Railway maintenance assistance system and railway maintenance assistance method

Abstract

The present invention proposes a component order-placement plan which reduces surplus stock of components. A railway maintenance assistance system for managing the order-placement timing and the stock quantity of railway equipment, the system comprising: a failure probability estimation unit that estimates a failure probability from the history of operation of a railway vehicle; a railway equipment required-quantity calculation unit that calculates the required quantity of a single type of railway equipment from the failure probability of the railway equipment; a railway equipment order-placement plan generation unit that receives a stochastic railway equipment required quantity, and generates an order-placement plan on the basis of an operation impediment risk pertaining to the occurrence of an operation impediment due to shortage in stock, cost incurred by excessive stock, and the delivery date and price of the railway equipment; and an input / output unit that outputs the generated order-placement plan.

Description

present invention. METHOD FOR CARRYING OUT THE INVENTION

[0012] Hereinafter, the embodiment of the present invention will be described using drawings. In each diagram used to explain embodiment, identical parts are given identical names and symbols as much as possible, omitting repeated explanations.

[0013] The present invention is not limited to the embodiments described below but encompasses various modifications and equivalent configurations within the scope of the appended claims. For example, the aforementioned embodiments have been described in detail to facilitate understanding of the present invention, and the present invention is not necessarily limited to those that include all the configurations described.

[0014] Also, the processing units and modules described in embodiments may be implemented in hardware by designing some or all of them, for example, as integrated circuits, or by interpreting and executing programs that implement each function in software.

[0015] The information described by embodiment can be table, database (DB), or data stored in main memory. [Embodiment 1 ]

[0016] <Overview of embodiment> In this embodiment, the railway maintenance assistance system estimates and aggregates the failure probability of each part constituting a railway vehicle, calculates the probabilistic replacement required amounts for all parts of the same type, and generates an ordering plan considering risk and cost based on this probabilistic replacement requirement.

[0017] In the present invention, the scope of railway vehicle parts is the part for which replacement parts are in inventory, such as wheels, brake shoes, and side sliding doors, with the assumption that they will be replaced. Note that the parts listed here are just examples and not limited to them. Also, since parts of the same type can be exchanged, this embodiment applies to orders for parts of the same type. <Configuration of the railway maintenance assistance system> Fig. 1 shows an example of a configuration diagram of a railway maintenance assistance system according to an embodiment of the present invention (Embodiment 1). It comprises a railway maintenance assistance system 10, an operations management system 20, a parts manufacturing information system 30, a financial management system 40, and an inventory management system 50.

[0018] The railway maintenance assistance system 10 comprises a probability of failure estimation unit 11, a replacement parts quantity calculation unit 12, a failure impact calculation unit 13, an order planning unit 14, an order planning and coordination unit 15, an input / output unit 16, vehicle equipment information 101, maintenance history 102, a maintenance plan 103, maintenance benefits 104, and replacement criteria information 105.

[0019] The operations management system 20 includes operational history 21, operating schedule 22, user history 23, and route information 24.

[0020] The parts manufacturing information system 30 includes delivery forecast information 31 and price forecast information 32.

[0021] The financial management system 40 includes budget information 41.

[0022] The inventory management system 50 comprises inventory information 51 and a purchasing unit 52.

[0023] In this embodiment, maintenance units are described as parts such as wheels, brake shoes, individual doors, lighting, air conditioning, ATS equipment, and other onboard equipment, as well as ground-based equipment such as branch, level crossings, and ticket gates; however, it is also possible to define maintenance units as individual vehicle or train compositions, and to apply this concept to railway equipment in general—that is, all equipment used to operate railway services.

[0024] Therefore, when applied to railway equipment, the name replacement parts quantity calculation unit 12 is the railway equipment required quantity calculation unit, and order planning unit 14 is called the railway equipment order planning unit.

[0025] Fig. 2 shows an example of the hardware configuration of the railway maintenance assistance system in the embodiment of the present invention.

[0026] The hardware comprises a railway maintenance support device 200, an operations management device 201, a parts manufacturing information device 202, a financial management device 203, and an inventory management device 204. These devices are equipped with processing units, storage units, and communication units. In addition, the railway maintenance support device 200 is equipped with an input device 205 and an output device 206, in addition to the processing unit, storage unit, and communication unit. Each of the units 200, 201, 202, 203, and 204 are implemented by computers connected via a network.

[0027] In this embodiment, each device is implemented on a standalone computer and explained as hardware connected via a network, but multiple devices may also be implemented on a single computer, A single device may be implemented by multiple computers.

[0028] Alternatively, it may be implemented using a cloud system that provides computing resources.

[0029] Note that the hardware configuration shown in Fig. 2 is merely an example and is not intended to be limiting. For example, in a railway maintenance assistance system, devices 200, 201, 202, 203, and 204 may be housed within a single device. <Process flow for generating replacement parts order plans> Fig. 3 shows an example flowchart of the replacement parts ordering plan generation process in the embodiment of the present invention. Using this flowchart and the overall structure described in Fig. 1, the processing flow will be explained. <Failure probability estimation> First, in step 301, the failure probability for each part of the railway vehicle is estimated. Step 301 is executed by the probability of failure estimation unit 11.

[0030] In Step 301, a failure probability model is used to estimate the failure probability of a part. A failure probability model is a model that estimates the failure probability of a part based on inputs such as the part’s operating time and the load accumulated on the part due to the operation of the railway vehicle.

[0031] Fig. 4 illustrates the configuration and operations of the probability of failure estimation unit. Process 301 is executed by the probability of failure estimation unit 11. The probability of failure estimation unit 11 comprises an operating time calculation unit 401, a failure probability model 402, and a maintenance benefits calculation unit 404.

[0032] The operational history 21 and operating schedule 22 are input into the operating time calculation unit 401. The operating time calculation unit 401 calculates the operating time of the target part for input into the failure probability model used to estimate the failure probability. Alternatively, the load accumulated on the target part may be calculated instead of the operating time.

[0033] After calculating the operating time, the operating time (accumulated load) is input into the failure probability model 402 to calculate the failure probability 403. While various models can be used for the failure probability model, the Weibull distribution function described in Japanese unexamined patent application publication No. 2023-157092 is used, for example. Note that the failure probability model 402 is not limited to the Weibull distribution function. Other failure probability models may also be used.

[0034] Additionally, maintenance can be used to determine the actual reduction in operating time, and the reduction in failure probability can be calculated from the desired actual reduction in operating time. By reflecting the required failure probability reduction in the failure probability, the failure probability reduction effect can also be obtained.

[0035] Generally, maintenance can range from minor maintenance tasks such as scheduled repairs to large-scale maintenance such as overhauls. Consider the rejuvenation effect of the equipment lifespan of the target parts according to the scale of these maintenance tasks.

[0036] Fig. 5 illustrates how maintenance reduces the probability of failure. For example, if the probability of failure based on accumulated operating hours was previously at point 501, maintenance—by effectively “rejuvenating” the railway equipment, i.e., reducing the effective operating hours—shifts this probability to point 502. Consequently, it is possible to calculate the reduction in the probability of failure achieved through maintenance.

[0037] Maintenance history 102 and maintenance plan 103 are entered into the maintenance benefits calculation unit 404 to calculate the actual reduction in operating time due to maintenance. The maintenance benefits calculation unit 404 calculates the actual reduction in operating hours according to the scale of maintenance work compared to maintenance benefits 104.

[0038] This effective reduction in operating time is summed with the operating time of the operating time calculation unit 401 and input into the failure probability model 402. Although the actual operating time reduced by maintenance is used as an example, when inputting accumulated load into the failure probability model, the amount reduced by accumulated load may be used.

[0039] The failure probability of 403 is calculated for each part.

[0040] Fig. 6 illustrates the failure probability of railway equipment parts. The failure probability of each part 403 is shown considering the reduction of failure rates through maintenance. The probability of failure up to the present point is calculated using operational history 21 and maintenance history 102.

[0041] On the other hand, the probability of future failures is calculated using the operating schedule 22 and the maintenance plan 103. Since future failure probabilities are not deterministic and variable, they have a confidence interval of 601. For example, the confidence interval 601 represents the 5%~95% confidence interval with dashed lines.

[0042] Additionally, the reduction in failure probability through maintenance is expressed as 602. As shown in Fig. 6, maintenance benefits are expressed as the actual reduction in operating hours. In other words, when maintenance is performed, the probability of failure of parts decreases discontinuously at that point. <Calculation of required replacement parts> Processing 302 aggregates the failure probabilities of each part estimated in processing 301 to calculate the required quantity of replacement parts. Processing 302 executes the failure probability of each part and replacement criteria information 105 as input in the replacement parts quantity calculation unit 12.

[0043] The failure probability of each part is aggregated over time to predict the required amount of replacement parts. For example, if wheel AA1 has a 20% failure probability and wheel AA2 has a 10% failure rate, the expected value is calculated by adding the failure probabilities of both to calculate the required quantity of 0.3 replacement parts. By performing similar calculations, it is possible to determine the total required replacement parts for the same type of part.

[0044] Fig. 7 is a diagram illustrating the required replacement parts for railway equipment. As shown in Fig. 6, the probability of future failures has variance. In other words, the required amount of replacement parts also has variance and can be calculated with confidence intervals.

[0045] Also, according to replacement criteria information 105, when a certain probability of failure is reached, the target part is replaced. Calculations are made to calculate the trend in the required quantity of replacement parts, considering future replacement timing. <Generating a replacement parts order plan> This section explains how to generate an order plan for replacement parts. It generates an optimal order plan that considers the required quantity of replacement parts calculated in step 302, as well as the costs associated with excess inventory, the risk of operational disruptions due to parts shortages, and the delivery times and price fluctuations of replacement parts.

[0046] In processing 303, various parameters and constraints are set to execute the order plan optimization calculation. Parameters refer to, for example, the cost factor of the objective function, which will be described later. Constraints refer to, for example, the budget for ordering parts. Parameters and constraints are not limited to those described herein, and any other necessary additions may be added.

[0047] Next, in step 304, the system performs the order planning optimization calculation. Step 304 is executed by the order planning unit 14. Using the required quantity of replacement parts calculated by the replacement parts quantity calculation unit 12, the failure impact amount calculated by the failure impact calculation unit 13, delivery forecast information 31, price forecast information 32, budget information 41, and inventory information 51 as inputs, the system calculates the parts order plan.

[0048] The part ordering plan, including the requested order quantity and ordering timing, is output on screen from the input / output unit 16 and obtained user approval. Once approved, the parts ordering plan is directed from order planning and coordination unit 15 to purchasing unit 52.

[0049] Fig. 8 shows an example of the replacement parts ordering plan generation screen for the railway maintenance assistance system in the embodiment of the present invention. This screen is used for order planning optimization calculations in processing 304. The order plan generation screen 800 is displayed by the output device 206, and user operations are accepted from the input device 205.

[0050] Order planning refers to the timing of ordering replacement parts and the order quantity for each order. When the order timing and quantity are determined for the required replacement parts, a solid line for the replacement parts inventory 801 is output.

[0051] In this embodiment, we divided the problem into two optimization problems: optimizing order timing and order quantity. Order timing optimization is calculated as a combination optimization that determines whether orders are placed from a regular selection of candidate order timing. Order volume optimization is calculated as linear optimization. The breakdown of this optimization problem is just one example; otherwise, for example, it can be computed simultaneously as a mixed integer programming problem.

[0052] Also, the determinant is the order quantity and order timing, but other variables may be added or reduced. For example, if the order timing is fixed on a regular basis, it is acceptable to formulate only the order quantity as a deciding variable. The decision variable may be selected by the user from the input device 205 in the determination variable selection section 803.

[0053] The order planning optimization calculation is formulated as a minimization problem for the objective function of the following formula 1. Determine the order quantity and timing for which this objective function is minimized. Note that it is not limited to the objective functions described here, other indicators may also be included.

[0054] Items included in the objective function may be selected by the user via input / output unit 16 from input device 205 in the objective function selection field 804. An example of an objective function is as follows.

[0055] [formula 1] (Property Tax + Inventory Maintenance Costs) x (Inventory Volume) + (Impact of Operational Disruptions) x (Probability of Inventory-outs) + (Parts Price) x (Order Quantity) x (Order Timing)    ..... (Formula 1) The first term represents the excess inventory costs associated with holding replacement parts for replacement. The second term represents the impact of cost—specifically, the risk of inventory-out—that occurs when a shortage of spare replacement parts renders the target railway vehicle unusable, resulting in an operational disruption. The third term represents the ordering cost incurred when ordering replacement parts.

[0056] Excess inventory costs and shortage risks are in a tradeoff relationship. For example, reducing the inventory of spare replacement parts to reduce the cost of excess inventory increases the risk of shortages.

[0057] On the other hand, holding more inventory to reduce shortage risks increases the excess inventory cost. By balancing excess inventory costs and shortage risks while minimizing ordering costs, we generate optimal replacement parts ordering plans.

[0058] The operational disruption risk is calculated as an impact amount based on the number of affected users and the importance of the target line when the target railway vehicles cannot operate due to parts shortages.

[0059] Although this embodiment explains optimization, it does not necessarily mean obtaining the most efficient maintenance plan; rather, it is about obtaining a reasonable maintenance plan for actual maintenance.

[0060] Although the maintenance plan may not always be optimal due to factors such as reviewing maintenance plans, fluctuations in part prices, and damage to railway equipment caused by disasters, it allows for a more rational maintenance plan compared to maintenance based on manual experience and intuition.

[0061] Fig. 9 is a diagram illustrating the probability of inventory shortages for replacement parts. Fig.9 is an example where, when a certain time section 802 is cut from Fig. 8, the probability density of required replacement parts is plotted on the horizontal axis and the required amount of replacement parts on the vertical axis.

[0062] This probability density means, for example, that on average 200 replacement parts are needed, but if multiple failures occur, there is a 5% chance of needing 300 replacement parts; conversely, if there are few failures, there is a 5% chance of needing 100 replacement parts.

[0063] When calculating inventory based on the order timing and quantity of the determinant variables for this probabilistic density of replacement parts requirements, the area enclosed by the line between the probability density and inventory quantity indicates the probability of an inventory shortage. For example, if the inventory is set at 260 units, the probability that the required replacement parts exceed 260 units, and the inventory shortage is calculated as 15%.

[0064] Fig. 10 is an example flowchart of the order planning optimization process for the railway maintenance assistance system in the embodiment of the present invention. Shows the computational process in processing 304.

[0065] First, processing 1001 performs optimization of the order timing. In the initial calculation, the variable determining the order quantity is fixed at any value, and the order timing is determined to minimize the objective function of (Formula 1).

[0066] Next, process 1002 fixes the order timing determined by process 1001 and executes order volume optimization to minimize the objective function of (Formula 1).

[0067] After the order quantity is determined, this order quantity is repeatedly optimized as a fixed value for the order quantity of processing 1001. That is, in processing 1001 and 1002, optimization is performed while alternately fixing the order timing and quantity. Processing 1003 confirms the convergence conditions of this iterative calculation. For example, when the difference in order timing and order quantity between loops becomes sufficiently small, the repeated calculation is terminated.

[0068] Finally, the optimal solution for the order timing and quantity obtained in processing 1004 is output as an order plan to complete the process.

[0069] Changes to order timing and quantity can be made by accepting user specifications or generated within this system. <Adjustment of the order plan for replacement parts> In processing 305, after generating the optimal order plan, the order plan is adjusted according to the situation. Order planning is a solution that minimizes the value of the set objective function. Include user adjustments to finalize the ordering plan.

[0070] Fig. 11 shows an example of the replacement parts ordering planning adjustment screen in the railway maintenance assistance system of the embodiment of the present invention.

[0071] The order plan adjustment screen 1100 displays the required quantity of replacement parts and the inventory plan based on the optimal order plan, allowing the user to refer to this information and adjust the order plan. The order planning adjustment screen 1100 is displayed on the output device 206 via the input / output unit 16, and the user performs operations from the input device 205. The input / output unit 16 receives user input from input device 205, and the order planning unit 14 recalculates the order plan.

[0072] The 1104 shown in the graph represents the inventory amount based on the order plan, 1105 is the average required replacement parts, 1106 is the maximum required replacement parts, and 1107 is the minimum required replacement parts.

[0073] For example, if the probability of out-of-inventory below the maximum required replacement parts in portion 1101 of the inventory plan cannot be tolerated, adjustments such as increasing the order quantity in 1102 to reduce the out-of inventory probability at 1101 are made.

[0074] This adjustment is performed by moving the user to the area where they want to change the order quantity via the input device 205. Additionally, rules for out-of-inventory probabilities may be set to automatically adjust the probability.

[0075] When changing the order plan, the change in the objective function before and after the plan adjustment is displayed in 1103. In the example in Fig. 11, increasing the order quantity raises excess inventory costs and ordering costs, while shortage costs decrease. However, when compared to the optimal order planning calculated by optimization calculations, the total objective function value increases. Users review these figures and decide on the final order plan.

[0076] Next, in processing 306, after adjusting the order plan, it is determined whether to modify the order plan again by adding new constraints. When adding constraints and recalculating, the process from processing 303 is repeated. For example, recalculation is done by adding constraints on the probability of inventory shortage.

[0077] Processes 305 and 306 are executed in the order planning and coordination unit 15.

[0078] Finally, processing 307 decides and outputs the finalized order plan after adjustment. This ordering plan is entered into purchasing unit 52, where the actual parts order is executed.

[0079] With this embodiment, maintenance departments managing the timing of parts replacement for railway vehicles can quantitatively assess operational disruption risks, ensure the reliability of railway systems, reduce surplus parts inventory, and formulate rational parts ordering plans. [Embodiment 2 ]

[0080] <Overview of Embodiment> In embodiment 2, the railway maintenance assistance system in embodiment 1 adds additional processing to the method for calculating the required quantity of replacement parts and generating the replacement parts order plan.

[0081] Fig. 12 is a diagram explaining the required amount of replacement parts when the timing of part replacement is changed.

[0082] In calculating the required amount of replacement parts in embodiment 1, from replacement criteria information 105, parts are replaced when a certain probability of failure is reached. Here, the replacement criteria for parts are constant, and the amount of parts used is calculated according to certain criteria. In other words, as shown in Fig. 12, the required amount of replacement parts 1201 once calculated does not change.

[0083] On the other hand, if the replacement criteria are set differently for each part, the number of parts used will vary, causing the required quantity of replacement parts to change each time. For example, as shown in Fig. 12 (1202), the required quantity of replacement parts will increase or decrease compared to the quantity (1201) when the replacement criteria are uniform.

[0084] Parts replacement should also be judged according to risk. Therefore, the determination of part replacement is also incorporated into the optimization calculations for replacement parts ordering plans. A part replacement check refers to adjusting the timing of part replacement.

[0085] Fig. 13 is a configuration diagram (Embodiment 2) of the railway maintenance assistance system in the embodiment of the present invention.

[0086] The difference from the configuration diagram of embodiment 1 is the parts usage calculation unit 1301. In order planning unit 14, the timing of replacing each part is also incorporated into the optimization problem. In other words, in addition to determining the timing and quantity of orders, the timing for replacing each part is also determined.

[0087] When the maintenance target is railway equipment rather than parts, the parts usage calculation unit 1301 may also be called the railway equipment usage calculation unit.

[0088] The timing of parts replacement is adjusted by balancing the cost of preparing replacement parts with the risk of operational disruptions. For example, delaying part replacement increases the probability of failure, but since fewer replacement parts are used, it is cost-effective.

[0089] If a part has minimal impact even if a failure occurs, it allows for greater risk and cost reduction. On the other hand, although the required replacement parts increase, it is also possible to reduce the risk of operational disruptions by timing replacement earlier and replacing parts with a lower probability of failure.

[0090] When the replacement timing of each part changes, the frequency of replacement is no longer fixed, so the amount of parts used also changes. The parts usage calculation unit 1301 calculates the amount of parts usage based on the part replacement determination as input. This part usage is entered into the replacement parts quantity calculation unit to update the required replacement parts quantity.

[0091] The required amount of replacement parts affects the timing and quantity of orders. Therefore, the order timing and quantity are determined according to the replacement timing for each part and the required quantity of replacement parts that change depending on the replacement timing.

[0092] With this embodiment, by formulating a reasonable parts ordering plan and considering the timing of parts replacement, it is possible to ensure the reliability of the railway system, further reduce surplus parts inventory, and lower costs. [Embodiment 3 ]

[0093] <Overview of embodiment> In embodiment 3, the railway maintenance assistance system in embodiment 1 is discussed, with the train composition of railway vehicles as the subject of the ordering plan.

[0094] Fig. 14 is a configuration diagram (Embodiment 3) of the railway maintenance assistance system in the embodiment of the present invention.

[0095] Since the target is the train composition of railway vehicles, based on the configuration diagram in embodiment 1, the railway maintenance assistance system 10 is divided into the order planning unit 14, operational train composition calculation unit 1403, train composition information 1404 and number of allocated train composition 1405.

[0096] Additionally, the vehicle manufacturing information system 1410 includes vehicle delivery forecast information 1411 and vehicle price forecast information 1412. Furthermore, the order management system 1420 includes a vehicle purchasing unit 1421.

[0097] When railway equipment is the maintenance target, the vehicle manufacturing information system 1410 is the railway equipment manufacturing information system, the vehicle delivery forecast information 1411 is the railway equipment delivery forecast information, and the vehicle price forecast information 1412 is the railway equipment price forecast information, vehicle purchasing unit 1421 may also be referred to as railway equipment purchasing unit.

[0098] In this embodiment, when generating the ordering plan, the failure probability for each train composition is estimated and the operational train composition is calculated.

[0099] Fig. 15 is a diagram illustrating the failure probability of a single train composition of railway vehicles. The probability of failure in the composition train increases with age.

[0100] Fig. 16 is a diagram illustrating the failure probability of the entire train composition of a railway vehicle. When estimating failure probability per train unit for all train compositions, the failure probability for each train composition can be calculated as shown in Fig. 16. In other words, based on the failure probability for each train composition, the operational train composition for the entire train composition can be calculated using (Formula 2).

[0101] [Formuler 2] (operational train composition) = S (1 - failure probability) ... (Formula 2) Generally, there are train compositions used for actual operations (number of allocated train compositions) and reserve train compositions. That is, if the operational train composition calculated from the failure probability of all train compositions meets the number of allocated train compositions, then even if a failure occurs in the train composition used for operation, a backup train composition can be assigned, this can prevent operational disruptions.

[0102] Fig. 17 is a diagram illustrating the method for generating railway vehicle order plans. In the ordering plan, the timing of ordering and removal and the number of train compositions for ordering and removal are determined. By provisionally determining the timing and number of train compositions for ordering and removal, the total number of train compositions is calculated as 1701.

[0103] Furthermore, from the failure probability of each train composition, operational train composition 1702 is calculated using (Formula 2). Since the probability of failure can be calculated with variance, the operational train composition 1702 also has variance.

[0104] If the operational train composition 1702 exceeds the number of allocated train composition 1405, the risk of operational disruption is zero; however, if it is likely to fall below the number of allocated train composition, the risk of operational disruption arises.

[0105] Fig. 18 is an example flowchart showing the optimization process of railway vehicle ordering planning in the railway maintenance assistance system in the embodiment of the present invention.

[0106] According to this processing flowchart, the method for generating order plans for railway vehicle train composition in this embodiment will be described.

[0107] First, processing 1801 sets parameters and constraints for the order plan optimization calculation. Parameters refer to, for example, the cost factor of the objective function, which will be described later. Constraints include, for example, the number of orders or removals, the number of train compositions per removal, and the budget for placing the order. Parameters and constraints are not limited to those described herein, and any other necessary additions may be added.

[0108] Next, in processing 1802, the timing of order / removal and the number of train compositions for order / removal are set as determinants.

[0109] Next, in processing 1803, the transition of all train compositions shown in 1701 is calculated according to the timing of order and removal and the number of train compositions set in processing 1803. In the case of removal, the oldest train composition is prioritized from all train compositions.

[0110] After determining the transition in the total number of train compositions, the failure probability of each train composition is estimated in processing 1804 and the operational train composition 1702 is calculated.

[0111] After calculating the operational train composition 1702, the objective function is calculated in processing 1805. The objective function is, for example, the following (Formula 3).

[0112] [Formula 3] (fixed asset tax + maintenance costs) x (total number of train compositions) + (amount of financial impact due to service disruptions) x (number of allocated train compositions insufficient probability) + (order price) x (number of allocated train compositions ordered) x (timing of order) + (removal costs) x (number of removal train compositions) ... (Formula 3) The first term represents the cost of maintaining the number of train compositions owned by the railway company. The second term represents the financial impact, that is the operational disruption risk—that occurs when the number of operational train compositions falls below the number of allocated train compositions. The third term represents the cost of ordering railway rolling stock. The fourth term represents the labor costs associated with removal train compositions.

[0113] In particular, the cost of owning train composition and the risk of operational disruptions are in a trade-off relationship. For example, reducing the cost of maintaining train composition, reducing the number of train compositions naturally reduces operational train composition, increasing the risk of operational disruptions due to not meeting the number of allocated train compositions.

[0114] On the other hand, increasing the number of train compositions to reduce operational disruption risks raises the cost of owning train compositions. Therefore, by balancing train composition ownership costs with operational disruption risks while minimizing ordering and removal costs, optimal vehicle ordering plans can be generated.

[0115] Note that the objective functions described here are not limited and may include other indicators. Also, the decision variable is set to the order / removal timing and the number of train compositions for ordering / removal, but other variables may be added or reduced. For example, if the timing of ordering and removal is fixed periodically, only the number of train compositions for ordering and removal can be used as a deciding variable and the objective function.

[0116] After calculating the objective function, the convergence conditions are checked in processing 1806. For example, when the difference in the determining variable between loops becomes sufficiently small, the repeated computation is terminated. If the convergence condition is not met, the timing and number of train compositions for ordering and removal are set to improve the objective function to a better value, and the process is repeated starting from process 1802. If the convergence condition is satisfied, the number of orders and order timing obtained in processing 1807 are output as the order plan and concluded.

[0117] With the above embodiment, even when targeting railway vehicles, it is possible to quantitatively assess the risk of operational disruptions, ensure appropriate reliability, reduce the number of surplus train compositions in railway vehicles, and formulate ordering plans.

[0118] It should be noted that the present invention is not limited to the above embodiments and includes various modifications. For example, the above embodiment is described in detail to clearly explain the present invention and is not necessarily limited to comprising all the parts described. REFERENCE SIGNS LIST

[0119] 10: railway maintenance assistance system 11: probability of failure estimation unit 12: replacement parts quantity calculation unit 13: failure impact calculation unit 14: order planning unit 15: order planning and coordination unit 16: input / output unit 20: operations management system 21: operational history 22: operating schedule 23: user history 24: route information 30: parts manufacturing information system 31: delivery forecast information 32: price forecast information 40: financial management system 41: budget information 50: inventory management system 51: inventory information 52: purchasing unit 101: vehicle equipment information 102: maintenance history 103: maintenance plan 104: maintenance benefits 105: replacement criteria information 1301: parts usage calculation unit 1403: operational train composition calculation unit 1405: number of allocated

Claims

1. A railway maintenance assistance system for managing the inventory levels and order schedules of railway equipment comprising:a probability of failure estimation unit estimates a probability of failure based on an operational history of vehicle,a railway equipment requirement calculation unit calculates a required quantity of railway equipment of same type based on the probability of failure of the railway equipment,a railway equipment order planning unit receives a probabilistic railway equipment requirement, generates an order plan based on risk of operational disruption caused byinventory shortage, cost resulting from excess inventory, anddelivery time and price of the railway equipment, andan input / output unit outputs generated order plan.

2. The railway maintenance assistance system described in claim 1, the railway equipment is a part.

3. The railway maintenance assistance system described in claim 1,the railway equipment is a train composition.

4. The railway maintenance assistance system described in claim 1,the railway equipment requirement calculation unit modifies a replacement schedule for the railway equipment andcalculates the usage volume of the railway equipment based on the replacement schedule.

5. The railway maintenance assistance system described in claim 1,the railway equipment requirement calculation unit estimates the failure probability for each railway equipmentbased on its operational history, future operating schedule,maintenance history, and future maintenance plan and calculates the required quantity of the railway equipment by summing the estimated the failure probabilities.

6. The railway maintenance assistance system described in claim 1,the input / output unit outputs a plan for the required quantity of railway equipment and a plan for the inventorylevel of railway equipment based on the order plan generatedby the railway equipment order planning unit,receives changes to the order quantity and timing, and the railway equipment order planning unit recalculates the order plan based on received order quantity and timing.

7. The railway maintenance assistance system described in claim 1,the operational disruption risk used by the railway equipment order planning unit to create an order plan is afigure representing a financial impact calculated based on the number of affected passengers and a importance of the affectedroute if railway operations cannot be continued due to a shortage of railway equipment.

8. The railway maintenance assistance system described in claim 1,the railway maintenance assistance system comprising a maintenance benefits calculation unit determines the reductionin the effective operating time of parts resulting from maintenance work and calculates the reduction in the failureprobability based on the reduction.

9. A railway maintenance assistance method for managing the inventory levels and ordering schedules of railway equipment, wherein:a probability of failure estimation unit estimates the probability of failure based on the operational history of vehicle,a railway equipment requirement calculation unit calculates the required quantity of railway equipment of the same type based on the probability of failure of the railway equipment,a railway equipment order planning unit receives the probabilistic railway equipment requirement, generates anorder plan based on the risk of operational disruptions causedby inventory shortages, costs resulting from excess inventory, and delivery time and price of railway equipment, andan input / output unit outputs generated order plan.

10. The railway maintenance assistance method described in claim 9, the railway equipment is a part.

11. The railway maintenance assistance method describedin claim 9,the railway equipment is a train composition.

Images