JP2005032202A - Resource allocation method and recording medium having program recorded therein - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To resolve the problem that a method is not established which considers how respective factors are linked with a problem event caused by several factors, to effectively allocate resources when allocating resources required for taking measures against factors in order to avoid the occurrence of the problem event. <P>SOLUTION: A method of fault tree analysis is used to set relations between the problem event caused by several factors and respective factors being causes of the occurrence of the problem event, and meanwhile, use resource functions showing correlations between inputs of resources required for taking measures to avoid respective factors and occurrence probabilities of respective factors are set, and an occurrence probability of the problem event obtained by the method of fault tree analysis and data of functions showing correlations between occurrence probabilities of respective factors prescribed by the use resource functions and inputs of resources are stored in a storage device, and these values are varied in accordance with restrictive conditions to calculate the most suitable inputs of resources, so that resources can be effectively allocated. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a resource allocation method for effectively allocating resources necessary for implementing measures to avoid the occurrence of a problem event caused by several factors, and a recording medium on which the program is recorded.

  In order to avoid the occurrence of problem events caused by some factors, it is necessary to take some measures against those factors. However, if the resources required to implement these measures are limited, the resources must be allocated to each measure in some way.

  Conventionally, there is a linear programming method for quantitatively allocating resources. The linear programming method is a method of formulating several constraint conditions to obtain the maximum value / minimum value of an output with respect to resource input, and is mainly used for a resource allocation method for production planning. However, when allocating resources to countermeasures for each factor implemented to avoid the occurrence of problem events caused by several factors as described above, it is necessary to consider how each factor is related to the problem event However, in linear programming, no method for formulating that part has been established.

  On the other hand, fault tree analysis (Fault Tree Analysis) can be cited as a method for clarifying how each factor relates to the problem event. Fault tree analysis is a method of reliability analysis for extracting the factors that cause problem events, expressing their relationships, and performing various analyzes and evaluations. The method was not devised.

  As described above, the conventional method considers the relationship of each factor to the problem event for each measure to be performed to avoid the occurrence of the problem event caused by some factors. Therefore, it has been difficult to effectively allocate resources necessary for implementing each of the above measures.

  Therefore, an object of the present invention is to solve the above-mentioned problems by establishing a resource allocation method that can cope with such problems and performing computations using a computer using the program of the method.

  Note that “perform resource allocation effectively” specifically means minimizing the input amount of resources necessary to execute each of the measures without changing the occurrence probability of the problem event from a given value. Reducing the probability of occurrence of the problem event without changing the input amount of the resource from a given value, and considering the damage caused by the occurrence of the problem event Is the optimal value, and the optimal allocation of resources is derived from the above constraints.

  In order to solve the above problems, the invention described in claim 1 is a computer having an input device, a storage device, a computing device, and an output device, and identifies the factors that cause the occurrence of a problem event and expresses the relationship between them. In a recording medium that records a program that calculates the occurrence probability of the top event from the occurrence probability of each basic event in the fault tree analysis, which is a method of reliability analysis, a problem event caused by several factors input from the input device is recorded. The occurrence data is a top event, and one or more factors that cause the occurrence of the problem event as a basic event, the configuration data of the fault tree diagram that represents the relationship structure, the occurrence probability of each factor and the occurrence of each factor To store data of a function (hereinafter referred to as “used resource function”) indicating a correlation with the input amount of resources of each countermeasure to be performed, The resource related to each of the measures that satisfies the condition that the occurrence probability of the problem event calculated using a program for calculating the occurrence probability of the top event in the fault tree analysis is not changed from a given value by a computing device Input amount is set by the used resource function, a combination in which the total amount of the resource input for each countermeasure is minimized is searched, and the combination that is a search result is displayed on the output device. This is a characteristic resource allocation method.

  Further, the invention according to claim 2 is a fault analysis method that is a reliability analysis technique for extracting a factor causing a problem event and expressing the relationship between the computer having an input device, a storage device, an arithmetic device, and an output device. In the recording medium recording the program for calculating the occurrence probability of the top event from the occurrence probability of each basic event in the tree analysis, the problem event caused by several factors input from the input device is regarded as the top event. Fault tree diagram configuration data having one or more factors that cause the occurrence of the occurrence as a basic event, the occurrence probability of each factor, and the input amount of resources for each measure to be performed to avoid the occurrence of each factor, The resource function data indicating the correlation between the resources is stored in a storage device, and the total input amount of the resource for each countermeasure is changed from a given value by the arithmetic unit. The occurrence probability of each factor that satisfies the condition of not to be set is set by the resource function used, and among them, the problem event calculated by the arithmetic unit using a program for obtaining the occurrence probability of the top event in the fault tree analysis It is a resource allocation method characterized by searching for a combination with the lowest occurrence probability and displaying the combination as a search result on an output device.

  Further, the invention according to claim 3 is a fault analysis method that is a reliability analysis technique for extracting a factor that causes a problem event and expressing the relationship between the computer having an input device, a storage device, an arithmetic device, and an output device. In the recording medium recording the program for calculating the occurrence probability of the top event from the occurrence probability of each basic event in the tree analysis, the occurrence of the problem event caused by some factors input from the input device is regarded as the top event. Fault tree diagram configuration data with one or more factors that cause the occurrence of the occurrence as basic events, the probability of occurrence of each factor, and the input of resources necessary for each countermeasure to be implemented to avoid the occurrence of each factor The resource function data indicating the correlation with the amount is stored in a storage device, and the probability of occurrence of the problem event arbitrarily set by the arithmetic device and The fault tree analysis with respect to the expected value of the damage indicated by the product of the damage values caused by the occurrence of the problem event expressed in units of the amount of resources and the value of the occurrence probability of the problem event set arbitrarily. Each of the countermeasures set by the resource function used based on the occurrence probability of each factor such that the value of the top event calculated by the arithmetic unit using the program for determining the occurrence probability of the top event in For each of the measures, the sum of the total input amount of the resources with respect to the minimum value becomes the minimum when the value of the probability of occurrence of the previously set problem event is changed within a certain range. The resource allocation method includes searching for a combination of the input amounts of the resources and displaying the combination as a search result on an output device.

  The above invention is realized by performing processing on a computer using a medium recording a program characterized by such a resource allocation method.

  Further, when executing the processing on the computer, the occurrence probability of one or more factors that cause the occurrence of a problem event caused by some factors, and each measure to be performed to avoid the occurrence of each factor Instead of inputting the resource function data indicating the correlation with the resource input amount from the input device, the generation time and frequency of each factor input by the input device and the resource input amount and input of each countermeasure By using on a computer a recording medium that records a program characterized by calculating the data of the used resource function based on time data, it is possible to derive a resource allocation result with less input burden.

  As described above, according to the present invention, with respect to each countermeasure to be implemented for avoiding the occurrence of a problem event caused by some factors, which is difficult with the conventional method, Establishing a resource allocation method characterized by effectively allocating resources necessary for implementing each of the measures in consideration of the relationship between the factors, and calculating with a computer using the program of the method The above-mentioned problem can be solved by performing the above.

  Note that “perform resource allocation effectively” specifically means minimizing the input amount of resources necessary to execute each of the measures without changing the occurrence probability of the problem event from a given value. Reducing the probability of occurrence of the problem event without changing the input amount of the resource from a given value, and considering the damage caused by the occurrence of the problem event Is the optimum value, and the optimum allocation of the resource input is derived from the above constraints, which are realized by the methods of the present invention.

  The best mode for carrying out the present invention will be described below. FIG. 1 shows a configuration example of a resource allocation support system related to maintenance of a certain device. The “failure mode” shown below is defined as an event that causes the device to stop when it occurs. The storage device 2 stores the failure mode / maintenance history information such as the name of the failure mode, the information on the occurrence time of each failure mode, the maintenance content corresponding to a certain failure mode, the maintenance execution time, and the maintenance cost from the input device 1. Store in the part 2a. The data is stored in a format like the table shown in FIG. It is assumed that a certain failure mode and a certain maintenance have a one-to-one relationship, and the occurrence probability of a certain failure mode varies depending on the corresponding maintenance execution period. Further, the configuration data of the fault tree diagram in which the stop of the device is a top event and the occurrence of each failure mode is a basic event is input from the input device 1 of FIG. 1 and stored in the storage unit 2 b of the storage device 2. In the fault tree diagram, when any failure mode that is a basic event occurs in any combination, the fact that the device that is the top event is stopped represents the relationship. An example of a fault tree diagram is shown in FIG.

  Based on the information stored in the storage unit 2a, the calculation unit 3a of the calculation device 3 calculates the occurrence probability of each failure mode and the maintenance cost corresponding thereto from the occurrence probability of each failure mode, the number of times of maintenance performed corresponding thereto, and the maintenance cost. A resource function that indicates the correlation is calculated. An example of the calculation method of the resource function used is described below. However, it is assumed that each failure mode has occurred n + 1 times necessary for creating a function. Regarding the occurrence interval (TBF (1 to n)) of each failure mode, the number of maintenance times (Nm (1 to n)) to be performed within each interval of TBF (1 to n) and the maintenance cost (Ce) per Nm The cost for each TBF (1 to n) obtained from the product (Ct (1 to n)) and the failure rate Pt (1 to n) obtained by 1 / TBF (1 to n) Plot the graph to find the distribution. Based on the distribution, by using a nonlinear least square method and applying it to the form of an exponential distribution, a resource function that is a correlation function between a failure rate, that is, a failure mode occurrence probability and a cost is obtained. An example of the function form of the resource function used is as shown in the graph of FIG.

On the condition that the operation stop probability is not changed from a given value based on the calculation result performed by the operation unit 3a and the configuration data of the fault tree diagram stored in the storage unit 2b in the operation unit 3b, A combination of the total minimum maintenance cost for the device, the maintenance cost for each failure mode that satisfies the result, and the maintenance execution cycle is calculated. An example of the calculation method will be described below with reference to FIG. FIG. 5 shows an example in which there are n failure modes with respect to equipment stoppage. In addition, due to calculation restrictions due to fault tree analysis, the occurrence probability of a basic event (in this case, each failure mode) must be within 0.1.
First, in step S1, the occurrence probability of the top event that the device stops is set as an arbitrary constraint value (Pa). On the other hand, in step S2, it is set what scale step (Sc1) the occurrence probability (Pf (1 to n)) of each failure mode (F (1 to n)) is changed. Here, it is assumed to be in increments of 0.001. Under such conditions, in step S3, the probability of occurrence of each failure mode (Pf (1 to n)) is arbitrarily set. The occurrence probability (Pf (1 to n)) of each failure mode is varied in increments of 0.001 in the range of 0.001 to 0.1, and 100 occurrence probability patterns for each failure mode. The following calculation is performed for all combinations by loop processing. In step S4, the occurrence probability (Pf (1-n)) of each failure mode (F (1-n)) and the occurrence probability of equipment stoppage using the fault tree analysis method based on the configuration data of the fault tree diagram (Pb) is obtained. In step S5, the occurrence probability (Pa) fixed in step S1 is compared with the occurrence probability (Pb) obtained in step S4. If the values do not match, the process proceeds to step S12. If they match, the process proceeds to step S6. move on. In step S6, from the occurrence probability (Pf (1-n)) of each failure mode (F (1-n)) satisfying the occurrence probability (Pb) set in step S4, each failure mode (F Each maintenance cost (Ca (1-n)) of (1-n)) is derived, and the total maintenance cost (Cs) is calculated. Next, it is determined in step S7 whether or not the series of operations up to step S6 is the first one, and if it is the first time, the process proceeds to step S9. In step S9, the value of the total maintenance cost (Cs) set in step S6 is stored in the storage area (Cb). On the other hand, if the series of operations leading to step S6 is not the first time, determination is made in step S7, and step S8 is performed. In step S8, the value stored in the storage area (Cb) is compared with the total maintenance cost (Cs). If the total maintenance cost (Cs) is smaller, step S9 is executed, and the storage area (Cb ) Is replaced with the total maintenance cost (Cs). On the other hand, if the value stored in the storage area (Cb) is less than or equal to the total maintenance cost (Cs), the process proceeds to step S12. Further, in step S10, each maintenance cost (Ca (1-n)) corresponding to the value stored in the storage area (Cb) is divided by each maintenance cost (Ce (1-n)). Each maintenance cycle (Mt (1 to n)) is calculated. In step S11, each maintenance cost (Ca (1-n)) and each maintenance cycle (Mt (1-n)) are stored in the output file (Op). In step S12, it is determined whether or not the calculation has been performed for all combinations of occurrence probabilities (Pf (1 to n)) of the respective failure modes (F (1 to n)). If not, the process returns to step S3 to perform the calculation. Perform the calculation again with a combination that has not. When the above loop is repeated and all combinations are performed, the loop is exited, and the value stored in the storage area (Cb) at that time is the total minimum maintenance cost. Therefore, in step S13, the value of the storage area (Cb) is output to the output file Store in (Op). Finally, in step S14, the value stored in the output file (Op) is output as a result.

On the condition that the calculation unit 3c does not change the total maintenance cost of the device from a given value based on the calculation result performed by the calculation unit 3a and the configuration data of the fault tree diagram stored in the storage unit 2b. A combination of the minimum occurrence probability of the stop, the maintenance cost for each failure mode that satisfies the result, and the maintenance execution cycle is calculated. An example of the calculation method will be described below with reference to FIG. FIG. 6 shows an example in which there are n failure modes with respect to equipment stoppage. In addition, due to calculation restrictions by fault tree analysis, the occurrence probability of a basic event (in the present case, each failure mode) must be within 0.1.
First, the maintenance cost total for each failure mode (F (1 to n)) that is step S1 is set to a certain arbitrary value as a constraint value (Cb). In step S2, the scale (Sc1) for changing the occurrence probability (Pf (1 to n)) of each failure mode (F (1 to n)) is set. Here, it is assumed to be in increments of 0.001. Under such conditions, in step S3, the occurrence probability (Pf (1 to n)) of each failure mode (F (1 to n)) is set, and the calculation after step S4 is performed. The occurrence probability (Pf (1 to n)) of each failure mode (F (1 to n)) is varied in increments of 0.001 in the range of 0.001 to 0.1, and 100 patterns per failure mode. The following calculation is performed for all combinations of the occurrence probability patterns by loop processing. In step S4, the maintenance cost (Ca) of each failure mode (F (1-n)) is calculated from the occurrence probability (Pf (1-n)) of each failure mode (F (1-n)) using the resource function used. (1-n)) is derived, and the total maintenance cost (Cs) is calculated. In step S5, the total maintenance cost (Cb) fixed in step S1 is compared with the total maintenance cost (Cs) obtained in step S4. If the values do not match, the process proceeds to step S12. Proceed to S6. In step S6, the occurrence probability (Pa) of equipment stoppage is determined using a fault tree analysis technique based on the occurrence probability (Pf (1 to n)) of each failure mode and the configuration data of the fault tree diagram. Next, it is determined in step S7 whether or not the series of operations up to step S6 is the first one, and if it is the first time, the process proceeds to step S9. In step S9, the value of the device stop occurrence probability (Pa) calculated in step S6 is stored in the storage area (Pb). On the other hand, if the series of operations leading to step S6 is not the first time, the determination is made in step S7, and the process proceeds to step S8. In step S8, the value stored in the storage area (Pb) is compared with the occurrence probability (Pa) of the equipment stop, and if the occurrence probability (Pa) of the equipment stop is smaller, execute step S9. The value stored in the storage area (Pb) is replaced with the value of the device stop occurrence probability (Pa). On the other hand, if the value stored in the storage area (Pb) is equal to or less than the value of the device stop occurrence probability (Pa), the process proceeds to step S12. Furthermore, in step S10, each maintenance cost (Ca (1-n)) of each failure mode (F (1-n)) stored in the storage area (Ca (1-n)) is stored for each maintenance. Dividing by the cost (Ce (1 to n)), each maintenance cycle (Mt (1 to n)) is calculated. In step S11, each maintenance cost (Ca (1-n)) and each maintenance cycle (Mt (1-n)) are stored in the output file (Op). In step S12, it is determined whether the calculation has been performed for all combinations of occurrence probability (Pf (1 to n)) of each failure mode. If not, the process returns to step S3, and the calculation is performed again with the combination that has not been calculated. To do. When the above loop is repeated and all combinations are performed, the loop is exited, and the value stored in the storage area (Pb) at that time is the minimum device stoppage occurrence probability, so in step S13 the value of the storage area (Pb) Are stored in the output file (Op). Finally, in step S14, the value stored in the output file (Op) is output as a result.

The calculation unit 3d obtains the product of the damage cost caused by the device stop and the occurrence probability of the device stop based on the calculation result performed by the calculation unit 3a and the configuration data of the fault tree diagram stored in the storage unit 2b. The minimum value of the assumed total cost that is the sum of the expected value (R) of damage and the total minimum maintenance cost related to the probability of equipment outage corresponding to the expected value (R), and the maintenance for each failure mode that satisfies the result Calculate the combination of cost and maintenance cycle. An example of the calculation method will be described below with reference to FIG. FIG. 7 shows an example in which there are n failure modes with respect to equipment stoppage. In addition, due to calculation restrictions by the fault tree, the occurrence probability of a basic event (in this case, each failure mode) is required to be within 0.1.
First, in step S1, a damage cost (Dc) due to equipment stop is arbitrarily set. In step S2, it is set what scale step (Sc1) the occurrence probability (Pf (1-n)) of each failure mode (F (1-n)) is changed. Here, it is assumed to be 0.001. In step S3, the scale increment (Sc2) at which the device stop occurrence probability (Pa) is changed is set. Here, it is assumed to be 0.01. In step S4, the occurrence probability (Pa) of equipment stop is arbitrarily set. In addition, the occurrence probability (Pa) of the equipment stop is varied in a range of 0.01 to 0.99 in increments of 0.01, and calculation is performed on 99 occurrence probability patterns. In step S5, an expected value (R) of damage is calculated from the product of the damage cost (Dc) and the occurrence probability (Pa) of equipment stoppage. In steps S6 to S15, calculation similar to the calculation in steps S3 to S12 of the calculation unit 3a is performed using the apparatus stop occurrence probability (Pa) as a constraint value, and the apparatus stop occurrence probability (Pa) is set. Find the minimum total maintenance cost. In step S16, an estimated total cost (Cz) that is the sum of the expected value (R) of damage and the minimum total maintenance cost stored in the storage area (Cb) is calculated. In step S17, it is determined whether or not the calculation of the estimated total cost (Cz) is the first calculation. If it is the first time, the process proceeds to step S19, and the value of the estimated total cost (Cz) is stored in the storage area (Cy). On the other hand, if the calculation of the estimated total cost (Cz) is not the first time, it is determined in step S17, and the process proceeds to step S18. In step S18, the value of the storage area (Cy) is compared with the estimated total cost (Cz). If the total estimated cost (Cz) is smaller, the process proceeds to step S19, where it is stored in the storage area (Cy). Is replaced with the value of the assumed total cost (Cz) and stored. If the value of the storage area (Cy) is equal to or less than the estimated total cost (Cz), the process proceeds to step S20. In step S20, it is determined whether or not the calculation has been performed for all patterns of the device stop occurrence probability (Pa). To implement. When the above loop is repeated and all patterns are executed, the value stored in the storage area (Cy) at that time is the minimum value of the estimated total cost (Cz). In step S21, the value of the storage area (Cy) is stored in the output file (Op). In the last step S22, the value stored in the output file (Op) is output as a result.

  When performing equipment maintenance using the above methods, effective maintenance according to the situation at hand, taking into account the relationship of several factors that cause the problem that the equipment has stopped. Cost allocation can be performed.

  FIG. 1 is a configuration example of a resource allocation support system related to maintenance of a certain device.   FIG. 2 is an example of a table for storing failure mode / maintenance history information in a computer.   FIG. 3 is an example of a fault tree diagram in which “device stop” is the top event.   FIG. 4 is an example of a graph of the used resource function regarding the occurrence probability of the failure mode and the maintenance cost corresponding thereto.   FIG. 5 is an example of a flowchart of each maintenance cost allocation calculation that minimizes the total maintenance cost when the probability of equipment stoppage is fixed to an arbitrary value.   FIG. 6 is an example of a flowchart of each maintenance cost allocation calculation that minimizes the probability of equipment outage when the total maintenance cost is fixed to an arbitrary value.   FIG. 7 is an example of a flowchart of each maintenance cost allocation calculation for minimizing the sum of the expected value of damage costs due to equipment shutdown and the total maintenance cost for each failure mode.

Claims (5)

  From the probability of occurrence of each basic event in fault tree analysis, which is a reliability analysis technique that extracts the factors that cause the occurrence of problem events and shows the relationship between computers with input devices, storage devices, arithmetic devices, and output devices In a recording medium recording a program for calculating the probability of occurrence of a top event, the occurrence of a problem event caused by several factors input from an input device is defined as a top event, and one or more causes of the occurrence of the problem event The fault tree diagram configuration data representing the relationship structure with the above factors as basic events, and the correlation between the occurrence probability of each factor and the input amount of resources of each countermeasure implemented to avoid the occurrence of each factor The data of the function (hereinafter referred to as “used resource function”) is stored in a storage device, and the data in the fault tree analysis is stored by the arithmetic device. The input amount of the resource according to each of the measures that satisfies the condition that the occurrence probability of the problem event calculated using a program for determining the occurrence probability of the event is not changed from a given value is set by the used resource function A resource allocation method comprising: searching for a combination that minimizes the total amount of the resources input for each of the measures, and causing the output device to display the combination that is a search result.   From the probability of occurrence of each basic event in fault tree analysis, which is a reliability analysis technique that extracts the factors that cause the occurrence of problem events and shows the relationship between computers with input devices, storage devices, arithmetic devices, and output devices In a recording medium in which a program for calculating the occurrence probability of a top event is recorded, the occurrence of the problem event caused by several factors input from the input device is regarded as the top event, and the one or more causes of the occurrence of the problem event Configuration data of a fault tree diagram with factors as basic events, data of resource functions used indicating the correlation between the probability of occurrence of each factor and the amount of resource input for each measure implemented to avoid the occurrence of each factor Is stored in the storage device, and the condition that the total input amount of the resource for each countermeasure is not changed from a given value by the arithmetic device is satisfied. The occurrence probability of each factor is set by the resource function used, and among them, the occurrence probability of the problem event calculated by the arithmetic unit using the program for obtaining the occurrence probability of the top event in the fault tree analysis is minimized. A resource allocation method comprising: searching for a combination to be displayed, and displaying the combination as a search result on an output device.   From the probability of occurrence of each basic event in fault tree analysis, which is a reliability analysis technique that extracts the factors that cause the occurrence of problem events and shows the relationship between computers with input devices, storage devices, arithmetic devices, and output devices In a recording medium in which a program for calculating the occurrence probability of a top event is recorded, the occurrence of the problem event caused by several factors input from the input device is regarded as the top event, and the one or more causes of the occurrence of the problem event Fault tree diagram configuration data with factors as basic events, resource usage function indicating the correlation between the probability of occurrence of each factor and the amount of resources required for each measure implemented to avoid the occurrence of each factor The data is stored in a storage device, and is expressed in units of the value of the probability of occurrence of the problem event and the amount of the resource arbitrarily set by the arithmetic device. A program for obtaining an occurrence probability of a top event in the fault tree analysis with respect to an expected value of damage indicated by a product of damage values caused by occurrence of the problem event and an arbitrarily set value of the occurrence probability of the problem event Based on the probability of occurrence of each factor such that the value of the top event calculated by the arithmetic unit using the same is used, the total amount of the resource input for each of the measures set by the used resource function A combination of the input amounts of the resources for each of the measures such that the sum of the values with the minimum value becomes the minimum when the value of the probability of occurrence of the previously set problem event is changed within a certain range. A resource allocation method comprising: searching and displaying the combination as a search result on an output device.   A recording medium, wherein the resource allocation method according to claim 1 is converted into a program and the program is recorded.   5. A recording medium storing the program of claim 4 and a computer executing the program, wherein the occurrence probability of one or more factors causing the occurrence of a problem event caused by several factors and the occurrence of each factor are avoided. Instead of inputting from the input device data on the resource function used indicating the correlation with the resource input of each measure to be implemented, the occurrence timing, the number of occurrences of each factor input by the input device and the A recording medium storing a program, wherein data of the used resource function is calculated based on data of a resource input amount and a resource input time.

Images