibility for each part between different facilities based on the failure history information.
The failure probability / weight calculation unit is
The equipment maintenance support system according to claim 1, wherein the failure probability for each part of the equipment weighted by the weight information is used. The weight information is
The equipment according to claim 4, wherein the equipment is calculated based on the ratio of the number of failures of the target part in the target equipment to the total number of failures of the target parts in all the equipments based on the failure history information. Maintenance support system. The weight information is
If the time-series tendency of the data indicating the operating status detected for each part of each facility matches the tendency at the time of failure, the first value is set, and if it does not match, the second value is smaller than the first value. The equipment maintenance support system according to claim 4, wherein the equipment is calculated as a value of. The failure probability / weight calculation unit is
Based on the life distribution of each equipment, the recommended maintenance date for which the failure probability of the equipment is equal to or higher than the threshold value is calculated for each equipment.
The equipment maintenance support system according to claim 1, wherein the failure probability for each part of the equipment on the recommended maintenance implementation date is calculated based on the life distribution of each part of the equipment. The failure probability / weight calculation unit is
Based on the failure probability of each part of each equipment on the recommended maintenance date, the priority of maintenance implementation of each part in each equipment is determined.
The equipment maintenance support system according to claim 7, wherein the priority of each part in each of the determined equipments is output to an output device. The failure probability / weight calculation unit is
Based on the life distribution of each equipment, the failure probability of the equipment on the scheduled maintenance date is calculated.
Based on the life distribution of each part of the equipment, the failure probability for each part of the equipment on the scheduled maintenance date is calculated.
The equipment maintenance support system according to claim 1, wherein the equipment maintenance support system is calculated based on the failure probability of the equipment and the failure probability of each part of the equipment on the scheduled maintenance date. The failure probability / weight calculation unit is
Information on equipment and parts where the failure probability for each equipment and part is equal to or higher than the threshold value is extracted.
The equipment maintenance support system according to claim 9, wherein the extracted information on the equipment and parts is output to an output device. The failure probability / weight calculation unit is
Based on the life distribution of each equipment, the failure probability of the equipment on the inventory information confirmation date is calculated.
Based on the life distribution of each part of each equipment, the failure probability for each part of each equipment on the inventory information confirmation date is calculated.
An inventory information processing unit that creates and manages current inventory quantity information in which the identification information of the portion, the identification information of the parts used by the portion, and the current inventory quantity of the parts are associated with each other.
With the required inventory quantity calculation unit that calculates the maintenance required inventory number required for maintenance for the part and each part based on the failure probability for each equipment and the failure probability for each part of the equipment on the inventory information confirmation date. The equipment maintenance support system according to claim 1, further comprising. The required inventory quantity calculation unit is
Based on the current inventory quantity information, the difference between the maintenance required inventory quantity and the current inventory quantity is calculated for each of the parts and the parts.
The equipment maintenance support system according to claim 11, wherein the maintenance required inventory number, the current inventory quantity, and the difference for each of the parts and the parts are output to the output device. It is an equipment maintenance support method executed by the equipment maintenance support system that supports the creation of maintenance plans for equipment that has parts.
Life distribution for each equipment based on the equipment identification information and the failure time of the failure history information accumulated by associating the equipment identification information and the part identification information with the failure time when the failure occurred at the part of the equipment. Equipment information processing to build
Part information processing that builds a life distribution for each part of each facility based on the identification information of the equipment, the identification information of the part, and the failure time of the failure history information.
It is provided with a failure probability / weight calculation process that calculates the failure probability of each equipment based on the life distribution of each equipment and calculates the failure probability of each part of each equipment based on the life distribution of each part of the equipment. Equipment maintenance support method characterized by this.

Images