KR20190071812A - Failure Risk Indicator Estimator and Failure Risk Indicator Estimation Method - Google Patents

Abstract

The model equation building unit 4a constructs a model equation representing the transition of the failure risk index according to the statistical distribution based on the FMEA result read from the FMEA result database 2 and the statistical evaluation information A. [ The parameter estimating unit 4b calculates the difference between the maintenance work interval calculated from the model equation and the already determined maintenance work interval based on the information read from the FMEA result DB 2 and the already determined work interval DB 3 And statistically estimates parameter values of the model expression.

Description

Failure Risk Indicator Estimator and Failure Risk Indicator Estimation Method

The present invention relates to a failure risk index estimating apparatus and an failure risk index estimating method for estimating an index of a failure in which a failure occurs in a facility.

The failure mode influence analysis (hereinafter referred to as FMEA) is an analysis technique for obtaining the result of assigning a level to each item related to failure in terms of the influence of the failure on the target equipment. The items concerning the failure include, for example, the frequency of occurrence of a failure in the facility and the magnitude of the influence of the failure on the facility. In addition, when it is difficult to evaluate the necessity of repair or repair of the equipment from the operation data because the operation data of the facility is not sufficiently accumulated, the result of analysis of the FMEA may be referred to when evaluating this.

For example, Patent Document 1 describes an operation prediction apparatus for predicting the operation rate of a mechanical system to be predicted by using an FMEA. The operation-amount predicting apparatus includes a map in which a correspondence relationship between a level of a plurality of evaluation items related to the failure rate and a failure coefficient is defined. With reference to the map, And specifies a corresponding failure factor. The operating rate predicting device estimates the failure rate of the component from the specific failure factor and predicts the operating rate of the mechanical system to be predicted based on the failure rate estimated by each of the plurality of components. The map is fitted to match the performance of the operating rate of the similar mechanical system.

On the other hand, in a stage where an attempt is made to accurately evaluate the risk of a failure occurring in a target facility, maintenance and management of the facility are already underway, and maintenance intervals are often determined according to the situation of the facility. Therefore, conventionally, a technique has been proposed that calculates the maintenance work interval expected for each component based on a failure factor of each component of the target facility.

For example, in the movable management apparatus described in Patent Document 2, the factor of the component of the component is selected in consideration of the factor of failure and the influence of the failure of the component in the static state, and the predicted life of the component The degradation rank coefficient of the component is selected. Based on the selected part rank coefficient and the deterioration rank coefficient, the abovementioned movable management apparatus calculates the inspection period expected for the component part and determines the effective maintenance period of the component, that is, the maintenance interval, based on the calculated inspection period .

Patent Document 1: JP-A-2016-126728 Patent Document 2: Japanese Patent Application Laid-Open No. 2004-252549

In order to obtain the map, the operating rate predicting apparatus disclosed in Patent Document 1 requires historical data sufficient to accurately reproduce the performance of the operating rate of the mechanical system similar to the mechanical system to be predicted.

Further, in the above-described operating rate predicting apparatus, only the magnitude of the failure rate is considered between the different levels of the evaluation items. For example, the operating rate is predicted under the constraint condition that a failure rate at which a fault with low frequency level of faults occurs is lower than a failure rate at which a fault with high frequency level of faults occurs. Therefore, it is not possible to appropriately evaluate the variation of the failure rate in an actual mechanical system.

In the movable management apparatus disclosed in Patent Document 2, the effective maintenance period is determined on the premise that the inspection period expected for the component is a constant multiple of the current inspection period. However, the premise is that the risk of failure is linear If we do not change, we do not establish.

Since the risk that a failure occurs in an actual facility generally changes nonlinearly, the above-described movable management apparatus can not be applied to an actual facility as it is.

The present invention solves the above problems and provides a failure risk index estimating apparatus and a failure risk index estimating method capable of appropriately estimating an index of a risk that a failure occurs in a facility, .

The failure risk index estimating apparatus according to the present invention comprises a model expression building unit and a parameter estimating unit. The model expression construction unit is a model expression unit that represents the transition of the failure risk index according to the statistical distribution based on the information indicating the FMEA result for each maintenance item of the components constituting the facility and the information indicating the statistical distribution used for the estimation of the failure risk index . The parameter estimating section calculates the maintenance work interval calculated from the model formula based on the information indicating the FMEA result for each maintenance item of the parts constituting the facility and the information indicating the maintenance work interval determined for each maintenance item of the part, And the parameter value of the model equation in which the difference of the interval becomes the smallest is statistically estimated.

According to the present invention, a model equation representing the transition of the failure risk index according to the statistical distribution is constructed, and the parameter value of the model equation is statistically estimated as an estimate of the failure risk index. By doing so, it is possible to appropriately estimate the index of the risk that a failure occurs in the facility, even when there is no or less maintenance data of the facility.

1 is a block diagram showing a functional configuration of a failure risk index estimating apparatus according to Embodiment 1 of the present invention.
2A is a diagram showing items of information stored in the FMEA result database (hereinafter referred to as DB). 2B is a diagram showing items of information stored in the already determined work interval DB. 2C is a diagram showing statistical evaluation information.
3 is a diagram showing items of information stored in the first storage unit in the first embodiment.
4A is a block diagram showing a hardware configuration for realizing the function of the failure risk indicator estimating device according to the first embodiment. 4B is a block diagram showing a hardware configuration for executing software for realizing the function of the failure risk indicator estimating device according to the first embodiment.
5 is a flowchart showing the operation of the failure risk indicator estimating apparatus according to the first embodiment.
6 is a diagram showing the relationship between the failure risk index and the intermediate evaluation index.
7 is a diagram showing the relationship between the evaluation items of the FMEA result and the intermediate evaluation index.
8 is a flowchart showing the operation of the model expression building unit and the parameter estimating unit.
9 is a diagram showing the result of merging the FMEA result DB and the table data of the already determined work interval DB.
10 is a diagram showing a result of assigning the parameter of the intermediate evaluation index to the evaluation item of the FMEA result.
11 is a diagram showing the maintenance work interval assumed from the model equation.
12 is a block diagram showing the functional configuration of a failure risk indicator estimating apparatus according to Embodiment 2 of the present invention.
13A is a diagram showing items of information stored in the facility information DB. 13B is a diagram showing items of information stored in the maintenance record DB. 13C is a diagram showing items of information stored in the failure history DB.
14A is a diagram showing items of information stored in the reduced-size data storage unit. 14B is a diagram showing items of information stored in the second storage unit. 14C is a diagram showing an estimate of the failure risk index.
15 is a flowchart showing the operation of the failure risk indicator estimating apparatus according to the second embodiment.
16 is a flowchart showing the operation of the range reduction unit in the second embodiment.
17 is a flowchart showing the operation of the land estimating unit in the second embodiment.
18 is a flowchart showing the operation of the merging unit in the second embodiment.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments for carrying out the present invention will be described with reference to the accompanying drawings.

Embodiment Mode 1.

Fig. 1 is a block diagram showing a functional configuration of a failure risk indicator estimation device 1 according to Embodiment 1 of the present invention. The failure risk indicator estimating apparatus 1 is an apparatus for estimating a failure risk indicator of a facility and includes an FMEA result DB 2, a previously determined work interval DB 3, an index estimating unit 4, and a first storing unit 5 . The index estimating unit 4 estimates the failure risk index of the facility based on the information read from each of the FMEA result DB 2 and the already determined work interval DB 3 and the statistical evaluation information A. Failure risk index is information that quantifies the size of the risk that a failure occurs in the components constituting the equipment.

The FMEA result DB (2) is a DB that stores FMEA results for each maintenance item of the components constituting the equipment. In the FMEA result DB 2, for example, items of information such as those shown in Fig. 2A are stored. In the item of "part ID ", the identification information of the part is set. In the item of "check item ID", the identification information of the check item is set. The inspection items include, for example, appearance inspection, energization inspection, insulation inspection, and friction inspection. Since both the energization inspection and the insulation inspection are to check the conduction state of electricity, they can be said to be similar inspection items.

"Fault frequency level" and "magnitude level of influence" and "detectability level" are evaluation items of FMEA. In the item of "fault frequency level ", an evaluation level of the FMEA regarding the frequency of failure is set. In the item of "magnitude level of influence", the evaluation level of the FMEA is set with respect to the magnitude of the influence of the failure on the component. An evaluation level of the FMEA concerning the ease of detecting a failure is set in the item of "detectability level ".

The already determined work interval DB 3 is a DB that stores information indicating the maintenance work interval already determined for each item of the inspection of the parts. In the already determined work interval DB 3, for example, information of an item as shown in FIG. 2B is stored. The "part ID" and "check item ID" are the same as those shown in FIG. The number of months of the maintenance work interval already determined for each maintenance item of the part is set in the item of "already determined work interval months".

The statistical evaluation information A is information indicating a statistical distribution used for statistical estimation of the failure risk index. In the statistical evaluation information A, for example, information on the item as shown in Fig. 2C is stored. In the item of "statistical distribution", the kind of statistical distribution used for the statistical estimation of the failure risk index is set. Types of statistical distribution include the theoretically derived distribution of Weibull distribution, gamma distribution, and lognormal distribution.

In the item of the "tolerance ", a ratio of an allowable error to the standard deviation of the estimated value of the failure risk index estimated by the index estimating unit 4 is set.

In the item of the "reliability rate ", the reliability ratio of the failure risk index estimated by the index estimation unit 4 is set. If the result of evaluating that the estimated value of the failure risk indicator falls within the tolerance is equal to or greater than the value of the deviation probability P1 and the result of evaluating that the estimated value of the failure risk indicator does not fall within the tolerance, (P1?, P2?).

Since the "tolerance" and the "reliability ratio" are information used for the merging process described later in the second embodiment, the statistical evaluation information A in the first embodiment may not include these pieces of information.

The index estimating unit 4 includes a model equation building unit 4a and a parameter estimating unit 4b.

The model expression construction unit 4a constructs a model expression representing the transition (failure) of the failure risk index based on the FMEA result read from the FMEA result DB 2 and the statistical evaluation information A.

Based on the information read from the FMEA result DB 2 and the already-read working space DB 3, the parameter estimating unit 4b finds the smallest difference between the maintenance work interval calculated from the model equation and the already determined maintenance work interval The parameter value of the model equation is statistically estimated.

The model equation is a model of a transition of the failure risk index according to the statistical distribution represented by the statistical evaluation information A, and the parameter value of this model equation is the first estimation value.

The first storage unit (5) stores a first estimate value that is a parameter value estimated by the index estimation unit (4). In the first storage unit 5, for example, information for each item as shown in Fig. 3 is stored. "Part ID" and "check item ID" are the same as those shown in Fig. 2A, and "statistical distribution"

In the item of "statistical distribution parameter ", a parameter is set which specifies the statistical distribution used in the statistical estimation of the first estimate. If the statistical distribution follows a Weibull distribution, the shape parameter γ and the measure parameter φ of the cumulative density function defining the Weibull distribution are set.

The "time scale factor "," risk weight factor ", and "safety margin" are parameters of the above-described model equation.

The "time scale factor" is a parameter related to the rate at which the failure risk increases in the above model equation. The "risk weighting factor" is a parameter related to the degree of weighting of the failure risk in the above model equation. The "safety margin" is a parameter indicating a time interval arising from the maintenance work date and time in the work interval estimated in the model formula. With this parameter, the maintenance work of the part is executed ahead of time.

The FMEA result DB 2 and the already determined work interval DB 3 are the database 100 shown in Figs. 4A and 4B. The information stored in each of the FMEA result DB 2 and the already determined work interval DB 3 is input to the index estimator 4 via the DB input /

The statistical evaluation information A is input to the failure risk index estimation device 1 via the information input interface 102. [ The estimated value of the failure risk index is output from the failure risk index estimation device 1 via the information output interface 103. [

The first storage unit 5 may be provided in the storage device having the database 100, but it may be provided in the internal memory of the processing circuit 104 shown in Fig. 4A. The first storage unit 5 may be provided in the memory 105 shown in Fig. 4B.

Each function of the model equation building unit 4a and the parameter estimating unit 4b in the failure risk index estimating apparatus 1 is realized by a processing circuit.

That is, the failure risk index estimating apparatus 1 constructs a model equation representing the transition of the failure risk index based on the FMEA result read from the FMEA result DB 2 and the statistical evaluation information A, 2) and the information read from the already determined work interval DB (3), a statistical estimation of the parameter value of the model formula in which the difference between the maintenance work interval calculated from the model equation and the already determined maintenance work interval becomes the smallest Processing circuit.

The processing circuit may be dedicated hardware or a CPU (Central Processing Unit) that executes a program stored in a memory.

When the processing circuit is dedicated hardware as shown in Fig. 4A, the processing circuit 104 may be a single circuit, a compound circuit, a programmed processor, a parallelized programmed processor, an ASIC (Application Specific Integrated Circuit) (Field-Programmable Gate Array), or a combination thereof.

The respective functions of the model expression building unit 4a and the parameter estimating unit 4b may be realized by respective processing circuits and these functions may be combined into one processing circuit.

When the processing circuit is the processor 106 as shown in FIG. 4B, the respective functions of the model expression building unit 4a and the parameter estimating unit 4b are implemented by software, firmware, or a combination of software and firmware .

The software or firmware is described as a program and is stored in the memory 105. [ The processor 106 reads out and executes the program stored in the memory 105 to realize the functions of the respective parts.

That is, the failure risk index estimating apparatus 1 includes a memory 105 for storing a program to be executed as a result of steps ST1 and ST2 shown in FIG. 5 when executed by the processor 106. FIG. These programs cause the computer to execute the procedure or method of the model expression construction unit 4a and the parameter estimation unit 4b.

Volatile or volatile semiconductor memories such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Electrically-EPROM) A magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, a DVD, and the like.

The functions of the model expression constructing unit 4a and the parameter estimating unit 4b may be partially realized by dedicated hardware, and some of them may be realized by software or firmware. For example, the function of the model expression constructing unit 4a is realized by a processing circuit as a dedicated hardware, and the processor 106 reads the program stored in the memory 105 and executes the program for the parameter estimating unit 4b The function may be realized.

As such, the processing circuit can realize each of the functions by hardware, software, firmware, or a combination thereof.

Next, the operation will be described.

5 is a flowchart showing the operation of the failure risk index estimating apparatus 1, and shows a series of processes from the estimation of the failure risk index to the storage of the failure risk index in the first storage unit 5. [ FIG. 6 is a diagram showing the relationship between the failure risk indicator R (t) and the intermediate evaluation indicators S, W, and M, and shows the transition of the failure risk indicator R (t) in the elapsed time after the maintenance work. 7 is a diagram showing the relationship between the evaluation items C, E, and D of the FMEA result and the intermediate evaluation indexes S, W, and M. FIG.

Hereinafter, the operation of the failure risk index estimation device 1 will be described with reference to FIG. 5, with reference to FIG. 6 and FIG.

First, the index estimating unit 4 constructs a model equation of the failure risk index based on the FMEA result read from the FMEA result DB 2 and the statistical evaluation information A (step ST1).

The failure risk indicator R (t) changes (changes with time) according to the statistical distribution f (t) as shown in the following equation (1). In the following equation (1), the time scale factor S is a coefficient for adjusting the change in R (t) in the time direction and is a parameter relating to the rate at which the failure risk increases. The risk weighting coefficient W is a coefficient for adjusting the magnitude of R (t) and is a parameter relating to the degree of weighting of the failure risk.

In the statistical estimation of R (t) by the index estimating unit 4, as shown in Fig. 6, an operation interval exceeding the permissible upper limit threshold value "1.0" as R (t) T _E , the work interval for performing the maintenance work in advance for safety is obtained. For example, the index estimating unit 4 sets the safety margin M, and calculates the working interval T assumed from the model equation from the following equation (2). The safety margin M is a time interval dating back from the elapsed time at the proper interval T _E.

In the statistical estimation of R (t) by the index estimating unit 4, the time scale coefficient S, the risk weighting coefficient W, and the safety margin M are parameters whose values fluctuate until an estimate of the failure risk index is obtained . Here, these are referred to as intermediate evaluation indexes.

If the statistical distribution f (t) follows the Weibull distribution, the statistical distribution f (t) can be expressed as a cumulative density function of the Weibull distribution as shown in equation (3) below. Here, the shape parameter γ and the scale parameter φ are statistical distribution parameters in the cumulative density function of the Weibull distribution.

The above equation (1) can be expressed by the following equation (4) using the shape parameter? And the scale parameter?. In the following equation (4), the scale parameter? Appears only as a product of the time scale coefficient S, and therefore,? = 1, and only S is a parameter to be estimated.

Since the failure risk indicator R (T _E ) = 1.0 when the elapsed time t is the appropriate interval T _E , the appropriate interval T _E can be expressed by the following equation (5) using the above equation (4). In the following formula (5), ln represents a natural logarithm.

As shown in FIG. 7, the evaluation items of the FMEA result include the frequency level of the fault, the magnitude level of influence, and the detectability level, and thereafter, the frequency level of the fault is denoted by C, the magnitude level of influence is denoted by E Let D be the probability level. There is a relationship shown in Fig. 7 between the evaluation item of the FMEA result and the intermediate evaluation index.

The frequency level of failure C is related to the time scale factor S with respect to the rate of increase in failure risk since it can be an index of the frequency at which a failure occurs in the part to be evaluated.

Since the magnitude level of influence E can be an index of the magnitude of the influence of the failure on the part to be evaluated, the risk weighting factor W relating to the degree of weighting of the failure risk and the safety margin M Are related to both.

Since the detectability level D can be an index of easiness of detection of a failure occurring in the part to be evaluated, it is related to the safety margin M about the degree to which the maintenance work is advanced.

The index estimating unit 4 statistically estimates the values of the parameters S, W, M, γ, and φ of the model equation so that the difference between the maintenance work interval T assumed from the model equation and the maintenance work interval T _S already determined is the smallest (Step ST2). Since the scale parameter? Is fixed at "1 ", the index estimating unit 4 calculates the values of the parameters S, W, M, and? Of the model formula for each part and for each check item, And stores it in the first storage unit 5.

Next, the detailed operation of the index estimating unit 4 will be described.

8 is a flowchart showing the operation of the model expression building unit 4a and the parameter estimating unit 4b. Hereinafter, it is assumed that the statistical distribution is a Weibull distribution.

The model expression construction unit 4a merges the table data of the FMEA result DB 2 and the already determined work interval DB 3 (step ST1a). The table data is data composed of information for each item as shown in Figs. 2A and 2B. Among the table data, the part ID, the check item ID, and the information on the side adjacent thereto are record data.

For example, in the information of the already determined work interval DB 3 shown in FIG. 2B, the part ID, the check item ID, and the already determined work interval number following this are the information constituting the table data.

The model expression construction unit 4a reads out the table data from the already determined work interval DB 3 and outputs the FMEA result having the same part ID and the check item ID based on the part ID and the check item ID in the table data The table data of the DB 2 is retrieved.

The model expression construction unit 4a executes so-called merge, which combines the FMEA result of the table data specified by this search with the table data read out from the already determined work interval DB (3).

The model expression construction unit 4a executes the above processing on all the table data in the already determined work interval DB 3 and generates information in which the determined repair work interval and the FMEA result are combined. This combination information is shown in Fig. In FIG. 9, the maintenance work interval T _S already determined is expressed by the number of months of the work interval.

Table data that does not have the same part ID and check item ID as the table data of the already determined work interval DB (3) among the table data of the FMEA result DB 2 is also directly added to the combination information shown in FIG.

The model expression construction unit 4a allocates the parameters of the intermediate evaluation indexes related to the FMEA result in accordance with the evaluation items and the evaluation levels of the FMEA results for each of the part IDs and the inspection item IDs (step ST2a).

The model expression establishing unit 4a assigns a common intermediate evaluation index parameter to the FMEA result of the combination information with respect to the same evaluation level in the evaluation item and the evaluation item related to the failure of the FMEA, Build.

The common parameters are set to the same value in the model equation between the data in which the part ID and the check item ID are different. Thus, it is possible to evaluate the influence of the failure common to different parts or maintenance items.

10 is a diagram showing a result of assigning the parameter of the intermediate evaluation index to the evaluation item of the FMEA result. For example, in FIG. 9, the fault frequency level C, which is an evaluation item of the FMEA result corresponding to (part ID, inspection item ID) = (EQ001, MT001), has an evaluation level of "2". (Part ID, check item ID) = (EQ002, MT002) exists in one of the FMEA results of the combination information shown in Fig. 9 in which the evaluation level of the frequency level C of the failure is "2". Expression model construction unit (4a), as shown in Figure 10, with respect to the parameters of the time scale factor S in relation to the frequency of level C of the failure, a common allocation of parameters S _2.

The magnitude level E of the influence, which is the evaluation item of the FMEA result corresponding to (EQ001, MT001), is the evaluation level "3 ".

(Part ID, check item ID) = (EQ001, MT002) is one of the evaluation levels of the influence magnitude level E of "3" among the FMEA results of the above combination information. The model equation construction unit (4a) is assigned to FIG. 10 shows, a common parameter with respect to the risk of the weighting coefficient W parameters that are relevant to the magnitude of the effect level E W ₃ as a.

(EQ001, MT001) and data of (EQ001, MT002), the detection level D and the evaluation level, which are evaluation items of the FMEA result, are "3" together. Therefore, as shown in Fig. 10, the model expression constructing unit 4a assigns the common parameters M3 _{, 3} to the parameters of the safety margin M that are related to the detectability level D. [

The model expression establishing unit 4a determines whether or not the parameter of the intermediate evaluation index has been allocated to the data of (part ID, check item ID) = (EQ001, MT001) Based on this, the work interval T, the failure risk indicator R (t), and the appropriate interval T _E are constructed as a model equation of the failure risk indicator.

Parameters of the intermediate evaluation index for the data in the (EQ001, MT001) is, because _{S 2, W 3, M 3} , 3, are as in the model to expression.

The safety margin M, may be expressed as a coefficient multiplied with the coefficient factor M _E M _D factor by detecting potential level D M × M _{E D} E of the effect due to the size level. As a result, the safety margin M becomes a fixed value, so that the number of parameters to be estimated can be reduced.

Next, the parameter estimation by the government (4b), the difference between the model equation construction unit (4a) on the basis of a parameter assigned by calculating the working distance T from the model formula, and the operation interval T to a predetermined working distance T _S The parameter that becomes the minimum is statistically estimated (step ST3a).

Since the statistical distribution f (t) depends on the Weibull distribution, the working interval T is obtained in the form of a variable determined by the parameter of the intermediate evaluation index, as shown in Fig. In Fig. 11, Z _i is a parameter represented by the following equation (6).

Parameter estimation unit (4b) is, for all parts ID and check item ID in the combination information, the parameters of the intermediate evaluation index is the sum of the squares of the difference between the working distance T and the previously determined task intervals T _S so as to minimize S, W , M, and the shape parameter gamma. As described above, the scale parameter? Is "1".

As a method of estimating these parameters, for example, a conjugate direction method can be used. However, the method is not limited to the conjugate direction method as long as it is a method capable of estimating a parameter that minimizes the error with the already determined work interval T _S.

The parameter estimating unit 4b classifies the results of estimation as described above for each part ID and for each inspection item ID and stores them in the first storage unit 5 in correspondence with the part ID and the inspection item ID (step ST4a ). Since the statistical distribution f (t) depends on the Weibull distribution, a "Weibull distribution" is set in the item of "Statistical distribution" shown in Fig. 3, Respectively.

The parameter estimating unit 4b may use the information used in the statistical estimation of the parameter values for one side between the similar inspection items for the statistical estimation of the parameter values for the other side between the inspection items. For example, since the energization inspection and the insulation inspection all check the conduction state of electricity, they can be said to be similar inspection items. Thus, the parameter estimating unit 4b uses the information used in the statistical estimation of the parameter values for the energization inspection for the statistical estimation of the parameter values for the insulation inspection. In this way, the information used for the statistical estimation can be reused, and the processing load required for estimation can be reduced.

As described above, in the failure risk index estimating apparatus 1 according to the first embodiment, when the model expression establishing unit 4a reads the FMEA result read from the FMEA result DB 2 and the statistical evaluation information A Based on this, we construct a model equation that shows the trend of failure risk indicators according to statistical distribution. The parameter estimating unit 4b calculates the difference between the maintenance work interval calculated from the model equation and the already determined maintenance work interval based on the information read from the FMEA result DB 2 and the already determined work interval DB 3 And statistically estimates parameter values of the model expression. With this configuration, even when there is no or little repairing data of the facility, it is possible to appropriately estimate the index of the risk that the failure occurs in the facility.

In the failure risk index estimating apparatus 1 according to the first embodiment, when the model expression establishing unit 4a determines that the evaluation item related to the failure of the FMEA is equal to the evaluation level in the evaluation item, And a common parameter is assigned to the parameter of the related model formula. By doing so, it is possible to evaluate the influence of the failure common to the different parts or the maintenance items.

In the failure risk index estimating apparatus 1 according to the first embodiment, the parameter estimating unit 4b may store information used in the statistical estimation of the parameter values for one side between the similar check items, For the statistical estimation of the parameter values for the other side.

In this way, the information used for the statistical estimation can be reused, and the processing load required for estimation can be reduced.

Embodiment 2:

12 is a block diagram showing the functional configuration of the failure risk indicator estimating device 1A according to the second embodiment of the present invention. In Fig. 12, the same components as those in Fig. 1 are denoted by the same reference numerals, and a description thereof will be omitted. The failure risk indicator estimating apparatus 1A includes a facility information DB 6, a maintenance performance DB 7, a failure history DB 8, a range reduction unit 9, A data storage unit 10, an index estimating unit 11, a second storage unit 12, and a merging unit 13.

The facility information DB 6 is a DB that stores facility information including equipment, components constituting the equipment, and maintenance items for each part. In the equipment information DB 6, for example, information of items such as those shown in Fig. 13A is stored. The item of "equipment ID" The "part ID" and "check item ID" are the same as those shown in FIG.

The date and time when the repair contract is started for each maintenance item of the part is set in the item of "repair start date & time ". In addition, during the maintenance contract, the maintenance date and time of the maintenance work performed individually is the maintenance work execution date and time described later with reference to FIG. 13B.

The maintenance performance DB 7 is a DB that stores the performance data of the maintenance work for each maintenance item of the components constituting the equipment. In the maintenance performance DB 7, information of the item shown in Fig. 13B is stored. In the item "maintenance result ID ", identification information of the performance data of the maintenance work is set.

The "facility ID", "part ID" and "check item ID" are the same as those shown in FIGS. 13A and 2A. The date and time of the maintenance work for each part is set in the item "maintenance work execution date & time ".

The failure history DB 8 is a DB for storing failure data for each component constituting the equipment. In the failure history DB 8, information of items as shown in Fig. 13C is stored. In the item of " fault history ID ", identification information of fault history data is set. The "facility ID" and "part ID" are the same as those shown in FIG. 13B. In the item of "fault occurrence date ", the date and time when a failure occurs in the component is set. The item of "related check item ID" is set to the identification information of the check item related to the failure that has occurred.

The range narrowing section 9 classifies the information stored in the equipment information DB 6, the maintenance performance DB 7 and the failure performance DB 8 into information corresponding to each FMEA result in the FMEA result DB 2 .

For example, the range narrowing section 9 compares the intervals of non-failure after the maintenance work is performed for each maintenance achievement data, classifies the information indicating the non-failure intervals to be aggregated for each FMEA result, (10).

The range reduction data storage unit 10 is a DB that stores information classified by the range reduction unit 9. [ The range reduced data storage section 10 stores information of the items shown in FIG. 14A.

In Fig. 14A, the "fault frequency level", the "influence magnitude level" and the "detectability level" are evaluation items of the FMEA and are the same as those shown in Fig. 2A.

The "part ID" and "check item ID" are the same as those shown in FIG. The "repair work execution date" is the same as that shown in FIG. 13B.

The items of "number of months without breakdown" include the number of months after the maintenance work has reached the next maintenance work, the number of months until the failure occurs after the maintenance work, and the number of months Is set. In the item of "fault occurrence flag ", a value indicating whether a failure has occurred in the part is set. For example, "1" is set when a component fails, and "0" is set when no failure occurs.

The index estimating section 11 estimates the second estimated value of the failure risk index according to the relationship between the performance data of the maintenance work and the performance data of the failure based on the information classified by the range reducing section 9. [

For example, the index estimating unit 11 reads information corresponding to the FMEA result to be processed from the reduced-range data storage unit 10, statistically calculates the failure risk index for each part and each maintenance item based on the read information . The estimated value of the failure risk index obtained by this estimation is stored in the second storage section 12 together with the actual data number used for the estimation.

The second storage unit 12 stores the estimated value of the failure risk index estimated by the index estimation unit 11 for each part and for each maintenance item. In the second storage unit 12, information of an item as shown in Fig. 14B is stored. 14B, the "part ID" and the "check item ID" are the same as those shown in FIG. 2A. The "statistical distribution" and "statistical distribution parameter" are the same as those shown in FIG. The "time scale factor "," risk weighting factor ", and "safety margin" are parameters that become the second estimate value of the failure risk indicator and are the same as those shown in FIG. In the item of "actual data number", the number of data used for the statistical estimation of the failure risk index by the index estimating unit 11 is set.

The merging unit 13 calculates the estimated value of the final failure risk index by subtracting the first estimated value estimated by the parameter estimating unit 4b and the second estimated value estimated by the index estimating unit 11. [

For example, the merging unit 13 calculates the first estimated value and the second estimated value according to the number of data assumed in the estimation of the first estimated value and the actual data used in the estimation of the second estimated value, . The information B indicating the estimated value of the failure risk index is output from the merging unit 13. [

The information B indicating the estimated value of the failure risk index is composed of the information of the items shown in Fig. 14C. 14C, "part ID" and "check item ID" are the same as those shown in FIG. 2A. The "statistical distribution" and "statistical distribution parameter" are the same as those shown in FIG. The "time scale factor "," risk weighting factor ", and "safety margin" are parameters that become the second estimate value of the failure risk indicator and are the same as those shown in FIG.

The FMEA result DB 2, the already determined work interval DB 3, the equipment information DB 6, the maintenance performance DB 7 and the failure performance DB 8 in the failure risk indicator estimation device 1A are stored 4A and 4B. The information stored in each of the FMEA result DB 2 and the already determined work interval DB 3 is input to the index estimator 4 via the DB input /

Information stored in each of the facility information DB 6, the maintenance performance DB 7 and the failure history DB 8 is inputted to the range narrowing unit 9 through the DB input /

The statistical evaluation information A is input to the failure risk index estimation device 1A via the information input interface 102. [ The information B indicating the estimated value of the final failure risk index is outputted from the merging unit 13 through the information output interface 103. [

It is conceivable that the first storage unit 5, the reduced size data storage unit 10 and the second storage unit 12 are provided in the storage device in which the database 100 is provided. However, the processing circuit 104 Or the like. The first storage unit 5, the reduced size data storage unit 10, and the second storage unit 12 may be provided in the memory 105 shown in FIG. 4B.

Each function of the index estimating unit 4, the range reducing unit 9, the index estimating unit 11 and the merging unit 13 in the failure risk index estimating apparatus 1A is realized by a processing circuit.

That is, the failure risk index estimating apparatus 1A has a processing circuit for performing the processing in the above-described functions of the respective units. The processing circuit may be dedicated hardware or a CPU that executes a program stored in a memory.

When the processing circuit is dedicated hardware as shown in Fig. 4A, the processing circuit 104 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, .

The respective functions of the index estimating unit 4, the range reducing unit 9, the index estimating unit 11 and the merging unit 13 may be realized by respective processing circuits, and these functions may be combined into one processing circuit May be realized.

When the processing circuit is the processor 106 as shown in FIG. 4B, the functions of the index estimating unit 4, the range reducing unit 9, the index estimating unit 11, and the merging unit 13 are implemented by software, Firmware, or a combination of software and firmware.

The software or firmware is described as a program and is stored in the memory 105. [ The processor 106 reads out and executes the program stored in the memory 105 to realize the functions of the respective parts.

That is, the failure risk index estimation device 1A includes a memory 105 for storing a program to be executed as a result of the processing from step ST1b to step ST8b shown in FIG. 15 when executed by the processor 106 .

These programs cause the computer to execute a procedure or a method of the index estimating unit 4, the range reducing unit 9, the index estimating unit 11, and the merging unit 13.

Some of the functions of the index estimating unit 4, the range reducing unit 9, the index estimating unit 11 and the merging unit 13 may be realized by dedicated hardware and some of them may be realized by software or firmware. For example, the function of the index estimating unit 4 and the range reducing unit 9 is realized by a processing circuit as a dedicated hardware, and the processor 106 performs the functions of the index estimating unit 11 and the merging unit 13, The function may be realized by reading and executing the program stored in the storage unit 105. [ As such, the processing circuit can realize each of the functions by hardware, software, firmware, or a combination thereof.

Next, the operation will be described.

FIG. 15 is a flowchart showing the operation of the failure risk index estimating apparatus 1A. FIG. 15 shows a series of processes until a first estimation value and a second estimation value of the failure risk index are obtained and a final estimation value is output.

The range narrowing section 9 classifies the information stored in the equipment information DB 6, the maintenance performance DB 7 and the failure performance DB 8 into information corresponding to each FMEA result (step ST1b).

The information classified by the range narrowing section 9 is stored in the range reduced data storage section 10.

The index estimating section 11 reads information corresponding to the FMEA result to be processed from the reduced range data storage section 10 and statistically estimates the failure risk index for each part and each inspection item based on the read information (Step ST2b).

The index estimating unit 11 stores the second estimated value and the actual data count obtained in the estimation in the second storage unit 12 in association with the part ID and the check item ID (step ST3b).

FMEA result If all of the FMEA results stored in the DB 2 have not been processed (step ST4b: NO), the processing from step ST1b is repeated.

FMEA result When all the FMEA results stored in the DB 2 have been processed (step ST4b; YES), the process proceeds to step ST5b.

In step ST5b, the index estimating unit 4 estimates the parameter value which is the first estimation value, as in the first embodiment.

The index estimating unit 4 stores the estimated first estimated value in the first storage unit 5 in association with the part ID and the check item ID (step ST6b).

The merging unit 13 reads out the first estimate stored in the first storage unit 5 and the second estimate stored in the second storage unit 12 and reads the first estimate and the second estimate read out And calculates an estimate of the final failure risk index (step ST7b). Thereafter, the merger 13 outputs the information B indicating the estimated value of the failure risk index (step ST8b).

15 shows a case where the estimation of the second estimation value by the indicator estimation unit 11 is performed before the estimation of the first estimation value by the indicator estimation unit 4, this is not restrictive.

For example, the estimation of the first estimate by the index estimator 4 may be performed before the estimation of the second estimate by the index estimator 11. [ The estimation of the first estimate value by the index estimating unit 4 and the estimation of the second estimate by the index estimating unit 11 may be performed in parallel.

Next, the operation of the range reduction unit 9 will be described in detail.

16 is a flowchart showing the operation of the range reducing section 9 and classifies the information stored in the equipment information DB 6, the maintenance performance DB 7 and the failure performance DB 8 and stores the information in the range reduction data storage section 10 In Fig. 2). Fig.

First, the range reducing unit 9 merges the table data of the FMEA result DB 2, the facility information DB 6, and the maintenance performance DB 7 (step ST1c).

For example, the range narrowing section 9 is configured to classify, based on the facility ID, the part ID, and the check item ID of the table data read from the facility information DB 6, (7) is retrieved. The range narrowing unit 9 merges the information set in the item of the "maintenance work execution date" of the table data specified by this search into the table data read from the facility information DB 6.

If there is no maintenance achievement data of the same facility ID, part ID, and maintenance item ID as the facility information, the range reduction unit 9 does not delete the record data of the facility ID, the part ID, And leaves it in the table data after merging.

Subsequently, on the basis of the facility ID, the part ID, and the check item ID in the merged table data, the range reduction unit 9 sets the range of the FMEA result DB (2) having the same facility ID, part ID, Retrieve table data. The range reduction unit 9 merges the FMEA result in the table data specified by the search into the table data after the merging.

The range narrowing section 9 generates the information in which the facility information and the maintenance performance data are classified for each FMEA result by executing the above-described processing on all the table data in the facility information DB 6.

Next, the range reduction unit 9 reduces the range only by the record data corresponding to the FMEA result of the evaluation target among the table data merged as described above (step ST2c).

The range narrowing section 9 adds items of "number of non-failure continuing months" and "failure occurrence flag" to the record data of the merged table data.

In step ST3c, based on the equipment ID and the part ID corresponding to the FMEA result to be evaluated, the range reducing unit 9 executes the maintenance operation for the part corresponding to the part ID, From the failure history DB (8). If the maintenance work has not been executed, the failure history data that occurred the earliest between the maintenance start date and time and the next maintenance work is searched.

The range reduction unit 9 calculates the number of non-failure continuing months in which no failure has occurred in the part based on the failure result data of the search result, sets the calculated number of months in the item of "number of non-failure continuing months" 1 ", which is a value indicating occurrence of the failure, is set to the item "fault occurrence flag ".

On the other hand, if there is no fault in the search, the range narrowing unit 9 calculates the time interval until the next maintenance work date and time, sets the number of months in the item of "number of non-failure continuing months" Quot; 0 "which is a value indicating " failure occurrence flag"

If the date and time of the maintenance work is not determined next, the interval to the present time is set to the item of "number of months without breakdown".

The range narrowing unit 9 executes the above processing on all the record data in the table data merged in step ST1c. As a result, information in which facility information, maintenance achievement data, and failure performance data are classified for each FMEA result is generated.

The range reduction unit 9 stores the table data of the processing result in the range reduction data storage unit 10 (step ST4c). As shown in FIG. 14A, the table data is classified into three categories: "frequency of failure level", "magnitude level of influence", "detectability level", "equipment ID", "part ID" Work execution date ", "no breakdown number of months ", and" fault occurrence flag ".

Next, the operation of the index estimating unit 11 will be described in detail.

FIG. 17 is a flowchart showing the operation of the index estimating unit 11, and shows a series of processes from estimation of the second estimation value to storage in the second storage unit 12. FIG.

In step ST1d, the index estimating unit 11 retrieves record data having the same part ID and check item ID as the processing target, from the table data stored in the range reduced data storing unit 10. [ The index estimating unit 11 calculates the number of record data whose value set in the item of "number of non-failure continuing months" in the searched record data is equal to or less than the elapsed time t (t = 1, 2, ...) do. The number of record data thus calculated is the total number of parts in accordance with the number of months after the maintenance work.

The index estimating unit 11 retrieves record data in which "1" is set in the item of "fault occurrence flag" from the table data stored in the range reduced data storage unit 10. [

The index estimating section 11 calculates the number of failures according to the number of elapsed days after the maintenance work for each part ID and for each maintenance item ID based on the retrieved record data (step ST2d).

Next, the index estimating unit 11 calculates an actual failure rate, which is a value obtained by dividing the number of failures according to the number of elapsed months after the maintenance work by the total number of the alerts (step ST3d).

Subsequently, the index estimating unit 11 specifies a statistical distribution approximating the trend of the actual failure rate calculated in step ST3d, and statistically estimates the parameter value of the model equation according to this statistical distribution (step ST4d). For example, the transition of the actual failure rate is estimated by the time scale factor S and the shape parameter? Which are approximated by the failure risk indicator R (t) when the risk weighting factor W is set to 1 in Equation (4).

Although it is conceivable to use the conjugate direction method for the estimation of the parameter, it may be an existing method for estimating the parameter of the Weibull distribution.

The value of the risk weighting factor W is set for each FMEA result. For example, a value received from the user may be set, or a value of a coefficient that minimizes the difference between the loss amount at the time of occurrence of a failure and the failure rate obtained from the performance data may be set. The value of the safety margin M may be set to a value received from the user or M = 0.

In addition, although the Weibull distribution is used as the statistical distribution, it may be a statistical distribution such as a gamma distribution or log normal distribution, and it may be a statistical distribution in which the error with the actual data is minimized.

The index estimating unit 11 checks whether there is any unprocessed combination among the combination of the part ID and the check item ID of the table data stored in the range reduced data storing unit 10 (step ST5d).

If there is a combination of unprocessed combinations of the part IDs and the check item IDs of the table data stored in the range reduced data storage unit 10 (step ST5d: NO), the process returns to step ST1d and the above process is repeated .

When all the combinations of the part ID and the check item ID are processed (step ST5d; YES), the index estimating unit 11 associates the second estimated value, which is the estimated parameter value, with the component ID and the check item ID, (Step ST6d).

Since the statistical distribution follows the Weibull distribution, a "Weibull distribution" is set in the item of "Statistical distribution " shown in Fig. 14B, and a value of the shape parameter y is set in the item of" Statistical distribution parameter ".

Next, the operation of the merging unit 13 will be described in detail.

FIG. 18 is a flowchart showing the operation of the merging unit 13, and shows a series of processes from the estimation of the final failure risk index to the calculation of the final failure risk index from the first estimation value and the second estimation value. The merging unit 13 reads out the part ID and the inspection item ID to be processed, the second estimation value corresponding to them, and the actual data number from the second storage unit 12 (step ST1e).

Next, the merging unit 13 reads the part ID, the inspection item ID to be processed and the first estimated value corresponding to them, from the first storage unit 5 (step ST2e).

A part ID that is the same as the part ID corresponding to the second estimated value read from the second storage unit 12 and the part ID that is the same as the check item ID and the record that does not have the check item ID among the first estimated values stored in the first storage unit 5 The data is read from the first storage unit 5 as it is.

The number of actual data set in the record data read from the second storage unit 12 is N _A , the total number of parts for each elapsed time after maintenance is always N _A , and the number of failures is N _A × Assume R _A (t). The merging unit 13 obtains N _I which is the number of data assumed in the statistical estimation of the first estimate value in the record data having the same part ID and the check item ID read out from the first storage unit 5 ST3e).

For example, the merging unit 13 calculates N _I from the following equation (7) by using the tolerance? And the confidence rate? In the statistical evaluation information A. In the following equation (7), z (?) Represents 100 alpha% above the standard normal distribution. If? = 0.1 and? = 0.99 (99%), then z (?) = 2.326 and N _I = 2168.

In the record data having the same part ID and the check item ID read from the first storage unit 5, the total number of parts for each elapsed time after the maintenance work is always N _I , And the number of failures is N _I x R _I (t).

The merging unit 13 has the failure risk index R _A (t) indicated by the record data read from the second storage unit 12 and the same part ID and the check item ID read from the first storage unit 5 The failure risk index R _I (t) indicated by the record data is calculated (step ST4e).

The merging section 13 sets the total number of parts for each elapsed time after the maintenance work to N _A + N _I and sets the number of failures to N _A R _A (t) + N _I R _I (t) The risk indicator R _F (t) is again estimated statistically. As the estimation method, for example, a conjugate direction method is used. The processing up to this point is an asymmetric processing. In this way, the merging unit 13 uses N _I, which is the number of data assumed in the statistical estimation of the first estimate, as the estimate of the second estimate. Thereby, the first estimate value and the second estimate value can be appropriately divided.

The merging unit 13 checks whether or not there is a combination of the unspecified part ID and the check item ID among the information stored in the first storage unit 5 and the second storage unit 12 (step ST5e ).

If there is a combination of the unspecified part ID and the check item ID among the information stored in the first storage unit 5 and the second storage unit 12 (step ST5e: NO) The processing from ST1e is repeated.

(Step ST5e; YES), the merging unit 13 obtains, from the parameter values of the failure risk index R _F (t), information indicating the estimated value of the final failure risk index B (step ST6e).

As described above, in the failure risk index estimating apparatus 1A according to the second embodiment, the range reducing unit 9 includes the facility information DB 6, the maintenance performance DB 7, and the failure history DB 8 ) Is classified into information corresponding to each FMEA result. The index estimating section 11 estimates an estimated value of the failure risk index according to the relationship between the performance data of the maintenance work and the performance data of the failure based on the information classified by the range reducing section 9. [ The merging unit 13 calculates the estimated value of the final failure risk index by subtracting the first estimated value estimated by the index estimating unit 4 and the second estimated value estimated by the index estimating unit 11. [ With this configuration, even when the maintenance performance data of the facility is small, it is possible to appropriately estimate the index of the risk that the facility is faulty.

In the failure risk index estimating apparatus 1A according to the second embodiment, the merging unit 13 uses the number of data assumed in the statistical estimation of the first estimate value for the difference between the second estimate value and the second estimate value. With this configuration, the first estimate value and the second estimate value can be properly calculated.

In addition, the present invention can be freely combined with each embodiment, or a modification of any constituent element of each embodiment, or omitting any constituent element in each embodiment within the scope of the invention.

(Industrial applicability)

The failure risk index estimating apparatus according to the present invention can appropriately estimate an index of a risk that a failure occurs in a facility even if there is no or little maintenance data of the facility and thus can be applied to various mechanical systems, for example.

1, 1A: Failure Risk Indicator Estimator
2: FMEA Result DB
3: The already determined work interval DB
4, 11:
4a: Model expression building section
4b: parameter estimating unit
5: First storage unit
6: Equipment Information DB
7: Performance record DB
8: Failure performance DB
9:
10: Range reduction data storage unit
12: Second storage unit
13:
100: Database
101: DB Input / Output Interface
102: Information input interface
103: Information output interface
104: Processing circuit
105: Memory
106: Processor

Claims (7)

A model equation which represents a transition of the failure risk index according to the statistical distribution, based on information indicating a failure mode influence analysis result for each maintenance item of components constituting the equipment and information indicating a statistical distribution used for estimation of the failure risk index, A model expression building unit for constructing a model expression,
Based on the information indicating the failure mode influence analysis result for each maintenance item of the parts constituting the equipment and the information indicating the maintenance work interval determined for each maintenance item of the part, A parameter estimating unit for statistically estimating a parameter value of the model formula in which a difference between the work intervals is minimized,
The failure risk index estimating apparatus comprising:
The method according to claim 1,
Equipment data including equipment and parts constituting the equipment and inspection items for each part, performance data of repair work for each inspection item of the parts constituting the equipment, and performance data of the failures of each part constituting the equipment, A range narrowing section for classifying the information into corresponding information for each effect analysis result,
An index estimator for estimating an estimate of the failure risk index according to the relationship between the performance data of the maintenance work and the performance data of the failure based on the information classified by the range reduction unit,
A first estimating unit that estimates a failure value of the failure risk indicator based on a first estimated value that is a parameter value estimated by the parameter estimating unit and a second estimated value that is an estimated value estimated by the index estimating unit,
The failure risk index estimating apparatus comprising:
3. The method of claim 2,
Wherein the merging unit uses the number of data assumed in the statistical estimation of the first estimation value for the difference between the second estimation value and the second estimation value.
The method according to claim 1,
Wherein the model expression construction unit assigns a common parameter to the parameter of the model expression related to the evaluation item when the evaluation item on the failure of the failure mode influence analysis and the evaluation level on the evaluation item are equal Fault risk indicator estimator.
The method according to claim 1,
Wherein the parameter estimating unit uses information used in statistical estimation of parameter values for one side between similar check items to statistical estimation of parameter values for the other side between the check items.
Based on the information indicating the result of the failure mode influence analysis for each check item of the components constituting the equipment, and the information indicating the statistical distribution used for the estimation of the failure risk index, the model expression building unit calculates the failure risk index A step of constructing a model expression indicating a trend,
The parameter estimating section calculates the maintenance work interval calculated from the model formula based on the information indicating the failure mode influence analysis result for each maintenance item of the components constituting the facility and the information indicating the maintenance work interval already determined for each maintenance item of the component, A step of statistically estimating a parameter value of the model formula in which the difference of the already determined maintenance work interval is minimized
And estimating the failure risk index.
The method according to claim 6,
The scope information of the parts constituting the equipment, the equipment and the equipment and the inspection items for each part, the performance data of the maintenance work for each inspection item of the parts constituting the equipment, and the performance data Into information corresponding to each of the failure mode influence analysis results,
Estimating an estimated value of the failure risk index according to the relationship between the performance data of the maintenance work and the performance data of the failure based on the information classified by the range reduction unit;
The merging unit calculates the estimated value of the failure risk index finally by subtracting the first estimated value, which is the parameter value estimated by the parameter estimating unit, and the second estimated value, which is the estimated value estimated by the index estimating unit,
And estimating the failure risk index.

Images