Stocked spare parts number calculation device, stocked spare parts number calculation method, and program

Abstract

To provide a technique for evaluating the number of stocked spare parts in consideration of probability distribution characteristics of the number of failures of a component.SOLUTION: A stocked spare parts number calculation device includes: means for calculating, regarding a component constituting an evaluation target system, the number of failures of the component during an evaluation target period and a failure number-based failure probability to indicate a probability that the number of failures will occur on the basis of data about time-series failure / maintenance results of the component; and means for calculating the number of stocked spare parts of the component for achieving a target value of a probability that shortage of the component does not occur when a failure occurs, on the basis of the failure number-based failure probability.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a spare parts inventory quantity calculation device, a spare parts inventory quantity calculation method, and a program. [Background technology]

[0002] Patent Document 1 discloses a method for evaluating the optimal number of spare parts for system components. This evaluation method calculates the average system downtime when a part is out of stock (average downtime due to a part being out of stock) based on the part's mean time between failures (MTBF), the work downtime required to replace the part when it fails, the time required to procure a new part, and the upper and lower limits of downtime due to a part being out of stock, and evaluates the optimal number of spare parts based on the number of spare parts and the average downtime due to a part being out of stock. This evaluation method does not take into account the probability distribution characteristics of the number of times a part fails. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 5999460 Summary of the Invention [Problem to be solved by the invention]

[0004] A method is provided for evaluating the appropriate number of spare parts to be held, taking into consideration the probability distribution characteristics of the number of failures of parts installed in a system.

[0005] The present disclosure provides a spare parts inventory quantity calculation device, a spare parts inventory quantity calculation method, and a program that can solve the above-mentioned problems. [Means for solving the problem]

[0006] The spare parts stock quantity calculation device according to the present disclosure includes: means for calculating, for parts constituting a system to be evaluated, the number of times a part has failed during an evaluation period based on performance data on time-series failures and maintenance of the part, a failure probability for each failure count that indicates the probability of a failure occurring at that number of failures; and means for calculating, based on the failure probability for each failure count, a spare parts stock quantity for the part that can achieve a target value for the probability that the part will not be out of stock in the event of a failure.

[0007] The method for calculating the quantity of spare parts held according to the present disclosure includes the steps of: calculating, for parts constituting a system to be evaluated, the number of times a part has failed during a period to be evaluated and the probability of that failure occurring for each number of failures, based on historical data on time-series failures and maintenance of the part; and calculating, based on the failure probability for each number of failures, the quantity of spare parts held for the part that can achieve a target value for the probability that the part will not be out of stock in the event of a failure.

[0008] The program according to the present disclosure causes a computer to execute the steps of: calculating, for components constituting a system to be evaluated, the number of component failures during a period to be evaluated based on performance data on time-series failures and maintenance of the components, the failure probability for each number of failures indicating the probability of a failure occurring at that number of failures; and calculating, based on the failure probability for each number of failures, the quantity of spare parts to be held for the components that will achieve a target value for the probability that the components will not be out of stock in the event of a failure. [Effects of the Invention]

[0009] According to the spare parts stock quantity calculation device, spare parts stock quantity calculation method, and program disclosed herein, it is possible to evaluate the appropriate number of spare parts to be stocked, taking into account the probability distribution characteristics of the number of times a part fails. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a block diagram illustrating an example of a spare parts inventory quantity calculation device according to an embodiment. [Figure 2] FIG. 10 is a diagram illustrating an example of importance of a part according to the embodiment. [Figure 3] 10 is a flowchart illustrating an example of a spare parts inventory quantity calculation process according to the embodiment. [Figure 4] FIG. 10 is a diagram showing an example of failure / maintenance data according to the embodiment. [Figure 5A] FIG. 1 is a diagram showing an example of a Weibull probability paper according to an embodiment. [Figure 5B] FIG. 10 is a diagram illustrating an example of a regular probability paper according to the embodiment. [Figure 5C] FIG. 10 is a diagram illustrating an example of a log-normal probability paper according to the embodiment. [Figure 6] 10A and 10B are diagrams illustrating the time series of the number of failures and the failure probability according to the embodiment. [Figure 7] 10A and 10B are diagrams illustrating a method for calculating a failure probability according to an embodiment. [Figure 8] FIG. 2 illustrates an example of a hardware configuration of a spare parts inventory quantity calculation device according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] <Embodiment> Hereinafter, the method of calculating the inventory quantity of spare parts according to the present disclosure will be described with reference to the drawings. (composition) FIG. 1 is a block diagram illustrating an example of a spare parts inventory quantity calculation device according to an embodiment. The spare part inventory quantity calculation device 10 includes a data acquisition unit 11, a calculation condition setting unit 12, a failure pattern discrimination unit 13, a failure probability prediction unit 14, an inventory quantity calculation unit 15, an output unit 16, and a memory unit 17.

[0012] The data acquisition unit 11 acquires data necessary for calculating the quantity of spare parts in stock. For example, the data acquisition unit 11 acquires failure and maintenance data for the devices and parts (hereinafter referred to as parts) that make up the system to be evaluated, the number of installations (the number of parts installed or attached to the system), the status of spare parts in stock, delivery dates, and information on replacement and obsolete parts. The failure and maintenance data includes the failure time and number of failures for each part, as well as the replacement / maintenance of parts. Obsolete and obsolete parts are parts whose production has been discontinued or is scheduled to be discontinued.

[0013] The calculation condition setting unit 12 sets calculation conditions for the available quantity of spare parts, such as the target stockout probability of spare parts, the importance of spare parts, and the evaluation period. Here, the probability index set to keep the probability of stockout occurrence below a target value in order to prevent stockouts when a part installed in the system fails is called the target stockout probability. For example, if the target stockout probability of part A is 10%, the number of parts in stock is 2, and the predicted probability of the number of failures being 2 or more is 5% in a specified evaluation period, the predicted stockout probability satisfies the target stockout probability. The target stockout probability is set based on the availability targets of the entire system, subsystems, and components, the target value of MTBF (Mean Time Between Failures), etc. For example, if the probability of damage to nuclear reactor equipment is 10 -6 A probability value such as [- / reactor·year] is used. The target out-of-stock probability may be set by a knowledgeable user. Alternatively, the MTBF and MTTR (Mean Time To Restoration) for each part when that part fails may be recorded in the storage unit 17, and when an operator inputs the availability target for the system, subsystem, or component, the calculation condition setting unit 12 may use a predetermined calculation formula and information such as the MTBF to calculate and set the target out-of-stock probability that must be reached in order for the availability to achieve the target.

[0014] Criticality refers to the part's importance in achieving the target stockout probability and is ultimately taken into consideration when determining the number of spare parts to stock. For example, for highly critical parts, the stocking quantity may be determined to be slightly higher than the quantity required to achieve the target stockout probability. For less critical parts, the minimum necessary quantity may be determined, taking into account cost and delivery time. Figure 2 shows an example of criticality. Highly critical parts include parts that directly affect system availability, parts with long delivery times, and obsolete or modified parts that have a significant impact on availability and economic efficiency (for example, replacing an old model of a part with a new one may require the replacement of surrounding connected devices, which may have a significant economic impact, including redesign). Parts of medium importance include parts with unstable market supply and obsolete or modified parts that have a moderate impact on availability and economic efficiency. Parts of low importance include parts with seasonal demand peaks (e.g., difficult to obtain in August), parts scheduled for large-scale maintenance or overhaul, and parts required for disaster or emergency preparedness. The importance may be set by a knowledgeable user. Alternatively, an operator may set evaluation values ​​regarding, for example, whether the part directly affects system availability, whether the part has a long delivery time, whether the part has unstable supply, or whether the part has a demand peak, for example, by setting "1" if each item applies and "0" if each item does not apply, and the calculation condition setting unit 12 may set the importance of the part based on these evaluation values ​​and a predetermined calculation formula. In a simple example, the calculation condition setting unit 12 may calculate the sum of the evaluation values ​​for items related to "high" importance, the sum of the evaluation values ​​for items related to "medium" importance, and the sum of the evaluation values ​​for items related to "low" importance, and set the importance with the largest sum to the part.

[0015] The failure pattern discrimination unit 13 plots the failure and maintenance data of each part on a probability paper to determine whether the part has a tendency to fail due to aging or whether it fails due to random failure. The failure pattern discrimination unit 13 plots the failure and maintenance data for each part on a Weibull probability paper, and if the slope of the approximation line of the plotted points is greater than 1, classifies the part as an aging deterioration type. If the slope is around 1 (within a predetermined range based on 1, the shape parameter of the Weibull distribution corresponds to a random failure when it is equal to 1, an initial failure when it is less than 1, and aging deterioration when it is greater than 1, for example, in the range of 0.7 to 1.3), the failure pattern discrimination unit 13 classifies the part as a random failure type. Furthermore, if the failure tendency of a spare part is classified as an aging deterioration type, the failure pattern discrimination unit 13 evaluates whether the distribution of the probability of the part failing due to aging conforms to a normal distribution, a log-normal distribution, or a Weibull distribution.

[0016] The failure probability prediction unit 14 estimates the failure probability for each number of failures during the evaluation period based on a Poisson distribution or a component lifespan distribution model. If the component is a random failure type, the failure probability prediction unit 14 assumes that the component failure probability follows a Poisson distribution and calculates the failure probability for each number of failures during the evaluation period by calculating the probability mass function of the Poisson distribution. If the component failure pattern is an aging degradation type, the failure probability prediction unit 14 assumes that the component failure probability follows the component lifespan distribution (either a normal distribution, a log-normal distribution, or a Weibull distribution) evaluated as suitable by the failure pattern discrimination unit 13, and calculates the component failure probability and the probability mass function of the Poisson binomial distribution to calculate the probability of component failure due to aging degradation for each number of failures. From the calculated failure probabilities for each number of failures, the failure probability prediction unit 14 calculates an upper limit for the number of failures corresponding to the target stockout probability during the evaluation period. For example, if the probability of a part failing two or more times during the target period is 10%, and the probability of three or more failures occurring is 1%, and the target stockout probability is 1%, then in order to achieve the target stockout probability of 1%, it will be necessary to deal with three or more failures that occur with a 1% probability, so the upper limit on the number of failures will be three.

[0017] The stock quantity calculation unit 15 calculates the required stock quantity from the upper limit of the number of failures calculated by the failure probability prediction unit 14. In the above example, the stock quantity of spare parts required to satisfy the target stockout probability of 10% is two. In addition, the stock quantity calculation unit 15 may calculate the number of spare parts to be newly procured by taking into consideration the stock quantity (inventory amount) of spare parts already in stock, the delivery date, the importance of the spare parts, etc. For example, if the importance of the part in the above example is "high" and the current inventory amount is 0, the stock quantity calculation unit 15 calculates the number of spare parts to be procured to be 2. For example, if the importance of the part in the above example is "low", the current inventory amount is 1, and the delivery date of the part is short, the stock quantity calculation unit 15 may calculate the number of spare parts to be procured to be 0. The stock quantity calculation unit 15 may calculate the order quantity using a predetermined formula for calculating the quantity of spare parts to be procured, using variables such as the necessary stock quantity calculated from the upper limit of the number of failures that satisfies the target stockout probability, the importance, the inventory amount, and the standard delivery time, or may output the number of failures, the importance, the standard delivery time, and the inventory amount of parts that satisfy the target stockout probability to a display device or the like via the output unit 16, and prompt the user to decide on the final stock quantity and the order quantity required for that.

[0018] The output unit 16 outputs the stocked quantity of spare parts calculated by the stocked quantity calculation unit 15 to a display device, an electronic file, or the like. The storage unit 17 stores various data acquired by the data acquisition unit 11, the inventory amount and standard delivery time of each part, the processing results by each of the functional units 12 to 15, a common dictionary of parts, etc. The common dictionary of parts is a dictionary for dealing with variations in the notation of part names, model numbers, etc. (for example, the same part may be called by name 1 in factory A and name 2 in factory B), and converting them into a unified name.

[0019] (operation) Next, the operation of the spare parts inventory quantity calculation device 10 will be described with reference to FIG. FIG. 3 is a flowchart illustrating an example of a spare parts inventory quantity calculation process according to an embodiment. The calculation condition setting unit 12 sets the importance of parts and the target out-of-stock probability (step S101). For example, when a user inputs the parts to be evaluated, the importance of each part, the target out-of-stock probability, and the target evaluation period into the spare parts inventory quantity calculation device 10, the calculation condition setting unit 12 acquires the input information and sets the system to be evaluated, the parts to be evaluated among the parts installed in the system, the importance of each part, the target out-of-stock probability, and the target evaluation period. The number of systems and parts to be evaluated is not limited to one. For example, if there are other sites or plants where similar systems are installed, the systems at those sites may be set. Next, the data acquisition unit 11 acquires failure and maintenance data, etc. (step S102). For example, a user inputs failure and maintenance data for the parts to be evaluated into the spare parts inventory quantity calculation device 10, and the data acquisition unit 11 acquires the failure and maintenance data and stores it in the memory unit 17. An example of the failure and maintenance data is shown in FIG. 4. Figure 4 shows the change in the number of failures of a component part of a certain system. The vertical axis of the graph in Figure 4 represents the number of failures of the part, and the horizontal axis represents time. The graph in Figure 4 shows that a part is installed and used at time T0, and the number of failures increases over time. When the part is replaced with a new one during maintenance at time T1, no failures occur for a while, but then the number of failures increases again as time passes.

[0020] Next, the data acquisition unit 11 acquires failure and maintenance data of other systems (step S103). Step S103 is not essential, but for example, if there is little failure and maintenance data on the evaluation target part in the system to be evaluated, the failure and maintenance data of other systems related to the evaluation target part can be used. When acquiring the failure and maintenance data of other systems, the data acquisition unit 11 corrects the names of the parts in the evaluation target system using a common dictionary. The data acquisition unit 11 saves the acquired failure and maintenance data of other systems in the memory unit 17.

[0021] Next, the failure pattern determination unit 13 develops the failure and maintenance data in a time series and determines whether the part being evaluated is of the random failure type or the aging deterioration type by plotting it on a Weibull probability paper (step S104). An example of a Weibull probability paper is shown in FIG. 5A. In the graph of FIG. 5A, the horizontal axis x is the logarithm of time t (ln(t)), and the vertical axis y is the failure rate according to time t, F(t), as follows: lnln(1-F(t)) -1 The failure pattern determination unit 13 calculates the unreliability F(t) over time from the failure and maintenance data acquired in steps S102 and S103, for example, by dividing the number of failures that occurred during the operating time (mounting time) of the part by the operating time. The failure pattern determination unit 13 calculates the unreliability F(t) over time from the failure and maintenance data acquired in steps S102 and S103, for example, by dividing the number of failures that occurred during the operating time (mounting time) of the part by the operating time. -1 is calculated and plotted on a Weibull probability paper. The failure pattern discrimination unit 13 calculates an approximate curve (linear equation: y = ax + b) connecting each of the plotted points and finds its slope a. The failure pattern discrimination unit 13 compares the slope a with 1, and if the slope a is greater than 1, it determines that the part's tendency to fail is of the aging deterioration type, and if the slope a is around 1, it determines that the part will fail due to random failure. Here, it is assumed that the system being evaluated undergoes daily maintenance and periodic inspections, and that aging deterioration is basically detected during daily inspections and repaired or replaced with a new part during maintenance. Therefore, the majority of parts are classified as the random failure type. However, there are some parts that are determined to be of the aging deterioration type.

[0022] If it is determined to be an aging deterioration type (step S105; No), the failure pattern determination unit 13 further specifies a life analysis model that is suitable for the part (step S106). The failure pattern determination unit 13 plots the relationship between the operating time (installation time) of the part based on the failure and maintenance data and the cumulative failure probability corresponding to the operating time on a log-normal probability paper and a normal probability paper, in addition to the Weibull probability paper created in step S104. Then, it estimates a linear line connecting the plotted points on the log-normal probability paper and the normal probability paper. Fig. 5B shows an example of the normal probability paper, and Fig. 5C shows an example of the log-normal probability paper. The failure pattern determination unit 13 calculates a coefficient of determination R 2 Calculate for each probability paper and R 2 In the case of Figures 5A to 5C, the R of the Weibull probability paper is selected. 2 = 0.9726 is closest to 1, the failure pattern discrimination unit 13 selects the Weibull probability paper. The failure pattern discrimination unit 13 considers that the failure probability due to aging deterioration of the part in question follows the Weibull distribution (specifies the life analysis model as the Weibull distribution). Similarly, the failure pattern discrimination unit 13 selects the R 2 If is closest to 1, the failure probability due to aging of the part follows a log-normal distribution, and the R 2 is closest to 1, the failure probability due to aging deterioration of the part is considered to follow a normal distribution. The failure pattern determination unit 13 outputs the identified type of life analysis model to the failure probability prediction unit 14.

[0023] If it is determined to be a random failure type (step S105; Yes), the failure probability prediction unit 14 predicts the failure probability for each number of failures using Poisson analysis (step S108). The failure probability prediction unit 14 calculates the failure probability for each number of failures using the following formula (1), assuming that component failures follow a Poisson distribution.

[0024]

number

[0025] Equation (1) is the probability mass function of the Poisson distribution. In equation (1), m represents the product of the target period T, the failure rate λ (the inverse of MTBF) when a part randomly fails, and the number n of target parts installed, and x represents the number of part failures. λ is set as the random failure rate analyzed by the user. The failure probability prediction unit 14 substitutes an arbitrary number for x and calculates the failure probability for each number of part failures using equation (1). For example, substituting 1 for x and calculating equation (1) calculates the probability that one part will fail during the target period, and substituting 2 for x calculates the probability that two failures will occur. As a result of such calculations, a graph such as that shown in FIG. 6 is obtained. The vertical axis of FIG. 6 represents the number of parts that will fail, and the horizontal axis represents time. For example, in the fourth year, there is a 10% probability that two or more failures will occur, and a 1% probability that three or more failures will occur. From this result, for example, if the part being evaluated is an important part, then preparing three spare parts will allow for dealing with a failure that occurs 99 times out of 100 (99% probability). Also, if it is considered sufficient to be able to deal with two failures that occur 90 times out of 100 (90% probability) and the part is readily available even if it does fail, then preparing two spare parts will suffice. The failure probability prediction unit 14 records the calculated occurrence probabilities for each number of failures in the memory unit 17 and outputs them to the inventory quantity calculation unit 15.

[0026] Furthermore, if the part is determined to be of the age-related deterioration type (step S105; No), the failure probability prediction unit 14 assumes that the part will fail according to the age-related deterioration type life distribution (normal distribution, log-normal distribution, Weibull distribution, etc.) identified in step S106, and calculates the failure probability for each number of failures at the end of the evaluation period using the following formula (2) (step S107).

[0027]

number

[0028] In the case of random failures, if the components are identical, a single failure probability m is assigned to all components. However, in the case of aging-related failures, because the operating time and other factors vary for each component, a failure probability pi at a certain time T is assigned to each component being evaluated (number of components installed: n). The number of failures is then given as a probability distribution (Poisson binomial distribution) of the sum of Bernoulli trials with different probability values. Equation (2) above is the probability mass function of the Poisson binomial distribution. Equation (2) can be used to calculate the probability of k components failing during a target period (e.g., the time until the next scheduled inspection or the time until the scheduled replacement of a component). In equation (2), pi and pj represent the probability of a component failing during the target period, and k represents the number of component failures. Figure 7 illustrates how pi is calculated. Figure 7 shows the failure probability when a component's failure probability is assumed to follow a Weibull distribution. The vertical axis of the graph in Figure 7 represents failure probability, the horizontal axis represents time, and line 71 represents the failure probability at each time. When the target period for evaluation is the time up to a certain time T, the failure probability prediction unit 14 calculates the area of ​​the region enclosed by the x-axis and line 71, i.e., the integrated value of the failure probability from the start of the target period to time T. The calculated integrated value is the failure probability pi. Furthermore, Fk in equation (2) is a set of combinations of components that will fail when the number of failures is k out of the number of installed components n. i is a subscript representing the k components that will fail, j is a subscript representing the nk components that will not fail, Ac is a corrected set of A, and Π represents a product. pi and pj are failure probabilities for the time up to time T calculated from the life distribution identified in step S107, as described with reference to FIG. 7. The failure probability prediction unit 14 substitutes an arbitrary number for the number of failures k and calculates the failure probability for each component by the number of failures using equation (2). For example, substituting 1 for k and calculating equation (2) calculates the probability that one component will fail during the target period, and substituting 2 for k and calculating the probability that two components will fail. Once the failure probability for each failure count has been calculated, a graph such as that shown in Figure 6 can be obtained, just as in the case of random failures. The user can determine the quantity of parts to stock based on the graph in Figure 6 and the importance and delivery date of each part, risk management policy, etc. The failure probability prediction unit 14 records the calculated occurrence probability for each failure count in the memory unit 17 and outputs it to the stock quantity calculation unit 15.

[0029] Next, the inventory quantity calculation unit 15 acquires the spare part inventory status and standard delivery time of the evaluation target part (step S109). The inventory quantity calculation unit 15 acquires the spare part inventory amount and standard delivery time through the data acquisition unit 11, and corrects variations in the spelling of the part name, etc., using a common dictionary. Next, the data acquisition unit 11 acquires the latest delivery time information and revision / elimination information (step S110). The latest revision / elimination information can be acquired by the following methods: (1) Scrape web information to acquire the part revision / elimination status; (2) Use the part revision / elimination information notified by the supplier; or (3) Interview the supplier about the planned revision / elimination and input the information obtained into the spare part inventory quantity calculation device 10. The revision / elimination information includes information such as the name of the part to be revised or eliminated, the planned revision / elimination date, the successor part, and the start date of its provision. The data acquisition unit 11 stores the acquired revision / elimination information in the memory unit 17 and outputs it to the inventory quantity calculation unit 15. Furthermore, the output unit 16 may display the revision / discontinuation information acquired by the data acquisition unit 11, for example, for each subsystem. This allows the user who is considering the number of items in stock to be notified of the revision / discontinuation status of parts.

[0030] Furthermore, the failure probability prediction unit 14 performs the same processing as steps S104 to S108 for the obsolete or obsolete product to calculate the failure probability for each number of failures while the obsolete or obsolete product continues to be used. For example, if the operation of the target system will end in five years, the failure probability for each number of failures of the obsolete or obsolete product over the five years is calculated. If the current stock quantity of obsolete or obsolete products is 0 and the target stockout probability is 10%, and it is calculated that the obsolete or obsolete product will fail two or more times within five years with a 10% probability, it is possible to order two obsolete or obsolete products from the supplier before production is discontinued.

[0031] Finally, the stock quantity calculation unit 15 calculates the stock quantity of the evaluation target part (step S111). For example, since the probability of occurrence of failure for each number of times of failure of a part can be calculated by the formula (1) or (2), the stock quantity calculation unit 15 calculates the stock quantity of the part that should satisfy the target stockout probability by the following formula (3).

[0032]

number

[0033] Here, i is the number of failures, P(i) is the probability of occurrence of the number of failures i during the target period calculated by Equation (1) or (2), and Pa is the target out-of-stock probability. The inventory quantity calculation unit 15 calculates the upper limit Xa of the number of failures that can satisfy the target out-of-stock probability Pa using Equation (3). If the inventory contains items equal to the upper limit Xa, the target out-of-stock probability Pa can be achieved. In addition, in the case of aging deterioration, if it is predicted that the target out-of-stock probability will not be reached at the time of the next scheduled replacement, it may be decided to bring forward the replacement schedule. In addition, if the part is a replacement or obsolescence product, the target period may be set to the scheduled time of replacement or obsolescence and the above processing may be performed. In addition, the inventory quantity calculation unit 15 may adjust Xa according to the importance of the part and the standard delivery date. For example, if the importance is high, the inventory quantity calculation unit 15 may set the target out-of-stock probability low or may calculate the inventory quantity by adding a predetermined number (e.g., 1) to Xa. Alternatively, if the part is of low importance and the standard delivery time is equal to or less than the threshold, the inventory quantity calculation unit 15 may calculate the inventory quantity by subtracting a predetermined number (for example, 1) from Xa. Next, the inventory quantity calculation unit 15 outputs Xa (inventory quantity) calculated by formula (3), the importance, and the standard delivery time to a display device or the like via the output unit 16. The inventory quantity calculation unit 15 may also output the value obtained by subtracting the inventory quantity of the currently-inventory part from Xa as the shortage in inventory quantity. The user refers to the output inventory quantity, importance, and standard delivery time to arrange for the part.

[0034] The process in the flowchart in Figure 3 is merely an example. For example, a target stockout probability may be set for each level of importance. Furthermore, the calculation of the parts inventory quantity based on the failure and maintenance data of other systems may be performed separately from the process of the system being evaluated, for example, before step S111. In step S111, the inventory quantities of the system being evaluated and those of other systems may be output in a format that allows comparison. This allows the user to determine the inventory quantity of parts while referring to examples from other systems. The process in Figure 3 may also be performed periodically. This allows the latest failure and maintenance data, spare part status, and information on replacement and obsolescence products and delivery dates to be collected and the evaluation to be revised based on this information, in response to constantly changing trends in the status of replacement and obsolescence, delivery dates, and failure rates. Failures of replacement and obsolescence products or parts with long delivery times can cause long-term system downtime, which can reduce availability and create a bottleneck in stable operation. However, the probability of long-term downtime can be reduced by performing the process in Figure 3 whenever possible to ensure that spare parts are available. Furthermore, when the evaluation period is long, the inventory quantity of spare parts tends to be calculated. You can repeat short-term evaluations and plan inventory management and spare part purchases depending on the availability of spare parts at any given time.

[0035] (effect) As described above, the spare parts inventory quantity calculation method of this embodiment allows for the evaluation of spare parts inventory quantities by taking into account the probability distribution characteristics of the number of part failures. While typical inventory quantity evaluations often only consider part failures caused by random failures, this embodiment allows for the analysis to distinguish between random failures and aging degradation, resulting in a more accurate calculation of part inventory quantities. Furthermore, when analyzing data from other systems, differences in equipment names and model numbers often prevent automatic data utilization. However, by standardizing names and model numbers using a common dictionary, it becomes easier to utilize failure and maintenance data from other systems, making it easier to calculate inventory quantities using a large amount of data and to compare the failure and part inventory status of both systems.

[0036] FIG. 8 is a diagram showing an example of the hardware configuration of the spare parts stock quantity calculation device 10 according to the embodiment. The computer 900 includes a CPU 901, a main storage device 902, an auxiliary storage device 903, an input / output interface 904, and a communication interface 905. The above-described spare parts stock quantity calculation device 10 is implemented in the computer 900. The above-described functions are stored in the auxiliary storage device 903 in the form of a program. The CPU 901 reads the program from the auxiliary storage device 903, loads it into the main storage device 902, and executes the above-described processing in accordance with the program. The CPU 901 also allocates a storage area in the main storage device 902 in accordance with the program. The CPU 901 also allocates a storage area in the auxiliary storage device 903 for storing data being processed in accordance with the program.

[0037] A program for implementing all or part of the functions of the spare parts inventory quantity calculation device 10 may be recorded on a computer-readable recording medium, and the program may be loaded into a computer system and executed to perform processing by each functional unit. The term "computer system" herein includes hardware such as an OS and peripheral devices. If a WWW system is used, the term "computer system" also includes a homepage provision environment (or display environment). The term "computer-readable recording medium" refers to portable media such as CDs, DVDs, and USBs, as well as storage devices such as hard disks built into the computer system. If the program is distributed to the computer 900 via a communication line, the computer 900 may load the program into the main storage device 902 and execute the processing described above. The program may be for implementing part of the functions described above, or may be capable of implementing the functions described above in combination with a program already stored in the computer system.

[0038] As described above, several embodiments according to the present disclosure have been described, but all of these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope of the invention and its equivalents as defined in the claims, as well as in the scope and spirit of the invention.

[0039] <Additional Notes> The spare parts stock quantity calculation device, the spare parts stock quantity calculation method, and the program described in the embodiments can be understood, for example, as follows.

[0040] (1) A spare parts inventory calculation device according to a first aspect includes: means for calculating, for a part constituting a system to be evaluated, a failure probability by number of failures, which indicates the number of times a part has failed during an evaluation period and the probability of a failure occurring at that number of failures, based on actual data on time-series failures and maintenance of the part; and means for calculating, based on the failure probability by number of failures, a spare parts inventory quantity for the part that can achieve a target value for the probability that the part will not be out of stock in the event of a failure (in other words, that can suppress the probability that the part will be out of stock in the event of a failure to (1 - the target value) or less). This makes it possible to evaluate the number of spare parts to be held, taking into account the probability distribution characteristics of the number of times a part fails.

[0041] (2) A second aspect of the spare parts stock quantity calculation device is the spare parts stock quantity calculation device of (1), further comprising a means for determining whether the part will fail due to an accidental failure or due to deterioration over time, and the means for calculating the failure probability by number of failures calculates the failure probability by number of failures for the part that will fail due to an accidental failure, assuming that the failure of the part occurs according to a Poisson distribution. This makes it possible to distinguish whether the failure mode of a part is a random failure or aging deterioration, and then calculate the failure probability of the part.

[0042] (3) A third aspect of the spare parts holding quantity calculation device is the spare parts holding quantity calculation device of (2), wherein the means for calculating the failure probability by number of failures determines, for the parts that fail due to aging, whether the relationship between the operating time of the part and the number of failures follows a lifespan distribution from among a Weibull distribution, a normal distribution, and a log-normal distribution based on a Weibull probability paper, a normal distribution probability paper, and a log-normal distribution probability paper, and calculates the failure probability by number of failures based on the determined lifespan distribution. This makes it possible to calculate the failure probability of a part that fails due to aging after determining the type of distribution that the failure probability follows.

[0043] (4) A spare parts inventory quantity calculation device according to a fourth aspect is the spare parts inventory quantity calculation device of (3), wherein the determining means determines that the part will fail due to deterioration over time if the slope of the approximation curve of the plotted points is greater than 1 when the relationship between the operating time of the part and the number of failures is plotted on a Weibull probability paper, and determines that the part will fail due to random failure if the slope is less than 1. This makes it possible to determine whether the failure mode of the part is a random failure type or an aging deterioration type.

[0044] (5) A fifth aspect of the spare part stock quantity calculation device is a spare part stock quantity calculation device according to any one of (1) to (4), further comprising a means for acquiring revision / discontinuation information on the part, wherein the means for calculating the failure probability by number of failures calculates the failure probability by number of failures until the time when production of the part will be discontinued when the revision / discontinuation information indicates that production of the part will be discontinued, and the means for calculating the spare part stock quantity calculates the spare part stock quantity based on the failure probability by number of failures until production of the part will be discontinued. This allows the calculation of a spare parts holding plan taking into account replacement or obsolete parts.

[0045] (6) A spare parts holding quantity calculation device according to a sixth aspect is a spare parts holding quantity calculation device according to any one of (1) to (5), further comprising a means for setting the importance of the part, and the means for calculating the spare parts holding quantity adjusts the spare parts holding quantity in accordance with the importance. This allows the number of spare parts held to be adjusted according to the importance of the parts and the delivery date.

[0046] (7) A seventh aspect of the spare part holding quantity calculation device is a spare part holding quantity calculation device according to any one of (1) to (6), further comprising a means for acquiring performance data of the part from another system other than the system to be evaluated, wherein the means for calculating the failure probability by number of failures calculates the failure probability by number of failures based on the performance data of the part in the other system, and the means for calculating the spare part holding quantity calculates the spare part holding quantity based on the failure probability by number of failures. This makes it possible to calculate the number of spare parts to be held for the part being evaluated by utilizing failure history data for parts in other plants, etc.

[0047] (8) The spare parts holding quantity calculation device according to the eighth aspect is the spare parts holding quantity calculation device of (7), wherein the means for acquiring the performance data of the parts from other systems other than the system to be evaluated has a common dictionary for converting the names of the parts in the other systems into the names of the parts in the system to be evaluated. This makes it possible to deal with variations in notation such as part names.

[0048] (9) A method for calculating spare part stock quantity according to a ninth aspect is a method for calculating spare part stock quantity executed by a computer, and includes the steps of: calculating, for parts constituting a system to be evaluated, the number of times a part has failed during an evaluation period and the probability of a failure occurring for each failure count, based on actual data on time-series failures and maintenance of the parts; and calculating, based on the failure probability for each failure count, the quantity of spare parts to be held for the parts that can achieve a target value for the probability that the part will not be out of stock in the event of a failure (in other words, can suppress the probability that the part will be out of stock in the event of a failure to be equal to or less than (1 - the target value)).

[0049] (10) A program according to a tenth aspect causes a computer to execute the following steps: calculating, for components constituting a system to be evaluated, the number of component failures during a period to be evaluated based on actual data on the time series of failures and maintenance of the components, the failure probability for each failure count, which indicates the probability of a failure occurring at that number of failures; and calculating, based on the failure probability for each failure count, the number of spare parts to be held for the components that can achieve a target value for the probability that the components will not be out of stock in the event of a failure (in other words, the probability that the components will be out of stock in the event of a failure can be kept to (1 - the target value) or less). [Explanation of symbols]

[0050] 10. Spare parts inventory calculation device 11. Data acquisition section 12. Calculation condition setting section 13. Failure pattern discrimination unit 14. Failure probability prediction section 15...Holding quantity calculation department 16 Output section 17...Storage section 900···Computer 901 CPU 902...Main memory 903...Auxiliary storage device 904 Input / Output Interface 905···Communication Interface

Claims

1. means for calculating the number of component failures during an evaluation period and the probability of occurrence of failures for each failure count, based on the time-series failure and maintenance performance data for the component constituting the evaluation target system; a means for calculating a spare part stock quantity of the part that can achieve a target value of the probability that the part will not be out of stock when a failure occurs, based on the failure probability for each failure count; A spare parts inventory quantity calculation device having the same.

2. means for determining whether the component has failed due to random failure or due to aging; and the means for calculating the failure probability by the number of failures calculates the failure probability by the number of failures for the component that fails due to a random failure, assuming that the failure of the component occurs according to a Poisson distribution; The spare parts inventory quantity calculation device according to claim 1.

3. The means for calculating the failure probability by number of failures determines, for the part that fails due to aging deterioration, whether the relationship between the operating time of the part and the number of failures follows a lifespan distribution of Weibull distribution, normal distribution, or log-normal distribution, based on a Weibull probability paper, normal distribution probability paper, or log-normal distribution probability paper, and calculates the failure probability by number of failures based on the determined lifespan distribution. The spare parts inventory quantity calculation device according to claim 2.

4. When the relationship between the operating time of the part and the number of failures is plotted on a Weibull probability paper, if the slope of the approximation curve of the plotted points is greater than 1, the determining means determines that the part will fail due to aging deterioration, and if the slope is within a predetermined range based on 1, determines that the part will fail due to random failure. The spare parts inventory quantity calculation device according to claim 2.

5. The system further includes a means for acquiring information on the revision or abolition of the part, The means for calculating the failure probability by the number of failures calculates, when the improvement / abolition information indicates the stop of production of the part, the failure probability by the number of failures until the time when production will be stopped, the means for calculating the spare part stock quantity calculates the spare part stock quantity based on the failure probability for each of the number of failures until production of the part is stopped; 3. The spare parts inventory quantity calculation device according to claim 1 or 2.

6. means for setting the importance of the component; the means for calculating the spare part stock quantity adjusts the spare part stock quantity in accordance with the importance.

3. The spare parts inventory quantity calculation device according to claim 1 or 2.

7. The system further includes means for acquiring the performance data of the part from a system other than the system to be evaluated, the means for calculating the failure probability by the number of failures calculates the failure probability by the number of failures based on the performance data of the component of the other system; the means for calculating the spare part stock quantity calculates the spare part stock quantity based on the failure probability for each failure count; 3. The spare parts inventory quantity calculation device according to claim 1 or 2.

8. the means for acquiring the performance data of the components from the other system other than the system to be evaluated has a common dictionary for converting the names of the components in the other system into the names of the components in the system to be evaluated; The spare parts inventory quantity calculation device according to claim 7.

9. A method for calculating spare part inventory quantity executed by a computer, comprising: A step of calculating the number of failures of components constituting the system to be evaluated during the evaluation period and the failure probability for each failure count, based on the time-series failure and maintenance performance data of the components; calculating a spare part stock quantity of the part that can achieve a target value of the probability that the part will not be out of stock when a failure occurs, based on the failure probability for each failure count; A method for calculating the quantity of spare parts held.

10. On the computer, A step of calculating the number of failures of components constituting the system to be evaluated during the evaluation period and the failure probability for each failure count, based on the time-series failure and maintenance performance data of the components; calculating a spare part stock quantity of the part that can achieve a target value of the probability that the part will not be out of stock when a failure occurs, based on the failure probability for each failure count; A program that executes the following.

Images