Adaptive asset management apparatus and method

Abstract

The present invention relates to an adaptive asset management apparatus and method, the apparatus comprising: a memory including at least one instruction; and at least one processor for executing the at least one instruction stored in the memory, wherein the processor identifies an investment time point and an investment cost for each of a plurality of facilities on the basis of time-based management (TBM), identifies a corresponding year in which the investment cost exceeds a preset budget constraint, and changes the investment time point by calculating an optimal solution for the excess amount by which the investment cost exceeds the budget constraint by using a genetic algorithm. Other embodiments are also applicable.

Description

Adaptive asset management device and method

[0001] The present invention relates to an adaptive asset management device and method.

[0002] A rational investment plan involves establishing a strategy to prepare a cross-investment budget in advance, which is broadly categorized into CAPEX (capital expenditures) and OPEX (operating expenses). While it is not difficult to establish such a strategy for small-scale power facilities, finding the optimal timing for budget investment for various facilities from the perspective of a large-scale plant is not easy because it requires checking real-time list fluidity for each facility.

[0003] Currently, a common method involves applying statistical lifespan-based time-based management (TBM) investment timing to multiple facilities installed in large-scale plants; when these facilities reach their statistically calculated limit lifespan, the relevant equipment is replaced with new ones, thereby investing the budget before an accident occurs.

[0004] Furthermore, in the case of large-scale plants, budget constraints are set on an annual basis to invest in facilities. However, a problem arises where facilities are excluded from the investment plan if the lifespan limits of multiple facilities are concentrated in a specific year, causing the investment costs for that year to exceed the budget constraints.

[0005] Therefore, to resolve these issues, there is a growing need for a method to determine the optimal timing for budget investment by considering the real-time risks of the facilities.

[0006] Embodiments of the present invention, designed to solve these conventional problems, provide an adaptive asset management device and method capable of determining the investment timing of equipment by comprehensively considering the time-based management (TBM) investment timing based on the statistical lifespan of the equipment and the risk-based management (RBM) investment timing identified based on genetic algorithms.

[0007] An adaptive asset management device according to an embodiment of the present invention comprises a memory containing at least one instruction and at least one processor that executes the at least one instruction stored in the memory, wherein the processor identifies an investment timing and investment costs for each of a plurality of facilities based on time-based management (TBM), identifies a year in which the investment costs exceed a preset budget constraint, and calculates an optimal solution for the excess costs exceeding the budget constraint from the investment costs using a generic algorithm to change the investment timing.

[0008] In addition, the processor according to the present invention is characterized by identifying the risk corresponding to each of the target facilities included in the corresponding year and sorting the target facilities in order of highest risk.

[0009] In addition, the processor according to the present invention is characterized by performing a second sort of the target equipment based on the risk after performing replacement or repair on each of the target equipment for which the first sorting has been completed.

[0010] In addition, the processor is characterized by calculating the investment value of the target equipment that has been sorted in the second order by computing an objective function using the genetic algorithm.

[0011] In addition, the processor is characterized by calculating the investment value based on the prior risk corresponding to each of the target facilities before the replacement or repair and the subsequent risk corresponding to each of the target facilities after the replacement or repair.

[0012] In addition, the processor is characterized by calculating the annual investment value based on the confirmed subsequent MHI while changing the investment timing for each of the aforementioned previous MHI and the aforementioned target facility.

[0013] Additionally, the processor is characterized by distributing the excess cost to at least one other year corresponding to the optimal solution calculated by applying the excess cost, the number of target facilities, the reinforcement year, the budget constraint, and the annual investment value to the genetic algorithm.

[0014] In addition, the adaptive asset management method according to an embodiment of the present invention is characterized by comprising the steps of: an electronic device identifying an investment timing and investment costs for each of a plurality of facilities based on time-based management (TBM); the electronic device identifying a year in which the investment costs exceed a preset budget constraint; the electronic device calculating an optimal solution for excess costs exceeding the budget constraint from the investment costs using a generic algorithm; and the electronic device changing the investment timing by distributing the excess costs to at least one other year based on the optimal solution.

[0015] As described above, the adaptive asset management device and method according to the present invention has the effect of determining the optimal budget investment timing for multiple facilities equipped in a large-scale plant by determining the investment timing of facilities by comprehensively considering the time-based management (TBM) investment timing based on the statistical lifespan of the facilities and the risk-based management (RBM) investment timing identified based on genetic algorithms.

[0016] FIG. 1 is a block diagram showing the main configuration of an adaptive asset management device according to an embodiment of the present invention.

[0017] FIG. 2 is a flowchart illustrating an adaptive asset management method according to an embodiment of the present invention.

[0018] FIG. 3 is a diagram illustrating a method for determining the corresponding year for calculating the optimal solution according to an embodiment of the present invention.

[0019] Figure 4 is a graph illustrating a method for calculating an objective function when investing in equipment replacement according to an embodiment of the present invention.

[0020] FIG. 5 is a diagram illustrating a method for calculating an objective function when investing in equipment repair according to an embodiment of the present invention.

[0021] Figure 6 is a diagram showing the investment value of a target facility by reinforcement year calculated by operating an objective function according to an embodiment of the present invention.

[0022] FIG. 7 is a drawing showing an investment plan generated by randomly arranging genes in each reinforcement year for each target facility according to an embodiment of the present invention, with a preset number of genes.

[0023] FIG. 8 is a diagram showing the objective function value corresponding to the investment plan obtained at the end of the operation of the genetic algorithm according to an embodiment of the present invention.

[0024] Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings. The detailed description disclosed below, together with the accompanying drawings, is intended to describe exemplary embodiments of the present invention and is not intended to represent the only embodiment in which the present invention can be practiced. In order to clearly explain the present invention in the drawings, parts unrelated to the description may be omitted, and the same reference numerals may be used for identical or similar components throughout the specification.

[0025] FIG. 1 is a block diagram showing the main configuration of an adaptive asset management device according to an embodiment of the present invention.

[0026] Referring to FIG. 1, an asset management device (100) according to the present invention (hereinafter referred to as the electronic device (100)) may include a processor (110) and a memory (120).

[0027] The processor (110) identifies the investment timing and investment costs for each of the multiple facilities based on time-based management (TBM) and identifies the year in which the investment costs exceed a pre-set budget constraint. The processor (110) calculates an optimal solution for the excess costs exceeding the budget constraint in the investment costs and changes the investment timing by distributing the excess costs to at least one other year corresponding to the optimal solution. To this end, the processor (110) may include an investment timing verification unit (111), a budget management unit (112), a priority management unit (113), and an optimization unit (114).

[0028] The investment timing verification unit (111) verifies the investment timing and investment costs for each piece of equipment based on TBM (time-based management). Generally, equipment such as power equipment has a set lifespan limit at the time of shipment, and statistical data regarding the types of failures that may occur over the years of use and the repair costs associated with those failures are distributed. Accordingly, the investment timing verification unit (111) verifies the lifespan limit or types of failures for each piece of equipment, and can verify the time when the equipment needs to be repaired and the repair costs, as well as the time when the equipment needs to be replaced and the replacement costs, based on the years of use from the start of use of the equipment until the present. In the embodiment of the present invention, the repair costs and replacement costs of the equipment are collectively referred to as investment costs, and the repair timing and replacement timing are collectively referred to as investment timing.

[0029] The budget management department (112) checks the budget constraint. At this time, the budget constraint is a budget set by the user operating the facility, and may mean the maximum budget set for investment in the facility when operating the factory where the facility is installed. The budget management department (112) checks the investment cost for each investment time point by setting the period up to the next critical year from the current time point as the reinforcement year. The budget management department (112) identifies the investment time point where the investment cost exceeds the budget constraint the most among the investment time points where the investment cost for each investment time point exceeds the budget constraint as the corresponding year.

[0030] The priority management unit (113) calculates the costs incurred when equipment included in the confirmed investment costs for the year is not replaced or repaired as a risk, and performs a first sorting in order of highest risk.

[0031] The priority management unit (113) replaces or repairs equipment at the current time based on a rule base and performs secondary sorting. More specifically, the processor (110) calls a risk matrix that is stored in memory (120). At this time, the risk matrix is ​​generated using the probability of failure (PoF) grade and the consequence of failure (CoF) grade, and can be generated in the form of a matrix having an x-axis with the CoF grade and a y-axis with the PoF grade. The CoF grade is a grade of the economic loss cost in the event of a failure of power equipment, and can be graded from A to E depending on the magnitude of the loss cost, and each section can be set differently depending on the workplace or power equipment. The PoF grade is set based on the health score, and the lowest score among the equipment's condition index, degradation index, and strategic index can be selected as the final health score and reflected in the PoF. In addition, the PoF grade can be graded from A to E according to the final health score.

[0032] The priority management unit (113) performs mitigation based on a rule base to change the position of the equipment within the risk matrix. The priority management unit (113) can perform risk mitigation measures on equipment with a PoF grade of at least a certain grade, such as D and E, to improve the condition index and degradation index of the equipment, thereby allowing the equipment to be positioned in grade A or B within the risk matrix.

[0033] In this way, if at least one piece of equipment is replaced or repaired at a cost below the budget constraint, the risk of the equipment that was replaced or repaired is eliminated or reduced, so the risk per piece of equipment that may occur in that year may differ from the result of the first sorting. Therefore, the priority management unit (113) can perform a second sorting using the risk of equipment whose risk has changed due to replacement or repair and the risk of equipment whose risk has not changed.

[0034] The optimization unit (114) calculates the investment value of the second-sorted target equipment by calculating the objective function using a genetic algorithm. The optimization unit (114) calculates the risk before equipment replacement or repair by year and the risk after equipment replacement or repair by year. At this time, the optimization unit (114) can calculate the MHI (modified health index), which is used to evaluate the deterioration state of the equipment by comprehensively considering the initial state of the equipment, the degree of deterioration, the statistical equipment lifespan, and the years of use of the equipment, as the risk.

[0035] The optimization unit (114) calculates the integrated result of the risk over time by integrating the value obtained by subtracting the subsequent risk from the previous risk.

[0036] The optimization unit (114) can calculate the investment value by subtracting the investment cost from the integrated result of the risk over time. Additionally, the optimization unit (114) can calculate the annual investment value within the reinforcement year by increasing the investment timing in critical units, for example, in units of one year.

[0037] The optimization unit (114) calculates an optimal solution by applying the number of target facilities, reinforcement year, budget constraint, and annual investment value related to the risk included in the excess cost exceeding the budget constraint in the investment cost of the corresponding year to the genetic algorithm. Generally, the genetic algorithm performs the processes of initialize, selection, crossover, mutation, and termination.

[0038] When the operation of the genetic algorithm is completed, the optimization unit (114) can obtain the result of the replacement year and the objective function value for each facility based on the investment plan, and can use this to distribute the excess cost exceeding the budget constraint in the corresponding year to years other than the corresponding year. More specifically, the optimization unit (114) compares the objective function value with the budget constraint for each year and forms a final set of investment plans that pass the constraint. At this time, the constraint may be the cost obtained by adding the investment cost and the budget constraint for each year.

[0039] The memory (120) stores operation programs of the electronic device (100). In particular, the memory (120) can store a risk matrix generated by a manager of a plant with multiple installed facilities, and can store various algorithms such as a genetic algorithm for calculating an optimal solution.

[0040] Thus, the present invention has the effect of enabling adaptive asset management by considering both the TMB-based investment timing and the RMB (risk-based management) investment timing that yields maximum investment value using a genetic algorithm.

[0041] FIG. 2 is a flowchart illustrating an adaptive asset management method according to an embodiment of the present invention. FIG. 3 is a diagram illustrating a method for identifying a corresponding year to calculate an optimal solution according to an embodiment of the present invention. FIG. 4 is a graph illustrating a method for calculating an objective function when investing in equipment replacement according to an embodiment of the present invention. FIG. 5 is a diagram illustrating a method for calculating an objective function when investing in equipment repair according to an embodiment of the present invention. FIG. 6 is a diagram showing the investment value by reinforcement year for a target facility calculated by computing the objective function according to an embodiment of the present invention. FIG. 7 is a diagram showing investment plans generated by randomly placing genes in each reinforcement year for each target facility according to an embodiment of the present invention, with a preset number of entries. FIG. 8 is a diagram showing the objective function value corresponding to the investment plan obtained at the end of the computation of the genetic algorithm according to an embodiment of the present invention.

[0042] Referring to FIGS. 2 through 8, in step 201, the processor (110) checks the investment timing and investment costs for each piece of equipment based on TBM. Generally, equipment such as power equipment has a set lifespan limit at the time of shipment, and statistical data regarding the types of failures that may occur over the years of use and the repair costs associated with those failures are distributed. Accordingly, the processor (110) checks the lifespan limit or types of failures for each piece of equipment and can determine the timing and repair costs for repairing the equipment, and the timing and replacement costs for replacing the equipment, based on the years of use from the start of use of the equipment until the present. In the embodiment of the present invention, the repair costs and replacement costs of the equipment are collectively referred to as investment costs, and the timing of repair and replacement are collectively referred to as investment timing.

[0043] In step 203, the processor (110) checks the budget constraint. At this time, the budget constraint is a budget set by the user operating the equipment, which may mean setting the maximum budget that can be invested in the equipment when operating the factory where the equipment is installed. The processor (110) checks the investment cost for each investment time by setting the reinforcement years from the current point in time to the next critical years.

[0044] In step 205, the processor (110) identifies the year in which the investment cost exceeds the budget constraint the most among the investment times in which the investment cost for each investment time exceeds the budget constraint. This will be explained using Figure 3 below.

[0045] Referring to FIG. 3, the processor (110) can check the investment cost (301) for each year among the reinforcement years of 20 years from 2025 to 2044 based on the current time (2024). At this time, the unit of the investment cost (301) may be 100 million won. The processor (110) can check that among the investment costs (301) from 2025 to 2044, the years in which the budget constraint (302) is exceeded are 2025, 2037, 2038, 2039, 2043, and 2044. The processor (110) can check the year 2038, in which the investment cost exceeded the budget constraint (302) the most among the six years, as the corresponding year (303). In addition, in the embodiments of the present invention, the budget constraint (302) for each reinforcement year is described as being the same, but this is for the convenience of explanation and it is clarified that the budget constraint for each reinforcement year may vary by year.

[0046] In step 207, the processor (110) identifies the risks of the equipment included in the investment costs for the year (303), 2038, and performs a first sorting in order of highest risk. More specifically, the processor (110) can calculate the risk of the costs incurred if each piece of equipment is not replaced or repaired in 2038.

[0047] In step 209, the processor (110) replaces or repairs the equipment at the current time based on a rule base and performs secondary sorting. More specifically, the processor (110) calls a risk matrix that is previously stored in memory (120). At this time, the risk matrix is ​​generated using the probability of failure (PoF) grade and the consequence of failure (CoF) grade, and can be generated in the form of a matrix having an x-axis with the CoF grade and a y-axis with the PoF grade. The CoF grade is a grade of the economic loss cost in the event of a failure of power equipment, and can be graded from A to E depending on the magnitude of the loss cost, and each section can be set differently depending on the workplace or power equipment. The PoF grade is set based on the health score, and the lowest score among the equipment's condition index, degradation index, and strategic index can be selected as the final health score and reflected in the PoF. In addition, the PoF grade can be graded from A to E according to the final health score.

[0048] The processor (110) performs mitigation based on a rule base to change the position of the equipment within the risk matrix. The processor (110) can perform risk mitigation measures on equipment with a PoF grade of a specific grade or higher, such as D and E, to improve the condition index and degradation index of the equipment, thereby enabling the equipment to be positioned in grade A or B within the risk matrix. At this time, the criteria for determining grade D and E based on the rule base and the risk mitigation method are as shown in Table 1 below, and the criteria for determining the criteria and the risk mitigation method may be set differently depending on the workplace or power equipment.

[0049] Grade Status PoF Assessment Criteria and Risk Mitigation Methods Replacement Review Additional Inspection DGA Analysis Cycle ECritical 81-100 - High probability of failure or end of life - Reached replacement age - Repair or replacement required within 1 year - Online monitoring required Within 1 year 3 months 1 month DPoor 61-80 - Poor condition or clear signs of deterioration - Consider repair or replacement within 3 years - Online monitoring required Within 3 years 1 year 3 months

[0050] In this way, if at least one piece of equipment is replaced or repaired for a cost less than or equal to the budget constraint (302), the risk of the equipment based on the replacement or repair is eliminated or reduced, so the risk per piece of equipment that may occur in the corresponding year (303), which is 2038, may differ from the result of the first sorting. Therefore, the processor (110) can perform a second sorting using the risk of equipment whose risk has changed due to replacement or repair and the risk of equipment whose risk has not changed. In step 211, the processor (110) calculates the optimal solution for the excess cost exceeding the budget constraint (302) in the corresponding year (303). At this time, since the investment cost for the corresponding year (303), which is 2038, is approximately 800 million and the budget constraint (302) is approximately 200 million, the processor (110) calculates the optimal solution for the excess cost exceeding the budget constraint (302), which is approximately 600 million, using a generic algorithm (GA). The processor (110) calculates the investment value of the second-sorted target equipment by calculating the objective function using a genetic algorithm. The processor (110) calculates the risk before equipment replacement or repair by year and calculates the risk after equipment replacement or repair by year. The processor (110) calculates the modified health index (MHI) using the following mathematical formula 1 and can verify the calculated HMI value as the risk. MHI is an improved version of the existing health index (HI) and refers to a technique used to evaluate the deterioration state of equipment by comprehensively considering the initial state of the equipment, the degree of deterioration, the statistical equipment lifespan, and the years of use of the equipment.

[0051]

[0052] In this case, A is calculated by multiplying the AHI (asset health index) and the measured CoF in the initial state of the equipment (baseline calibration, initial degradation). is calculated through log-log regression with respect to the degradation slope. represents the statistical lifespan of the equipment, and t represents the current age of the equipment. Conventional investment value calculation methods calculate the value by subtracting the investment cost from the integral of risk over time, based on the equipment's PoF, CoF, and investment cost. However, since the equipment's PoF cannot exceed 100%, there is a problem where the investment value is calculated identically after a certain period due to saturation. In optimization solutions, if the objective function is identical or similar, a solution may result in an investment made well after the actual statistical lifespan of the equipment. Therefore, to improve upon this problem, the present invention calculates the investment value using MHI.

[0053] More specifically, the processor (110) calculates the integrated result of the risk over time by integrating the value obtained by subtracting the risk after replacement from the risk before replacement as shown in FIG. 4. At this time, the y-axis represents the risk by investment time point (unit: won), and the x-axis represents the year of replacement. For example, if the equipment is replaced in 2088, the processor (110) subtracts the risk after equipment replacement from the risk before equipment replacement. The processor (110) calculates the integrated result of the risk over time by integrating the subtracted value to obtain an area such as reference numeral 401. That is, the integrated result is calculated differently depending on the time of replacement, and the processor (110) can confirm that the risk has been minimized with the same cost as the value of the integrated result is larger.

[0054] The processor (110) calculates the integrated result of the risk over time by integrating the value obtained by subtracting the risk after the repair from the risk before the repair, as shown in FIG. 5. For example, if the equipment is repaired in 2088, the processor (110) subtracts the risk after the equipment repair from the risk before the equipment repair. The processor (110) calculates the integrated result of the risk over time by integrating the subtracted value to obtain an area such as reference numeral 501. That is, the integrated result is calculated differently depending on the time of the repair, and the processor (110) can confirm that the risk has been minimized at the same cost as the integrated result is larger.

[0055] The processor (110) can calculate the investment value by subtracting the investment cost from the integral result over time. Additionally, the processor (110) can calculate the annual investment value within the reinforcement year as shown in FIG. 6 by increasing the investment timing by a critical unit, for example, one year.

[0056] The processor (110) calculates an optimal solution by applying the number of target facilities (e.g., 100), reinforcement years (e.g., 20 years), budget constraints, and annual investment value related to the risk included in the excess cost of approximately 600 million that exceeds the budget constraint (302) in the investment cost of the corresponding year (303) to the genetic algorithm. The genetic algorithm in the present invention performs the processes of initialize, selection, crossover, mutation, and termination.

[0057] More specifically, during the initialization process, the processor (110) generates a preset number of investment plans by randomly placing genes in each reinforcement year for each target facility as shown in FIG. 7. At this time, the genes can be 0 and 1, and an investment plan can be generated by randomly placing 1 1 and 99 0s in each reinforcement year for each facility.

[0058] In the selection process, the processor (110) calculates an objective function for the generated investment plan and sorts the investment plans based on the calculated annual investment value. At this time, the processor (110) sorts the investment plans among the sorted investment plans so that the investment plan has the highest rank and selects an investment plan above the upper threshold.

[0059] In the crossover process, the processor (110) randomly selects two investment plans from among the investment plans above the upper threshold, exchanges 50% of the investment plans from each of the two selected investment plans to create a new descendant generation investment plan, and replaces the investment plans below the lower threshold with the created descendant generation investment plan.

[0060] In the mutation process, the processor (110) randomly applies the order of genes in the offspring generation investment plan generated in the crossover process. That is, by changing the order of each gene in the investment plan, it can be made easier to search for a global optimal solution. The processor (110) can replace investment plans below an upper threshold with offspring generation investment plans and repeat the selection, crossover, and mutation processes until there is almost no change overall to produce an optimal solution. For example, the processor (110) can calculate and sort the objective function for the newly generated offspring generation investment plan and repeat the selection, crossover, and mutation processes until the investment plan with the newly improved objective function is maintained within the upper threshold and no new investment plans are derived anymore, thereby producing an optimal solution.

[0061] In step 213, when the operation of the genetic algorithm is completed, the processor (110) can obtain the result of the replacement year and the objective function value for each facility based on the investment plan as shown in FIG. 8, and can use this to distribute the excess cost exceeding the budget constraint in the corresponding year (303) to years other than the corresponding year (303). More specifically, the processor (110) compares the objective function value with the budget constraint for each year and forms a final set of investment plans that pass the constraint. At this time, the constraint may be the cost obtained by adding the investment cost and the budget constraint for each year.

[0062] The embodiments of the invention disclosed in this specification and drawings are provided merely as specific examples to facilitate the explanation of the technical content of the invention and to aid in understanding the invention, and are not intended to limit the scope of the invention. Accordingly, the scope of the invention should be interpreted to include all modifications or variations derived based on the technical concept of the invention, in addition to the embodiments disclosed herein.

Claims

1. Memory containing at least one instruction; and It includes at least one processor that executes at least one instruction stored in the memory, The above processor is, An asset management device characterized by verifying the investment timing and investment costs for each of multiple facilities based on TBM (time-based management), identifying the year in which the investment costs exceed a preset budget constraint, and using a generic algorithm to calculate an optimal solution for the excess costs exceeding the budget constraint from the investment costs to change the investment timing.

2. In Paragraph 1, The above processor is, An asset management device characterized by identifying the risk corresponding to each target facility included in the aforementioned year and sorting the target facilities in order of highest risk.

3. In Paragraph 2, The above processor is, An asset management device characterized by performing a second sort of target equipment based on the risk after performing replacement or repair for each of the target equipment for which the first sorting has been completed.

4. In Paragraph 3, The above processor is, An asset management device characterized by calculating the investment value of the target equipment sorted in the second order by computing an objective function using the above genetic algorithm.

5. In Paragraph 4, The above processor is, An asset management device characterized by calculating the investment value based on the prior risk corresponding to each of the target facilities before the replacement or repair and the subsequent risk corresponding to each of the target facilities after the replacement or repair.

6. In Paragraph 5, The above processor is, An asset management device characterized by calculating annual investment value based on identified subsequent risks while changing the investment timing for each of the aforementioned prior risks and the aforementioned target facilities.

7. In Paragraph 6, The above processor is, An asset management device characterized by distributing the excess cost to at least one other year corresponding to the optimal solution calculated by applying the excess cost, the number of target facilities, the reinforcement year, the budget constraint, and the annual investment value to the genetic algorithm.

8. A step in which an electronic device verifies the investment timing and investment costs for each of multiple facilities based on TBM (time-based management); A step in which the electronic device identifies the year in which the investment cost exceeds a pre-set budget constraint; The electronic device calculates an optimal solution for excess costs exceeding the budget constraints in the investment costs using a generic algorithm; and The electronic device changes the investment timing by distributing the excess costs to at least one other year based on the optimal solution; An asset management method characterized by including

Images