JP2016224862A - Maintenance work interval determination device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a maintenance work interval determination device capable of determining a maintenance work interval such that maintenance quality is suppressed from decreasing when an actual fault occurrence period fluctuates.SOLUTION: A maintenance work interval determination device comprises: a fault distribution parameter setting part 11 which sets a fault distribution that a fault occurrence rate at facilities as an object of maintenance work is assumed to accord with; an actual result data storage part 3 which previously store actual result data on a fault occurrence actual result and a maintenance work actual result of the facilities; a fault occurrence actual result calculation part 4 which calculate the number of actual result faults upon the basis of the actual result data; an assumed fault number calculation part 12 which calculates the number of assumed faults assumed to occur at work intervals of the actual result data when it is assumed that the fault occurrence rate accords with the fault distribution; a fault number matching condition determination part 13 which determines whether a difference between the assumed fault number and actual result fault number meets a fault number matching condition, as for the fault distribution; and a work interval determination part 17 which determines a maintenance work interval at the facilities upon the basis of the fault distribution determined to meet the fault number matching condition.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a maintenance work interval determination device.

  Conventionally, as a device for evaluating a maintenance plan for preventive maintenance of machine equipment, a strategy input unit for inputting a strategy for determining the maintenance time of the machine equipment as a conditional expression and outputting strategy information is expected for the machine equipment. Is assigned to numerical values of 1 and 0 in the time interval expressed in a discretized manner to form a binary number sequence, which is used as a failure scenario, a failure scenario generation unit that generates the failure scenario, and the strategy information and the failure scenario information. Based on the maintenance scenario generation unit for generating a maintenance scenario consisting of a binary sequence for determining the execution timing of maintenance measures in each failure scenario, the probability of failure in the future of the machine equipment and the machine equipment after maintenance A probability database that stores the probability of failure of the machine, costs associated with normal operation of the machine equipment, costs due to loss at the time of failure, and maintenance of the machine equipment. Cost database storing costs required for maintenance and costs associated with normal operation after maintenance, the failure scenario, the maintenance scenario, the failure probability of the machine facility in the future, and the subsequent failure probability when the machine facility is maintained Applying a strategy for determining the maintenance timing of the mechanical equipment based on the expenses associated with the normal operation of the mechanical equipment, the expenses due to the loss at the time of the failure, and the maintenance expenses and the expenses associated with the normal business after the maintenance A calculation unit that calculates a cost related to the entire operation of the machine equipment in the case of, and a display unit that graphically displays a calculation result of the cost by the calculation unit, the maintenance time determination strategy of the machine facility is displayed in the display What is comprised so that evaluation can be performed based on the graph display by a part is known (for example, refer patent document 1).

JP-A-2005-202585

  As described above, in the prior art disclosed in Patent Document 1, the failure probability in the future of the mechanical equipment and the failure probability in the subsequent maintenance of the mechanical equipment are determined as the future failure probability of the mechanical equipment and the maintenance after the maintenance. For the value of the probability of failure of the mechanical equipment, one stored in advance in the probability database is used. The parameter of the probability distribution function used for calculating the probability of failure in the future stored in the probability database uses a value statistically calculated from the failure history of the actual machine equipment.

  However, the parameters calculated statistically in this way do not always give a true failure probability distribution function. For this reason, fluctuations may occur in the actual failure occurrence probability and failure occurrence time, and the maintenance quality may be deteriorated due to a shift in the failure occurrence time.

  The present invention has been made in order to solve such a problem, and includes a constant value that includes a true value for a parameter that determines a statistical distribution that is assumed to follow a failure rate in a facility that is a subject of maintenance work. A maintenance work interval determination device is provided that can determine a maintenance work interval in consideration of a value within the range, and can suppress deterioration in maintenance quality when the failure occurrence time fluctuates.

  In the maintenance work interval determination device according to the present invention, failure distribution setting means for setting a failure distribution which is a statistical distribution assumed to follow a failure occurrence rate in a facility to be subjected to maintenance work, and a failure occurrence record in the facility And storage means for preliminarily storing actual data of maintenance work results, actual number calculation means for calculating the actual failure number based on the failure occurrence results of the actual data, and the failure occurrence rate according to the failure distribution In addition, for the failure distribution, a difference between the assumed failure number and the actual failure number is set in advance. A number-of-cases matching condition determining means for determining whether or not the number of matched conditions is satisfied, and the failure distribution determined to satisfy the number of matching conditions Zui it, a structure in which and a working distance determining means for determining a maintenance interval in the facility.

  In the maintenance work interval determination device according to the present invention, a value within a certain range that includes a true value is considered for a parameter that determines a statistical distribution that is assumed to follow a failure occurrence rate in a facility that is a maintenance work target. Thus, it is possible to determine the maintenance work interval and to suppress the deterioration of the maintenance quality when the failure occurrence time fluctuates.

It is a block diagram which shows the functional structure of the maintenance work interval determination apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows an example of the failure distribution used with the maintenance work space | interval determination apparatus which concerns on Embodiment 1 of this invention. It is a figure explaining the calculation method of the number of assumption faults in the maintenance work space | interval determination apparatus which concerns on Embodiment 1 of this invention. It is a figure explaining the correlation test used for determination of the distribution matching condition in the maintenance operation | work interval determination apparatus which concerns on Embodiment 1 of this invention. It is a figure explaining the determination method of the maintenance work interval using the failure distribution which satisfy | fills each condition in the maintenance work interval determination apparatus which concerns on Embodiment 1 of this invention. It is a flowchart which shows the flow of operation | movement of the maintenance work interval determination apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows the outline of the probability distribution plane centering on the parameter which the failure distribution used by the maintenance work interval determination apparatus concerning Embodiment 2 of this invention can take, and the cumulative probability distribution corresponding to each coordinate point. It is the figure which showed the number of assumption failures of the accumulation probability distribution corresponding to each coordinate point in the probability distribution plane of FIG. It is the figure which showed the area | region which satisfy | fills the isoline of the number of assumption failures corresponding to each coordinate point in the probability distribution plane of FIG. It is the figure which showed the isoline of the maintenance operation | work interval corresponding to the area | region which satisfy | fills number-of-match conditions in the probability distribution plane of FIG. 7, and each coordinate point.

  DESCRIPTION OF EMBODIMENTS Embodiments of the present invention will be described with reference to the accompanying drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and overlapping descriptions are simplified or omitted as appropriate. The present invention is not limited to the following embodiments, and various modifications can be made without departing from the spirit of the present invention.

Embodiment 1 FIG.
FIGS. 1 to 6 relate to Embodiment 1 of the present invention, FIG. 1 is a block diagram showing a functional configuration of a maintenance work interval determination device, and FIG. 2 shows a failure distribution used in the maintenance work interval determination device. FIG. 3 is a diagram illustrating an example, FIG. 3 is a diagram illustrating a method for calculating the number of assumed failures in the maintenance work interval determination device, FIG. 4 is a diagram illustrating a correlation test used for determining a distribution matching condition in the maintenance work interval determination device, and FIG. FIG. 6 is a diagram for explaining a method for determining a maintenance work interval using a failure distribution that satisfies each condition in the maintenance work interval determination device, and FIG. 6 is a flowchart showing an operation flow of the maintenance work interval determination device.

  As shown in FIG. 1, the maintenance work interval determination device according to the present invention includes a failure distribution selection unit 1, a subdivision condition setting unit 2, a failure occurrence result calculation unit 4, a result data storage unit 3, and a work interval calculation unit 10. I have. The fault distribution selection unit 1 is for selecting the type of statistical distribution used as the fault distribution. Here, the failure distribution is a statistical distribution that is assumed to follow the failure occurrence rate in the facility to be maintained.

  Further, here, the failure occurrence rate C (m) is defined as “probability that a failure will occur during one month after m months have passed since the previous maintenance work”. More precisely, regarding “one month after m months”, the month in which the maintenance work was performed is treated as “one month after zero months”, and the next month (from the first day to the end of the month) is “one month has passed. “For one month,” and so on. In the following description, the failure occurrence rate refers to the monthly failure occurrence rate. When evaluating the annual failure rate, a value obtained by adding 12 months of the monthly failure rate may be used.

  As shown in FIG. 2, the failure occurrence rate C (m) defined as described above has a property that the value increases as the work interval is longer, that is, as m is larger, and gradually approaches 100%. have. Specifically, for example, a Weibull distribution, a gamma distribution, a lognormal distribution, or the like can be used as the statistical distribution followed by the failure occurrence rate C (m) having such properties. Each of these statistical distributions has two parameters (for example, the shape parameter and the scale parameter in the Weibull distribution and the gamma distribution, and the average and the standard deviation in the lognormal distribution).

  The user of the maintenance work interval determination device uses the failure distribution selection unit 1 as a failure distribution that is a statistical distribution that the failure occurrence rate C (m) follows, such as the Weibull distribution, the gamma distribution, and the lognormal distribution described above. Of these, a specific distribution to be used is selected. Hereinafter, a case where the Weibull distribution is selected as the statistical distribution used as the failure distribution will be described as an example.

  The subdivision condition setting unit 2 shown in FIG. 1 is for setting subdivision conditions. Subdivision conditions are factors that may affect the failure rate C (m), such as the model of the facility subject to maintenance work or the content of maintenance work, and are subdivided for each of these factors for maintenance. This is a condition set when it is desired to determine the work interval more accurately. The user of the maintenance work interval determination device sets the presence / absence of the subdivision condition and the content thereof by the subdivision condition setting unit 2.

  The performance data storage unit 3 stores in advance performance data for a facility to be subjected to maintenance work. The performance data is information related to the failure occurrence record and the maintenance work record. The actual occurrence of failure is the date, content, number, etc. of failures that have actually occurred in the past at each site where the facility to be maintained is installed. However, here, since the failure occurrence rate C (m) is a failure occurrence rate in units of one month, the failure occurrence results may be at least the number of failure occurrences at each site per month.

  In addition, the maintenance work results are the date, contents, number of cases, and the like of maintenance work actually performed in the past at each site where the facility to be subjected to the maintenance work is installed. However, the maintenance work results should have at least a maintenance work interval in units of one month at each site. In this way, the record data storage unit 3 constitutes storage means for storing in advance the failure occurrence record and the maintenance record of the maintenance work in the facility that is the object of the maintenance work.

  The failure occurrence result calculation unit 4 calculates the actual failure number and the result distribution based on the failure occurrence result of the result data stored in the result data storage unit 3. The actual number of failures is the total number of failures actually occurring at each site where a facility to be subjected to maintenance work is installed within a predetermined period.

  The actual distribution is the total number of failures actually occurring at each site where the facility to be subjected to maintenance work is installed for each number of months elapsed since the previous maintenance work. In other words, it can be said that the actual distribution is a re-expression of the actual failure occurrence in the same format as the failure distribution with the failure occurrence rate C (m).

  In this way, the failure occurrence record calculation unit 4 includes a record number calculation unit that calculates the number of record failure cases based on the failure occurrence record of the record data stored in the record data storage unit 3 that is a storage unit, and a storage unit. It also serves as a result distribution calculating means for calculating a result distribution which is a failure occurrence distribution of results based on the result of failure occurrence of the result data stored in the result data storage unit 3.

  The work interval calculation unit 10 calculates the failure distribution selected by the failure distribution selection unit 1, the subdivision conditions set by the subdivision condition setting unit 2, the actual data stored in the actual data storage unit 3, and the actual occurrence of failure Based on the number of actual failures calculated by the unit 4 and the actual distribution, a maintenance operation interval that can be commonly applied to each site where the facility to be subjected to the maintenance operation is installed is calculated. As shown in FIG. 1, the work interval calculation unit 10 includes a failure distribution parameter setting unit 11, an assumed failure number calculation unit 12, a number match condition determination unit 13, a distribution match condition determination unit 14, a work interval candidate calculation unit 15, A work interval candidate storage unit 16 and a work interval determination unit 17 are provided.

  As described above, the failure distribution followed by the failure rate C (m) includes parameters, and by determining these parameters, the failure rate C (m) as a function of the number of elapsed months m is determined. Can do. The failure distribution parameter setting unit 11 enumerates possible parameter values for the failure distribution selected by the failure distribution selection unit 1 and sequentially sets the enumerated parameter values in the failure distribution. In this way, the failure distribution selection unit 1 and the failure distribution parameter setting unit 11 provide failure distribution setting means for setting a failure distribution that is a statistical distribution that is assumed to follow the failure occurrence rate in the facility to be maintained. It is composed.

  The assumed failure number calculation unit 12 calculates the number of assumed failures when the failure occurrence rate C (m) follows the failure distribution set by the failure distribution selection unit 1 and the failure distribution parameter setting unit 11 which are failure distribution setting means. calculate. The number of assumed failures is based on the assumption that the failure rate C (m) follows the set failure distribution, and at the work intervals that correspond to the maintenance work results at each site where the facility subject to maintenance work is installed. This is the total number of failures that are expected to occur during the predetermined period when maintenance work is performed. As the maintenance work results used in the calculation of the number of assumed failures, those of the result data stored in the result data storage unit 3 are used.

  The method for calculating the number of assumed failures will be described in detail below. First, assuming that an M-month interval is left from the previous maintenance work to the next maintenance work for one piece of equipment at a certain site, the expected number of failures of the equipment during the M month is the failure occurrence rate C described above. From the definition of (m), it can be obtained by the following equation (1).

  (Estimated number of failures) = C (1) + C (2) +... + C (M−1) + C (M) (1)

  In addition, it is unclear whether the failure occurred before or after the work execution month (in this example, the 0th month and the Mth month). For this reason, in order to be more accurate, it is desirable to exclude the failure occurrence rates (C (0) and C (M)) in the work execution month from the calculation of the number of assumed failures. However, for the convenience of explanation, it is assumed that C (M) is also used for calculating the number of assumed failures.

  Next, a method for calculating the number of assumed failures of the equipment from a plurality of work results for the same equipment and work contents will be described with reference to an example shown in FIG. As shown in FIG. 3, in this example, it is assumed that there are 12 work results for the same equipment and work content. Of these 12 work achievements, there is one achievement with a work interval of 3 months, 1 achievement with a work interval of 6 months, 6 achievements with a work interval of 12 months, Assume that there are 3 achievements in which the work interval was 15 months and 1 achievement in which the work interval was 18 months. In this case, the total number of expected failures of these results can be obtained by the following equation (2) because the sum is obtained by applying equation (1) for each result.

  (Estimated number of failures) = 12 × C (1) + 12 × C (2) + 12 × C (3) + 11 × C (4) + 11 × C (5) + 11 × C (6) + 10 × C (7) + 10 × C (8) + 10 × C (9) + 10 × C (10) + 10 × C (11) + 10 × C (12) + 4 × C (13) + 4 × C (14) + 4 × C (15) + C (16) + C (17) + C (18) (2)

  The number coincidence condition determination unit 13 includes the number of assumed failures calculated by the number of assumed failures calculated by the number of assumed failures calculated for the failure distribution set by the failure distribution selection unit 1 and the failure distribution parameter setting unit 11 as failure distribution setting means, It is determined whether or not the difference from the actual failure number calculated by the occurrence result calculation unit 4 satisfies a preset number of cases matching condition. The number matching condition is specifically set to whether or not the difference between the number of assumed failures and the number of actual failures falls within an allowable range of ± 5% with respect to the number of actual failures. Alternatively, it may be possible to set the allowable range of the difference between the assumed failure number and the actual failure number with respect to the actual failure number to a certain number.

  By imposing the number matching condition in this way, the statistical distribution in which the number of failures assumed when following the candidate statistical distribution is in the vicinity of the number of actual failures is extracted, and a distribution that is robust to fluctuations in the time of failure occurrence is obtained. Can be extracted.

  The distribution matching condition determination unit 14 first performs a correlation test on the failure distribution determined by the number matching condition determination unit 13 to satisfy the number matching condition with the actual distribution calculated by the failure occurrence result calculation unit 4. Then, it is determined whether or not the result of this correlation test satisfies a preset distribution matching condition.

  Here, the correlation test (correlation test) is for testing whether or not the two series of sequences {Xi} and {Yi} (i-th values correspond to each other) are linked to each other. It is. The correlation test is a test in which, when (Xi, Yi) is plotted on a scatter diagram, the stronger the tendency to be distributed on a straight line, the more correlated there is. When a correlation test is performed, a value called a p-value is obtained as a test result.

  For example, when (Xi, Yi) is plotted for the two series {Xi}, {Yi} shown in FIG. 4A, a scatter diagram as shown in FIG. 4B is obtained. And when the correlation test was performed in this example, p value was 0.001832. In the correlation test, generally, if the p value is less than (1-reliability coefficient), it can be determined that there is a correlation. Therefore, after setting the reliability coefficient in advance, the distribution matching condition is set to, for example, “p value obtained as a result of the correlation test between the failure distribution and the actual distribution is less than (1−reliability coefficient)”.

  By imposing the distribution matching condition in this way, it is possible to extract a failure distribution parameter that becomes a statistical distribution having a certain degree of similarity by paying attention to the similarity of the distribution with the actual results.

  As described above, the failure distribution parameter setting unit 11 lists possible parameter values in the failure distribution, and sequentially sets the listed parameter values in the failure distribution. Then, through the determination by the number matching condition determination unit 13 and the distribution matching condition determination unit 14, only parameters that satisfy both the number matching condition and the distribution matching condition are extracted.

  The work interval candidate calculation unit 15 calculates a work interval candidate for each failure distribution in which the values of the parameters extracted in this way are set. More specifically, the work interval candidate calculation unit 15 determines the number of expected failures when all maintenance work at each site where the facility to be subjected to maintenance work is performed at the same work interval N months, as desired. The work interval N months, which is the number of cases (for example, the number of cases equivalent to the number of actual failures), is calculated as a candidate for the work interval.

  The calculation of the work interval candidate will be described with reference to FIG. 5 with a specific example. In the example of FIG. 5, a failure distribution in which a certain parameter (which satisfies these conditions) extracted based on the number matching condition and the distribution matching condition is set is C1 (m). Then, it is assumed that the assumed number of failures for C1 (m) when all maintenance work is performed at the same work interval N months is as shown in the table in FIG.

  That is, for example, when N is 10, the number of assumed failures is 15.2, when N is 12, the number of assumed failures is 18.1, and when N is 14, the number of assumed failures is 22.4. The number of assumed failures when N is 16 is 23.1, the number of assumed failures is 31.7 when N is 18, and the number of assumed failures is 45.5 when N is 20. Suppose that

  Here, if the actual number of failures is 23, the number of assumed failures equal to 23 is that N is 14 (the number of assumed failures is 22.4) and N is 16 (the number of assumed failures is 23. 1 case). Here, when there are a plurality of N equivalent to the actual number of failures, a larger N is set as a work interval candidate, that is, in this example, a work interval candidate in the case of a parameter value failure distribution of C1 (m). Calculated to be 16 months.

  In this way, the work interval candidate calculation unit 15 detects not only the parameter value to be C1 (m) but also the failure distribution of all parameter values extracted by satisfying the number matching condition and the distribution matching condition (satisfying these conditions). The work interval candidate for is calculated.

  The work interval candidate storage unit 16 stores the work interval candidates calculated by the work interval candidate calculation unit 15. In other words, the work interval candidate storage unit 16 stores each of the work interval candidates in the failure distribution of each parameter value extracted by the number matching condition and the distribution matching condition (satisfying these conditions).

  Then, the work interval determination unit 17 determines a maintenance work interval in a facility to be subjected to maintenance work from among the work interval candidates stored in the work interval candidate storage unit 16. Specifically, the work interval determination unit 17 determines the shortest one of the work interval candidates stored in the work interval candidate storage unit 16 as the maintenance work interval in the facility to be subjected to the maintenance work. In this way, in any possible situation, it is possible to determine a maintenance work interval that keeps the statistically assumed failure occurrence rate below a specified threshold value.

  As described above, the work interval candidate calculation unit 15, the work interval candidate storage unit 16, and the work interval determination unit 17 perform maintenance based on the failure distribution determined by the distribution match condition determination unit 14 as satisfying the distribution match condition. Work interval determination means for determining the maintenance work interval in the facility to be worked is configured. Here, as described above, the determination by the distribution matching condition determination unit 14 is performed for the failure distribution determined by the number matching condition determination unit 13 to satisfy the number matching condition. Therefore, the work interval determination means performs maintenance work based on the failure distribution determined by the number matching condition determination unit 13 that the number matching condition is satisfied and determined by the distribution matching condition determination unit 14 that the distribution matching condition is satisfied. This means that the maintenance work interval at the target facility is determined.

  The maintenance work interval determination device having the functional configuration as described above can be implemented using, for example, a known computer. Then, next, an example of a hardware configuration for mounting the maintenance work interval determination device will be described.

  The maintenance work interval determination device includes a CPU, a ROM, a RAM, and a communication interface. The CPU is a central processing unit that executes a given instruction (including a program that is a set of instructions) to calculate or process information. The ROM is a read-only memory that stores data or programs in a nonvolatile manner. The RAM is a readable / writable random access memory (volatile access memory) for storing data or programs in a volatile manner.

  The communication interface is hardware necessary for the maintenance work interval determination device to communicate with the outside. Here, the maintenance work interval determination device is connected to the Internet, which is a communication network, via a communication interface. These CPU, ROM, RAM, and communication interface are connected to each other via a bus so that information can be exchanged between them.

  The maintenance work interval determination device may include a hard disk, a CD-ROM, or the like as an external storage device, or may include a display output unit (for example, a monitor), an input unit (for example, a keyboard), or the like.

  The ROM of the computer constituting the maintenance work interval determination device stores in advance a program for operating the computer configured as described above as the maintenance work interval determination device. The CPU first reads the program from the ROM and stores the program in a format that can be executed in the RAM. The CPU executes the program stored in the RAM while reading it.

  In this way, when the software program is read into the computer, the calculation or processing of the information is realized by specific means in which the software and the hardware resources cooperate, as described above. A maintenance work interval determination device having each function shown in FIG. 1 is constructed.

  That is, the program executed in the CPU of the computer of the maintenance work interval determination device is the same as the computer of the maintenance work interval determination device, the failure distribution selection unit 1, the segmentation condition setting unit 2, the result data storage unit 3, and the failure occurrence result calculation. Unit 4 and work interval calculation unit 10 (failure distribution parameter setting unit 11, assumed failure number calculation unit 12, number match condition determination unit 13, distribution match condition determination unit 14, work interval candidate calculation unit 15, work interval candidate storage unit 16 And function as each part of the work interval determination unit 17).

  The program may be stored in an external storage device instead of the ROM, or the program may be downloaded from the outside to a computer via a communication interface.

  An example of the flow of maintenance work interval determination by the maintenance work interval determination device configured as described above will be described once again with reference to the flowchart of FIG. First, in step S1, the failure distribution selection unit 1 selects a failure distribution (for example, Weibull distribution or the like) that is assumed to follow the failure occurrence rate with respect to the work interval months. In subsequent step S2, the subdivision conditions are set by the subdivision condition setting unit 2.

  Then, the process proceeds to step S3, and the failure occurrence result calculation unit 4 calculates the actual failure number and the result distribution with respect to the work interval months based on the failure occurrence result of the result data stored in the result data storage unit 3. .

  After step S3, the process proceeds to step S10. This step S10 is a step in which the work interval calculation unit 10 obtains a lower limit of the number of work interval months in which the number of assumed failures falls within the allowable range from the failure distribution that matches the actual results. This step S10 is further subdivided into the following steps S11 to S16.

  First, in step S11, the failure distribution parameter setting unit 11 enumerates possible parameter values for the failure distribution selected in step S1, and selects one of the enumerated parameter values (when there are a plurality of parameters). 1 set) is set as the failure distribution.

  In subsequent step S12, first, the assumed failure number calculation unit 12 calculates the number of assumed failures when the failure occurrence rate follows the failure distribution set in step S11. Then, the number matching condition determination unit 13 determines whether or not the difference between the number of assumed failures and the actual number of failures calculated in step S3 satisfies the number matching condition for the failure distribution set in step S11. If the failure distribution set in step S11 does not satisfy the number matching condition, the process returns to step S11, sets another value for the parameter, and repeats the determination in step S12. If the failure distribution set in step S11 satisfies the number matching condition, the process proceeds to step S13.

  In step S13, the distribution matching condition determination unit 14 performs a correlation test on the failure distribution set in step S11 with the actual distribution calculated in step S3, and whether the result of this correlation test satisfies the distribution matching condition. Determine whether. If the failure distribution set in step S11 does not satisfy the distribution matching condition, the process returns to step S11, sets another value for the parameter, and starts again from the number matching condition determination in step S12. On the other hand, if the failure distribution set in step S11 also satisfies the distribution matching condition, the process proceeds to step S14.

  In step S14, the work interval candidate calculation unit 15 calculates work interval candidates based on the failure distribution set in step S11. Then, the calculated work interval candidates are stored (saved) in the work interval candidate storage unit 16, and the process proceeds to step S15.

  In step S15, the work interval calculation unit 10 confirms whether or not the number matching condition determination process of at least step S12 has been completed for the failure distribution of all the parameter values listed in step S11. If the processing has not been completed for all the parameter values, the process returns to step S11, another value is set for the parameter, and the number matching condition determination in step S12 is repeated. When the processing is completed for all parameter values, the process proceeds to step S16.

  In step S <b> 16, the work interval determination unit 17 determines a maintenance work interval in a facility to be subjected to maintenance work from among the work interval candidates stored in the work interval candidate storage unit 16. When step S10 (steps S11 to S16) is completed, the process proceeds to step S4.

  In step S4, it is confirmed whether or not the processing in step S10 has been completed for all the subdivision conditions. If the process of step S10 has not been completed for all the subdivision conditions, the process returns to step S2, another subdivision condition is set, and steps S3 and S10 are performed. On the other hand, if the process of step S10 is completed for all the subdivision conditions, the series of processes ends.

  In the above, the failure distribution of the parameter value that satisfies both the number matching condition and the distribution matching condition is extracted, and the maintenance work interval in the facility is determined. Only the number matching condition may be determined. That is, the work interval determination means may determine the maintenance work interval in the facility based on the failure distribution determined to satisfy the distribution matching condition.

  The maintenance operation interval determination device configured as described above is a failure distribution selection unit that is a failure distribution setting unit that sets a failure distribution that is a statistical distribution that is assumed to follow a failure occurrence rate in a facility that is a subject of maintenance work. 1 and a failure distribution parameter setting unit 11, a record data storage unit 3 which is a storage means for storing record data of failure occurrence results and maintenance work results in the facility, and the number of actual failure cases based on the failure occurrence record of the result data If the failure occurrence rate is in accordance with the failure distribution, the number of assumed failures that are assumed to occur at the work operation maintenance interval of the result data is calculated. Assumed failure number calculation unit 12 determines whether or not the difference between the assumed failure number and the actual failure number satisfies a preset number matching condition for the failure distribution. A work interval candidate calculating unit 15 which is a work interval determining means for determining a maintenance work interval in the facility based on the failure distribution determined to satisfy the number match condition, and a work interval candidate storage. Unit 16 and work interval determination unit 17.

  For this reason, the maintenance work interval should be determined in consideration of a value within a certain range that includes a true value for the parameter that determines the statistical distribution assumed to follow the failure rate in the facility subject to maintenance work. It is possible to suppress the deterioration of the maintenance quality when the failure occurrence time fluctuates.

  In addition, since the number matching condition for extracting parameters can be set based on the fact that the number of failures has occurred, the setting of the maintenance quality standard assumed for determining the maintenance work interval is clear.

Embodiment 2. FIG.
FIGS. 7 to 10 relate to Embodiment 2 of the present invention, and FIG. 7 is a probability distribution plane centered on parameters that can be taken by the failure distribution used in the maintenance work interval determination device and the accumulation corresponding to each coordinate point. FIG. 8 is a diagram showing an outline of the probability distribution, FIG. 8 is a diagram showing the number of assumed failures of the cumulative probability distribution corresponding to each coordinate point in the probability distribution plane of FIG. 7, and FIG. 9 is each coordinate point in the probability distribution plane of FIG. FIG. 10 is a diagram showing an isoline of the number of assumed failures corresponding to the number and a region satisfying the number matching condition, FIG. 10 is a diagram showing the region satisfying the number matching condition in the probability distribution plane of FIG. It is the figure which showed the value line.

  In the first embodiment described above, parameter values that can be taken by the failure distribution are listed, and these values are set in order to determine the number-match condition or the like. In contrast, the second embodiment described here associates a failure distribution with each coordinate point in the parameter space with the failure distribution parameter as a space axis, and sets a set of parameters satisfying the number matching condition and the like. The work interval is determined by expressing it as an area in the parameter space.

  The functional configuration of the maintenance work interval determination device according to the second embodiment is the same as that shown in FIG. 1 described in the first embodiment. As described above in the first embodiment, the Weibull distribution, the gamma distribution, and the lognormal distribution adopted as the statistical distribution (failure distribution) that the failure occurrence rate C (m) follows are all two parameters (shape parameters). And scale parameters).

  Therefore, when a plane (parameter space) represented by a coordinate system having a shape parameter as the first parameter on the vertical axis and a scale parameter as the horizontal axis as the second parameter as shown in FIG. 7 is considered, One coordinate point corresponds to a set of shape parameter values and scale parameter values. If the combination of the shape parameter value and the scale parameter value is determined, the failure distribution corresponding to this set is also determined. Therefore, each point on the plane represented by the coordinate system with each parameter as a coordinate axis corresponds to one failure distribution, that is, a cumulative probability distribution.

  That is, also in the second embodiment, the statistical distribution assumed that the failure occurrence rate in the facility to be subjected to the maintenance work is obeyed by the failure distribution selection unit 1 and the failure distribution parameter setting unit 11 constituting the failure distribution setting means. A failure distribution is set. However, the set result is expressed as a probability distribution plane as shown in FIG.

  When the parameter value is determined and the failure distribution is determined, the number of assumed failures can be calculated in the same manner as in the first embodiment. That is, as shown in FIG. 8, the number of assumed faults is obtained corresponding to the fault distribution corresponding to each point on the probability distribution plane of FIG.

  Therefore, also in the second embodiment, the assumed failure number calculation unit 12 includes the failure occurrence rate C (m) in the failure distribution set by the failure distribution selection unit 1 and the failure distribution parameter setting unit 11 which are failure distribution setting means. Calculate the number of assumed failures if However, the calculated number of assumed failures is expressed as a value (isoline) on the probability distribution plane as shown in FIGS.

  Then, the number-of-cases matching condition determination unit 13 determines the number of assumed failures in the failure distribution corresponding to the coordinate points and the number of actual failures calculated by the failure occurrence result calculation unit 4 for each coordinate point on the probability distribution plane of FIG. It is determined whether or not the difference between and satisfies a preset number matching condition. A set of coordinate points that are confirmed to satisfy the number matching condition by this determination form a region 20 on the probability distribution plane of FIG.

  For example, assuming that the number of actual failures is 23 and the allowable range that satisfies the number matching condition is the number of actual failures ± 1, the set of coordinate points of the failure distribution satisfying the number matching condition is an area as shown in FIG. 20 Note that there are coordinate points of a set of parameters that give a true failure occurrence distribution to any one of the areas 20.

  In addition, the work interval candidate calculation unit 15, the work interval candidate storage unit 16, and the work interval determination unit 17 constituting the work interval determination unit determine the maintenance work interval in the facility to be subjected to the maintenance work as follows. .

  That is, first, the work interval candidate calculation unit 15 calculates a work interval candidate corresponding to a failure distribution corresponding to each point on the probability distribution plane of FIG. 7 by the same method as in the first embodiment. Thus, the value of the work interval candidate calculated corresponding to each point on the probability distribution plane draws an isoline 30 as shown in FIG.

Then, the work interval determination unit 17 selects the one with the smallest value (the shortest period) among the isolines 30 of the work interval candidates passing through the region 20 of the coordinate point of the failure distribution satisfying the number matching condition. Determine as work interval. In the example illustrated in FIG. 10, the work interval determination unit 17 determines 15 to 16 months as the maintenance work interval. By determining the maintenance work interval in this way, it is possible to prevent the maintenance quality from being insufficient even if any parameter in the region 20 gives a true failure occurrence rate distribution. is there.
Other configurations are the same as those in the first embodiment, and detailed description thereof is omitted.

  In the maintenance work interval determination device configured as described above, in addition to having the same effects as those of the first embodiment, the failure distribution parameters satisfying the number matching condition as the region 20 on the probability distribution plane can be obtained. Can be extracted and the maintenance work interval determination process can be efficiently performed.

DESCRIPTION OF SYMBOLS 1 Failure distribution selection part 2 Subdivision condition setting part 3 Actual data storage part 4 Failure occurrence result calculation part 10 Work interval calculation part 11 Failure distribution parameter setting part 12 Expected failure number calculation part 13 Number matching condition determination part 14 Distribution matching condition determination 15 Work interval candidate calculation unit 16 Work interval candidate storage unit 17 Work interval determination unit

Claims (2)

A failure distribution setting means for setting a failure distribution which is a statistical distribution assumed to follow a failure occurrence rate in a facility to be subjected to maintenance work;
Storage means for preliminarily storing performance data of failure occurrence and maintenance work in the facility;
Based on the actual occurrence of failure in the actual data, the actual number calculating means for calculating the actual number of failures,
Assuming that the failure occurrence rate follows the failure distribution, an assumed number calculating means for calculating the number of assumed failures that are assumed to occur in the work interval of the maintenance work results of the actual data;
For the failure distribution, the number match condition determination means for determining whether or not the difference between the assumed failure number and the actual failure number satisfies a preset number match condition;
A maintenance work interval determination device comprising: work interval determination means for determining a maintenance work interval in the facility based on the failure distribution determined to satisfy the number matching condition. Based on the failure occurrence record of the record data, a record distribution calculating means for calculating a record distribution that is a failure occurrence distribution of the record;
Distribution failure condition determination means for determining whether or not the result of correlation test with the actual result distribution satisfies a preset distribution match condition for the failure distribution determined to satisfy the number match condition,
The maintenance work interval determination device according to claim 1, wherein the work interval determination unit determines a maintenance work interval in the facility based on the failure distribution determined to satisfy the distribution matching condition.

Images