Transportation plan creation apparatus, transportation plan creation method, and recording medium

Abstract

In this transportation plan creation apparatus, a train service plan creation means creates a train service plan on the basis of a planning condition. A railcar operation plan creation means creates a railcar operation plan on the basis of the train service plan. A crew operation plan creation means creates a crew operation plan on the basis of the railcar operation plan. An adjustment means adjusts the train service plan, the railcar operation plan, and the crew operation plan on the basis of user's preferences.

Description

Transport Plan Creation Device, Transport Plan Creation Method, and Recording Medium

[0001] The present disclosure relates to the creation of transport plans for transportation agencies.

[0002] Trains, buses, airlines, etc. operate and are utilized according to pre-established operation plans. Patent Document 1 discloses a system for creating an operation plan using an operation plan creation algorithm.

[0003] Japanese Unexamined Patent Application Publication No. 2023 - 82606

[0004] The formulation of an operation plan needs to be carried out integrally with the vehicle operation plan and the crew operation plan. However, since the goals of each are different, if the formulation of each plan is carried out independently, there is a problem that the optimality of each plan deteriorates.

[0005] One object of the present disclosure is to provide a transport plan creation device capable of optimizing each plan as a whole by adjusting the goals of individual plans.

[0006] In one aspect of the present disclosure, a transport plan creation device includes: operation plan creation means for creating a train operation plan based on plan conditions; vehicle operation plan creation means for creating a vehicle operation plan based on the train operation plan; crew operation plan creation means for creating a crew operation plan based on the vehicle operation plan; and adjustment means for adjusting the train operation plan, the vehicle operation plan, and the crew operation plan based on user preferences.

[0007] In another aspect of the present disclosure, a transport plan creation method executed by a computer includes: creating a train operation plan based on plan conditions; creating a vehicle operation plan based on the train operation plan; creating a crew operation plan based on the vehicle operation plan; and adjusting the train operation plan, the vehicle operation plan, and the crew operation plan based on user preferences.

[0008] In yet another aspect of this disclosure, the recording medium records a program that causes a computer to perform the following processes: create a train operation plan based on planning conditions; create a vehicle operation plan based on the train operation plan; create a crew operation plan based on the vehicle operation plan; and adjust the train operation plan, the vehicle operation plan and the crew operation plan based on user preferences.

[0009] This diagram shows the overall configuration of the transportation planning system related to this disclosure. This is a block diagram showing the hardware configuration of the transportation planning device. This is a block diagram showing the functional configuration of the transportation planning device. This diagram schematically shows an example of a train timetable. This diagram schematically shows an example of a vehicle operation schedule showing a vehicle operation plan. This diagram shows an example of a route schedule showing a crew operation plan. This diagram shows the functional configuration of the train timetable planning unit. This diagram illustrates how to create a train timetable. This is another diagram illustrating how to create a train timetable. This is another diagram illustrating how to create a train timetable. This is another diagram illustrating how to create a train timetable. This is another diagram illustrating how to create a train timetable. This diagram illustrates how to create a train timetable. This diagram illustrates how to create a vehicle operation plan. This diagram illustrates how to create a vehicle operation plan. This diagram illustrates how to create a vehicle operation plan. This diagram shows the functional configuration of the crew operation plan unit. This diagram illustrates how to create a vehicle operation plan. This diagram illustrates how to create a vehicle operation plan. This diagram illustrates how to create a crew operation plan. This diagram shows one example of extracting decision timing and train status. This diagram shows another example of extracting decision timing and train status. This diagram shows an example of simulator display. This is a flowchart of the transportation planning process. This is a block diagram showing the configuration of another transportation planning device related to this disclosure. This is a flowchart of processing by another transportation planning device related to this disclosure.

[0010] Preferred embodiments of this disclosure will be described below with reference to the drawings. <Background> Planning railway operations must be carried out in conjunction with vehicle operation plans and crew operation plans. However, since the objectives of each plan are different, if each plan is planned independently, the optimality of each plan may decrease. For example, the objective of a train timetable plan is to satisfy transportation demand. The objective of a vehicle operation plan is to make the most use of the limited number of vehicles and to avoid changing vehicles midway through a train operation. In addition, in a crew operation plan, crew members may be replaced at intermediate stations during a train operation, but the objective is to satisfy the working conditions of the crew members, such as working hours and rest periods. Thus, the measure of optimality in planning differs for each plan. In other words, available resources (vehicles, crew members, etc.) are finite, and the resources currently available change depending on the allocation and assignment of resources in the past (up to that point), but it is required to optimize all plans by considering all such resources and demand simultaneously. Therefore, in the following embodiment, modules for creating operation plans, vehicle operation plans, and crew operation plans are provided separately, and a module for automatically adjusting the targets of each plan is also provided to optimize the overall transportation plan.

[0011] <First Embodiment> [Overall Configuration] Figure 1 shows the overall configuration of the transportation plan creation system according to the first embodiment. The transportation plan creation system 1 is a system for creating railway transportation plans and comprises a user terminal 2 and a transportation plan creation device 100. The transportation plan creation device 100 is composed of a server device and the like. The user terminal 2 is a terminal device such as a personal computer used by a user who creates a transportation plan, for example, a planning officer of a railway company. The user terminal 2 and the transportation plan creation device 100 are connected so as to be able to communicate via a network or the like.

[0012] The user transmits the planning conditions for the transportation plan to be created to the transportation plan creation device 100 via the user terminal 2. The planning conditions are the conditions that the transportation plan should meet, such as the number of trains running and the volume of transport, based on the demand forecast, and may also include requests for improvements to the transportation plan and targets for each plan. For example, the planning conditions may be the amount of passenger transport to be achieved and the number of trains running. Regarding the amount of passenger transport, for example, the conditions may be "X million people on weekdays, Y million people on holidays" for a certain railway line or section (between stations). Also, regarding the number of trains running, the conditions may be "six trains in each direction per hour during the morning rush hour, and four trains in each direction per hour during the evening rush hour."

[0013] The transportation plan creation device 100 creates a transportation plan based on the planning conditions. Specifically, the transportation plan creation device 100 creates a train operation plan (also called a "train timetable plan"), a vehicle operation plan, and a crew operation plan, and transmits them to the user terminal 2. In the following, "transportation plan" will include the train timetable, vehicle operation plan, and crew operation plan.

[0014] [Hardware Configuration] Figure 2 is a block diagram showing the hardware configuration of the transportation planning device 100. As shown in the figure, the transportation planning device 100 comprises a processor 11, an interface (IF) 12, a memory 13, a database (DB) 14, and a recording medium 15. These elements are connected by a bus 18.

[0015] The processor 11 is a computer such as a CPU (Central Processing Unit), and controls the entire transportation planning device 100 by executing a pre-prepared program. The processor 11 can be a CPU, GPU (Graphics Processing Unit), DSP (Digital Signal Processor), MPU (Micro Processing Unit), FPU (Floating Point Number Processing Unit), PPU (Physics Processing Unit), TPU (Tensor Processing Unit), quantum processor, microcontroller, or a combination thereof. The processor 11 executes the transportation planning process described later.

[0016] IF12 receives planning conditions from user terminal 2. IF12 also transmits the transportation plan created by the transportation plan creation device 100 to user terminal 2.

[0017] Memory 13 is composed of ROM (Read Only Memory), RAM (Random Access Memory), and the like. Memory 13 stores various programs executed by the processor 11. Memory 13 is also used as working memory while the processor 11 is executing various processes.

[0018] DB14 stores various types of data necessary for creating transportation plans. For example, DB14 stores railway route information, train (vehicle) information used by railway companies, and crew information.

[0019] The recording medium 15 is a non-volatile, non-temporary recording medium such as a disk-shaped recording medium or semiconductor memory, and is configured to be detachable from the transport planning device 100. The recording medium 15 stores various programs that the processor 11 will execute. When the transport planning device 100 performs various processes, the programs stored on the recording medium 15 are loaded into the memory 13 and executed by the processor 11.

[0020] [Functional Configuration] Figure 3 is a block diagram showing the functional configuration of the transportation planning device 100. The transportation planning device 100 includes a user IF 21, an integrated control unit 22, a route DB 23, a vehicle DB 24, a crew DB 25, a train timetable planning unit 30, a vehicle operation planning unit 40, and a crew operation planning unit 50.

[0021] User IF21 is composed of IF12 as shown in Figure 2 and performs interface processing for data input and output with user terminal 2. The integrated control unit 22 is composed of an integrated control module and provides integrated control of the train timetable planning unit 30, the vehicle operation planning unit 40, and the crew operation planning unit 50. The integrated control unit 22 optimizes the creation of each plan by the transport plan creation device 100. Details of the integrated control will be described later.

[0022] Route DB23 stores railway route information. Route information includes information on the railway track network, stations, and train depots. Vehicle DB24 stores information on vehicles used in train operations. Vehicle information includes the number of seats and passenger capacity of each vehicle, the distance they can travel, and the area they operate in. Crew DB25 stores information on the crew of the railway company. Crew information includes information on crew members who can operate the vehicles, such as age, experience, type of work (driver, conductor, etc.), and working conditions. Route DB23, Vehicle DB24, and Crew DB25 are composed of DB14 as shown in Figure 2.

[0023] The train schedule planning unit 30 is composed of, for example, an AI module for train schedule planning, and creates train schedules. A train schedule is a diagram that shows the railway operation plan and operating status, and is also called an operation diagram or timetable. Figure 4 schematically shows an example of a train schedule. A train schedule is a graph that shows time on the horizontal axis and stations on the vertical axis. Each line on the graph corresponds to a train, and it shows the operation of each train, such as when each train departs from which station, when it arrives at which station, and how long it stops at each station. The train schedule planning unit 30 takes into account the planning conditions entered from the user terminal 2, creates a train schedule based on the route information stored in the route DB 23 and the vehicle information stored in the vehicle DB 24, and outputs it to the vehicle operation planning unit 40 and the integrated control unit 22. Specific examples of how to create a train schedule will be described later.

[0024] The Vehicle Operation Planning Unit 40 is composed of, for example, an AI module for vehicle operation planning, and creates a vehicle operation plan. A vehicle operation plan is part of the transportation plan that a railway company creates in advance for train operation, and it is a plan that determines how to use the vehicles. Figure 5 schematically shows an example of a vehicle operation table that shows a vehicle operation plan. The vehicle operation table is a graph that shows time on the horizontal axis and stations on the vertical axis. Each line on the graph corresponds to a train, and it specifies when and where each train runs, where it stops, when and from which train depot it departs, and when and to which train depot it returns. The Vehicle Operation Planning Unit 40 creates a vehicle operation plan based on the train timetable input from the Train Timetable Planning Unit 30, the route information stored in the Route DB 23, and the vehicle information stored in the Vehicle DB 24, and outputs it to the Crew Operation Planning Unit 50 and the Integrated Control Unit 22. Specific examples of how to create a vehicle operation plan will be described later.

[0025] The crew operation planning unit 50 is composed of, for example, an AI module for crew operation planning and creates a crew operation plan. The crew operation plan shows the assignment of crew members (drivers, conductors, etc.) to each train. The history of crew members' movements (boarding and alighting) for each train is also called a "route," and crew members perform their duties according to their assigned route. Figure 6 shows an example of a route table that shows a crew operation plan. In the route table, stations are shown on the horizontal axis and times are shown on the vertical axis. The line segments in the route table specify when and at which station the crew member boards the train, when and at which station they alight from that train, when and for how long they wait at that station, and when they board the next train. The crew operation planning unit 50 creates a crew operation plan based on the vehicle operation plan input from the vehicle operation planning unit 40 and the crew information stored in crew member D25, and outputs it to the integrated control unit 22. Specific examples of how to create a crew deployment plan will be discussed later.

[0026] The integrated control unit 22 automatically adjusts the train schedule obtained from the train schedule planning unit 30, the vehicle operation plan obtained from the vehicle operation planning unit 40, and the crew operation plan obtained from the crew operation planning unit 50. The integrated control unit 22 uses, for example, an integrated control module, which is a pre-trained AI module, to automatically adjust the train schedule, vehicle operation plan, and crew operation plan. The integrated control module is a higher-level module than the train schedule planning module, the vehicle operation planning module, and the crew operation planning module, and automatically adjusts the targets of each planning module based on the plans created by each planning module, the requests and demands entered by the user, and the degree of vehicle congestion that is the premise of each plan.

[0027] For example, the integrated control unit 22 modifies parts of the train schedule, vehicle operation plan, and crew operation plan based on user preferences. User preferences refer to the user's requests and specifications regarding the satisfaction level and achievement level of each plan in relation to the plan conditions, the priority of items to be considered when creating each plan, etc. For example, if the degree of achievement of the plan conditions is insufficient, the user can request that the degree of achievement of the plan conditions be increased to a certain target value. The user can also specify, for example, whether to prioritize measures to increase the number of trains running or measures to increase the train length (the number of cars that make up a train) in order to achieve the goal of reducing congestion, that is, which of the items, the number of trains running or the train length, should be prioritized.

[0028] Based on this change, other plans will need to be modified. For example, the integrated control unit 22 instructs the train timetable planning unit 30 to modify part of the train timetable in accordance with the user's request to "reduce congestion." The integrated control unit 22 also instructs the vehicle operation planning unit 40 to modify part of the vehicle operation plan to correspond to the modified train timetable. Furthermore, the integrated control unit 22 instructs the crew operation planning unit 50 to modify part of the vehicle operation plan to correspond to the modified vehicle operation plan. In this way, the integrated control unit 22 modifies parts of the train timetable, vehicle operation plan, and crew operation plan in accordance with the user's input. If the intended change cannot be made in any of the plans, the integrated control unit 22 instructs another planning unit to make a different change. In this way, the integrated control unit 22 repeatedly modifies parts of each plan as needed, and when a change (solution) that satisfies the user's input is obtained, the adjustment is terminated. This makes it possible to optimize each plan while simultaneously considering limited resources such as routes, vehicles, and crew, as well as user demand. Specific examples of integrated control will be described later.

[0029] [Train Schedule Planning Unit] Next, the train schedule planning unit will be explained in detail. Figure 7 shows the functional configuration of the train schedule planning unit 30. As shown in Figure 7, the train schedule planning unit 30 includes a setting unit 31, a layout unit 32, an output unit 33, and a layout storage unit 35.

[0030] The setting unit 31 first obtains setting information for creating train timetables from the route DB 23. The setting information includes information about the location and section on which the train runs, such as stations and the distance between stations, and information about the time period for which the train timetable is to be created. It is assumed that the section on which the train runs is an exclusive section, such as a single-track section. On the other hand, the single-track section is between stations, and the stations are located on double-track sections and are not exclusive sections.

[0031] The setting unit 31 prepares to create a train timetable based on the acquired setting information. Specifically, the setting unit 31 generates a coordinate space for the train timetable, with time (time) on the horizontal axis and station (location) on the vertical axis, as shown in Figure 8(A), for example. The setting unit 31 also sets cards corresponding to trains traveling between stations in the coordinate space of the train timetable. A card is a rectangular figure that graphically represents the planned position and time of the train's movement, and includes a timetable line represented by the diagonal within the rectangular figure corresponding to the train's movement. As an example, as shown in Figure 8(A), a card can be placed between each station, with the width representing the travel time and the height representing the distance between stations. The timetable line located on the diagonal of the card is formed to extend from the coordinates of (departure time, departure station) to the coordinates of (arrival time, arrival station). In this way, the height of the card corresponds to the distance between stations, which is the single-track section on which the train travels, and the width corresponds to the time the train travels between stations. Therefore, the area represented by the card represents the area occupied by the corresponding train between stations on a single-track section, and thus represents the occupied area within the coordinate space. In other words, the real space corresponding to the time period in which the card is placed within the coordinate space is inaccessible to other trains.

[0032] The placement unit 32 places cards corresponding to trains in the coordinate space of the train timetable. Specifically, the placement unit 32 first places card C1 corresponding to the first train, as shown in Figure 8(A). At this time, the placement unit 32 places one card C1 between each station. In the example in Figure 8(A), it shows the placement of card C1 for a train that departs from "Station 0" at "Time 04:30" and arrives at "Station 6" at "Time 04:48".

[0033] Next, as shown in Figure 8(B), the placement unit 32 places cards C2 corresponding to other trains that intersect with the first train at a station. Specifically, the placement unit 32 places cards C2 so that the other train departs in the opposite direction at the station and time where the first train arrives and departs. Note that the spaces between the cards C1 of the first train, that is, the spaces between the vertices of each card C1, are located at the departure and arrival stations, and since the stations are double-track sections, these are areas not occupied by the first train. Therefore, between the cards, i.e., at the stations, when the first train is stopped, the other train can also stop and intersect so that they can depart in opposite directions. When placing cards C2 for other trains, the placement unit 32 selects multiple stations evenly from the station where the first train arrives and departs, and places cards C2 so that other trains depart from each of the selected stations. In this way, the placement unit 32 places cards C2 corresponding to other trains that intersect with the first train at a station in a leveled manner within the coordinate space of the train timetable. In the example in Figure 8(B), cards C2 for other trains that cross paths with the arrival and departure times of the first trains and depart in the opposite direction are placed at each of "Station 2," "Station 3," "Station 5," and "Station 6," showing that four different train cards C2 are placed at each station.

[0034] Next, as shown in Figure 9(A), the placement unit 32 places cards C3 corresponding to other trains based on the movement status of the first train and other trains, that is, the placement of card C1 for the first train and card C2 for the other train. For example, the placement unit 32 places cards C3 corresponding to other trains so that they depart in the same direction as the first train, with a predetermined time interval from the time the first train departs. At this time, the placement unit 32 further places cards C3 so that other trains intersect at some stations and times where the other train (card C2) arrives and departs. The placement unit 32 also places cards C3 corresponding to other trains in the same manner as described above, with a further predetermined time interval. For example, the placement unit 32 places cards C3 for multiple other trains in a pattern, such as leaving a certain interval from the first train or arranging for them to arrive and depart at each station at regular times. In the example in Figure 9(A), the following cards C3 are placed: another train C3 that departs from "Station 0" at "Time 04:36" and crosses with another train at "Station 5" at "Time 04:52"; another train C3 that departs from "Station 0" at "Time 04:48" and crosses with another train at "Station 3" at "Time 04:57"; and another train C3 that departs from "Station 0" at "Time 04:55" and crosses with another train at "Station 2" at "Time 05:02". In this example, three cards C3 for other trains are placed. Note that in this example, the cards for other trains are placed only in a portion of the section on the departure station side.

[0035] Next, as described above, when the placement unit 32 places the card C3 of another train, if the card C3 of the other train overlaps with other cards C1 and C2, it moves the cards in coordinate space so that the cards do not overlap. At this time, if there are multiple places where cards overlap, the placement unit 32 moves the cards sequentially to resolve the overlaps, starting from the places where the overlap occurred earlier. Furthermore, at this time, the placement unit 32 moves one of the overlapping cards toward the later time to resolve the overlap. Note that the placement unit 32 is not limited to moving the cards; it may also change the width of the cards, for example. In other words, since the width of the cards corresponds to the time it takes for a train to travel between stations, the placement unit 32 may also change the travel time between stations by changing the speed of the train within the speed limit. In this way, the placement unit 32 sets the cards in coordinate space to resolve the overlaps by moving the placement of one of the overlapping cards or changing the width of the cards.

[0036] In the example shown in Figure 9(A), overlapping cards occur in multiple locations. At this time, card C3 of another train departing from "Station 0" at time "04:36" overlaps with card C2 of another train at location R1. Since this location is the earliest time, card C3 of the other train is shifted to a later time to eliminate the overlap between cards. Accordingly, the placement unit 32 sequentially shifts other cards C3 corresponding to the same other train to later times, and if further overlapping occurs as a result, the overlap is eliminated in the same manner as described above. As a result, as shown in Figure 9(B), the overlap of card C3 of the train departing from "Station 0" at time "04:36" is eliminated. Subsequently, as shown by R2 in Figure 9(B), there is further overlapping of cards at a later time, so the placement unit 32 moves the card of the train corresponding to this card to a later time, as described above, to eliminate the overlap between cards as shown in Figure 10(A). Next, as shown by the symbol R3 in Figure 10(A), since overlapping of cards occurs at a later time, the placement unit 32 moves the train card corresponding to the overlapping card to a later time, as described above, to eliminate the overlapping of cards as shown in Figure 10(B).

[0037] Furthermore, when resolving the overlapping of cards as described above, the arrangement unit 32 may move cards as appropriate to eliminate unnecessary stopping time at each station for each train. For example, as shown in Figure 10(B), a train departing from "Station 0" at time "04:36" has unnecessary stopping time at the next station "Station 1". Therefore, the arrangement unit 32 moves the corresponding card C3 as shown in Figure 11(A) to shorten the stopping time at "Station 1" for such a train. In the example described above, the overlapping of cards is resolved by setting the card to move, but the overlapping of cards may also be resolved by setting the card width to change.

[0038] Next, the placement unit 32 places cards C3 for the continuation of other trains, since the train timetable is not yet complete and the placement of cards is not finished. For example, as shown in Figure 11(B), for cards C3 of other trains, it places additional cards C3 for the continuation of the section from the point where it intersects with card C2 of another train. Then, similar to the card C2 of another train described above, the placement unit 32 places cards C4 corresponding to yet another train that intersects with the other train at a station, as shown in Figure 12(A). At this time, if the placement of card C4 causes overlapping of cards, the placement unit 32 moves the cards sequentially from the earlier time zones toward the later time zones to resolve the overlap, as described above. For example, the placement unit 32 resolves the overlapping of cards sequentially from the situation shown in Figure 12(A) to Figure 12(B) and Figure 13(A). Subsequently, if the arrangement unit 32 determines that the overlapping of cards has resulted in unnecessary stopping time at each station for each train, it moves the corresponding cards as shown in Figure 13(B) to shorten the stopping time. In addition, whenever the arrangement unit 32 arranges or moves cards as described above, it always stores the arrangement of the cards within the coordinate space of the train timetable in the arrangement storage unit 35.

[0039] Once the desired train schedule is complete and the cards have been placed, the output unit 33 creates and outputs the train schedule based on the placement of the cards in the final train schedule. The output unit 33 may also extract the departure and arrival times for each station for each train from the card placement and the schedule within the cards, and create and output this as train timetable data.

[0040] [Vehicle Operation Planning Department] Next, the Vehicle Operation Planning Department will be explained in detail. Vehicle operation plans are created to meet the constraints on the number of vehicles at stations and between stations. At this time, vehicle operation plans may also be created to minimize the cost of vehicles in accordance with the operator's requests, etc. In the following example, vehicle operation plans will be created taking into consideration the cost of vehicles. In this disclosure, a "train set" will be used to refer to a single or multiple vehicles, and a train set that operates between stations will be used to refer to a "train."

[0041] Figure 14 shows the functional configuration of the vehicle operation planning unit 40. As shown in Figure 14, the vehicle operation planning unit 40 comprises a graph construction unit 41, a solution unit 42, an allocation unit 43, and a graph storage unit 45.

[0042] The graph construction unit 41 first obtains the train schedule, constraints, and operational requirements created by the train schedule planning unit 30. The train schedule is information in the form of a diagram (timetable) that represents the train operation plan, and is represented, for example, as shown in the left diagram of Figure 15.

[0043] Constraints are information representing the limitations on vehicles on the tracks at and between stations, and are obtained, for example, from the route database DB23. For example, constraints may include information such as the number of vehicles that can wait at each station or the number of vehicles that can be parked overnight, or information such as the number of vehicles that can enter each track. Note that constraints are not limited to the number of vehicles at stations or between stations, but may also include constraints on the type of vehicles.

[0044] Operational requests are information representing train-related requests from train operators. For example, operational requests may include information such as the maximum number of cars, desired number of cars, and minimum number of cars for each station and each time period. Operational requests may also include the desired number of cars for each station regardless of time, and may also include requests for the types of cars used between stations.

[0045] The graph construction unit 41 constructs a graph using the train timetable, constraints, and operation requirements. In this graph, station information representing stations for each time period is used as nodes, and information representing the movement of vehicles between stations is associated with these nodes as edges. The edge represents at least one of the trains between stations or the formation (hereinafter referred to as "waiting formation") that waits (or stays) at a station across time periods. Constraints such as the number of vehicles in the formation that can wait at a station and the number of vehicles in a train between stations may be associated with the edge. As an example, the graph construction unit 41 constructs a graph as shown in the right diagram of FIG. 15 from the train timetable shown in the left diagram of FIG. 15 and the above-mentioned constraints and operation requirements. Here, an example of constructing the graph will be described with reference to FIG. 15.

[0046] First, in the train timetable in the left diagram of FIG. 15, the train indicated by the arrow within symbol T0' is the train arriving at Station B (hereinafter referred to as "arriving train"), and the train indicated by the arrow within symbol T0" is the train departing from Station B (hereinafter referred to as "departing train"). Also, the group of trains indicated by the arrow within symbol T1' represents the trains arriving at Station B. In such a case, each station related to "arriving train T0' at Station B → departing train T0" from Station B" is represented as one node at time T0. As a result, as shown in the right diagram of FIG. 15, Station A, Station B, and Station C at time T0 are represented as nodes A_0, B_0, and C_0 respectively, and Station B at time T1 is represented as node B_1. In addition, for each node corresponding to each station at each time, edges represented by arrows corresponding to each train in operation are connected. Specifically, when there is a train running between stations at each time, the nodes corresponding to each station at each time are connected by an edge represented by an arrow corresponding to the running train. Also, when a formation can wait at each station at each time, an edge represented by a dotted arrow corresponding to the waiting of the formation at the station is connected to the node corresponding to each station at each time.

[0047] Then, the graph construction unit 41 sets, based on the constraints and operation requirements, the capacity representing the constraints on the number of vehicles at stations and between stations, and the cost per vehicle, for the edges connected to the nodes as described above.

[0048] Capacity represents the constraints when creating a vehicle plan. For example, the capacity is a value set based on constraints. When an edge represents a train running between stations, the maximum number of vehicles that can run between such stations is set. In this case, the capacity may represent the maximum number of vehicles determined by factors such as the number of tracks. For example, when an edge represents the waiting of formations at a station, the capacity is set to the maximum number of vehicles that can wait at such a station. In this case, the capacity may represent physical constraints such as the number of platforms for waiting formations at the station. The capacity may represent constraints based on the operation schedule. In this case, the capacity may represent, for example, the maximum number of vehicles that can operate during a specific time period based on the departure and arrival times set during that time period or the density of the train timetable. The capacity may represent safety and operation constraints. In this case, the capacity may be a value set based on safety criteria for ensuring safe train operation on tracks and stations. The capacity may represent the constraints for detaining vehicles at a station from the time of the last train of a day to the time of the first train of the next day. In this case, the capacity may represent, for example, in addition to the number of platforms at the station, the number of vehicles that can be detained in the vehicle depot at the station.

[0049] Cost is set for an edge and represents a numerical value related to the evaluation of vehicle operation. For example, the cost represents the cost incurred when a vehicle runs between stations or the cost incurred when a vehicle waits at a station. At this time, the cost is set to a smaller value as the operation requirement desires more vehicle operation at stations and between stations. As an example, for the cost between stations, the value is set small or even negative in sections where vehicle running is desired. Therefore, in a time period or between stations where a large number of passengers are expected and the desired number of vehicles etc. in the operation requirement is a large value, the cost is set to a small value. Also, when a vehicle waits at a station, if a large number of vehicle uses are desired afterwards, the cost value is small, and if vehicle use is not desired or the number of vehicles that can wait is small afterwards, the cost value is set to a large value.

[0050] Furthermore, the cost may represent the expenses required to operate the train. In this case, the cost represents economic expenses such as electricity, maintenance, and fuel costs associated with operating the train. The cost may also represent expenses based on the demand for the train. In this case, the cost is set according to passenger demand in a particular time period or section. For example, when demand is high, train allocation is prioritized, so the cost is set low, and when demand is low, the cost is set high. The cost may also represent expenses based on the characteristics of the vehicle. In this case, for example, when maintenance is required more frequently in operation, the cost is set high, and when maintenance is required less frequently, the cost is set low. The cost may also be set according to the intention to operate the train. In this case, for example, the cost is set low when there is a sudden increase in demand, such as during an event, and high under normal circumstances.

[0051] As described above, the graph construction unit 41 constructs a graph, as shown in the right-hand diagram of Figure 15, by connecting nodes representing stations for each hour with edges representing the number of train cars waiting at stations, constraints on the number of train cars between stations, and the cost of the cars. At this time, nodes representing virtual start and end points may be set at the beginning and end of the day to construct a group of graphs representing the train schedule for the entire day, or the train schedule may be divided into predetermined time periods, and a group of graphs may be constructed for each time period. The graph construction unit 41 then stores the constructed graphs in the graph storage unit 45.

[0052] The problem-solving unit 42 (creation unit) uses the graph constructed as described above to calculate the number of vehicles to operate in order to satisfy the constraints set for stations and the spaces between stations. At this time, the problem-solving unit 42 calculates the number of vehicles to operate in order to minimize the cost for the entire group of graphs. In other words, the problem-solving unit 42 solves the minimum cost flow problem in the graph.

[0053] Here, Figure 16 shows an example of a graph constructed by the graph construction unit 41 and solved by the solution unit 42. In this graph, node 1 is the starting point and node 5 is the ending point, and each node (1 to 5) represents a station at each time point. Each node is connected to an edge corresponding to the train being operated, and each edge is further assigned the aforementioned "capacity" and "cost". In this case, the "flow rate" shown in Figure 16 represents the number of operational "vehicles". Therefore, in the example of Figure 16, the problem is to minimize the cost across the entire graph and allocate "number of vehicles = 10" while satisfying the capacity constraint set for each edge. Specifically, the solution unit 42 calculates the allocation of the number of vehicles (shown as "?" in Figure 16) at each edge such that it does not exceed the maximum number of vehicles, which is the "capacity", and the sum of the values ​​obtained by multiplying the "cost" by the number of vehicles is minimized.

[0054] For example, the edge connecting node 1 and node 2 is set to "Capacity: 5" and "Cost: 3". Therefore, at this edge, the number of vehicles, which is the "flow rate," will be "5" or less, and the cost will be calculated as "number of vehicles × 3 (cost)". Similarly, at all edges, the calculation of "number of vehicles × cost" is performed so that the "flow rate" does not exceed the "capacity" of the number of vehicles, and the allocation of the number of vehicles, which is the "flow rate," at each edge is calculated to minimize the total cost.

[0055] The allocation unit 43 assigns train sets to each train in the train timetable so that the number of trains is the same as the number of trains calculated by the solving unit 42 as described above. In other words, the number of trains used in each train in the train timetable is the number of trains calculated. Then, as a result of assigning train sets to each train as described above, the allocation unit 43 creates and outputs a train operation plan consisting of a series of trains operated by each train set.

[0056] Furthermore, the allocation unit 43 may control the configuration of each train according to the number of vehicles determined. For example, the allocation unit 43 may control the configuration of trains in the station depot and on the tracks according to the number of vehicles determined by allocating multiple train sets to the same train so that the total number of vehicles is equal to the number of vehicles determined.

[0057] [Crew Operation Planning Unit] Next, the crew operation planning unit will be explained in detail. Figure 17 shows the functional configuration of the crew operation planning unit 50. As shown in Figure 17, the crew operation planning unit 50 comprises an extraction unit 51, a learning unit 52, and a simulator storage unit 53.

[0058] The Crew Operations Planning Department 50 creates a crew operations plan using a simulator that reproduces the movement of crew members. This simulator is created through prior training. Based on the vehicle operations plan, the simulator determines which crew members are assigned to which trains. The basic idea is to assume that trains are moving according to a given vehicle operations plan and to assign crew members to each vehicle. For example, if a train must have a driver and a conductor on board, the simulator will assign crew members so that each train always has a driver and a conductor on board. Also, crew members do not get on or off at all stations; the stations where crew members get on and off are predetermined. In reality, the stations where crew members get on and off are limited to stations with rest rooms, etc. In addition, there is usually a rule that each crew member leaves work at the same station where they reported to work in the morning. Furthermore, there are constraints on working hours for each crew member. The rules and constraints regarding each crew member's work style and working hours are stored in the Crew DB 25.

[0059] Therefore, the simulator can be thought of as playing a game in which, under the rules and constraints regarding the crew, it determines which crew members board and alight at which stations, and when, according to a given train operation plan. Accordingly, the crew operation planning unit 50 extracts the status of each train from the train operation plan and also extracts the timing for deciding on the boarding and alighting of crew members at specific stations (hereinafter referred to as the "decision timing"). Then, by learning the game of deciding on the boarding and alighting of crew members at the decision timing, it creates a pre-trained simulator.

[0060] The extraction unit 51 extracts decision timings and train statuses based on the train schedule and the vehicle operation plan. Figure 18 shows one example of extracting decision timings and train statuses. The train schedule is a timetable showing the departure and arrival times at each station for each train with a train number. The vehicle operation plan, on the other hand, is a list showing which trains a particular train set will be responsible for from the time it leaves the depot until it returns to the depot. The extraction unit 51 extracts vehicle operations such as entering and leaving the depot and turning around from the vehicle operation plan, and extracts only the departure and arrival times for stations where crew members can board and alight from the train schedule. The extraction unit 51 then correlates the "departure and arrival times at stations where crew members can board and alight for each train" with the status of the train, such as whether or not it "enters the depot or turns around" at that station. As a result, the extraction unit 51 creates the decision timing list shown in Figure 18. The decision timing list shows the time step when considering only one vehicle operation. Theoretically, crew members cannot take action outside of stations where they are permitted to board or alight, so the decision timing list can skip time periods outside of those stations. In other words, the decision timing list will only show the timings for stations where crew members are permitted to board or alight. Furthermore, the decision timing list can limit the actions (possible state transitions) that crew members can take. For example, if there is a rule that "the crew must continue their duties when turning around," then the action can be determined accordingly.

[0061] Figure 19 shows another example of extracting decision timings and train status. In this example, the extraction unit 51 associates train timetables with vehicle operation plans, merges each event of each associated operation across all trains (all lines), sorts by time, and creates a decision timing list. Each time in the created decision timing list becomes the decision timing for the action.

[0062] The learning unit 52 learns a simulator that determines the crew's actions (state transitions) at each decision timing indicated by the created decision timing list. Specifically, the simulator is thought to play a game at each decision timing indicated by the decision timing list, where it decides which crew member or crew pair to board, alight, or wait, based on how many crew members are at the boarding / alighting stations. Figure 20 shows an example of the display screen for this game. The simulator plays a game at each boarding / alighting station (stations A to C in this example) to decide the actions (boarding / alighting) of a crew member or crew pair.

[0063] The learning unit 52 learns the above simulator using reinforcement learning. In this case, the action in reinforcement learning is the decision of whether to board or disembark crew members at the decision timing. The reward function is set such that, for example, a higher score is given for fewer crew members on duty, or a penalty of -10,000 points is given if a train with no crew members occurs. As a result, when the learning unit 52 receives a game screen (observation) like the one in Figure 20, it pre-trains a policy function to determine the action that maximizes the reward function. The simulator operates according to the policy function obtained through learning. The learning unit 52 stores the pre-trained simulator in the simulator memory unit 53.

[0064] In the example above, the simulator is created using reinforcement learning, but it could also be created using a rule-based approach. In this case, a rule such as "in this situation (observation), take this action" could be prepared for every situation, and a policy could be created based on these rules to build the simulator.

[0065] In actual crew operation planning, the crew operation planning unit 50 reads a pre-created simulator from the simulator storage unit 53, operates the simulator based on the input train schedule and vehicle operation plan, and creates and outputs an optimal crew operation plan. Figure 20(B) shows an example of a crew operation plan created by the simulator. Each route 1 to 5 indicates the route of the corresponding crew member. Actions such as boarding and alighting by crew members are associated with each decision timing indicated by the triangular mark "▲". When each crew member's route is displayed individually, the route table shown in Figure 6 is obtained.

[0066] [Transportation Plan Creation Process] Next, the transportation plan creation process by the transportation plan creation device 100 will be described. Figure 21 is a flowchart of the transportation plan creation process. This process is realized when the processor 11 shown in Figure 2 executes a pre-prepared program and operates as each element shown in Figure 3.

[0067] First, the transportation plan creation device 100 obtains planning conditions from the user (step S11). Next, the train timetable planning unit 30 creates a train timetable based on the obtained planning conditions (step S12). Next, the vehicle operation planning unit 40 creates a vehicle operation plan based on the created train timetable (step S13). Next, the crew operation planning unit 50 creates a crew operation plan based on the created vehicle operation plan (step S14).

[0068] Next, the integrated control unit 22 adjusts the created train schedule, vehicle operation plan, and crew operation plan (step S15). Specifically, the integrated control unit 22 determines whether any of the train schedule, vehicle operation plan, or crew operation plan needs to be modified based on the requests and demands entered by the user and the goals of each plan (step S16). If the integrated control unit 22 determines that any of the plans needs to be modified (step S16: Yes), it returns to step S12, S13, or S14 and causes the plan that needs modification to be created again. For example, if it determines that the train schedule needs to be modified to alleviate congestion, the integrated control unit 22 gives a modification instruction to the train schedule planning unit 30 and causes the train schedule to be modified. Once the new train schedule is created, the vehicle operation planning unit 40 creates a new vehicle operation plan based on the modified train schedule, and the crew operation planning unit 50 creates a new crew operation plan based on the new vehicle operation plan. In this way, if any of the plans need to be modified, that plan and other related plans are modified. If no modifications are needed for any of the plans (Step S16: No), the integrated control unit 22 outputs the transportation plan, including the three created plans, to the user terminal 2 (Step S17). The transportation plan creation process then ends.

[0069] <Second Embodiment> Figure 22 is a block diagram showing the configuration of a transportation plan creation device according to the second embodiment. The transportation plan creation device 70 includes an operation plan creation means 71, a vehicle operation plan creation means 72, a crew operation plan creation means 73, and an adjustment means 74.

[0070] Figure 23 is a flowchart of the processing performed by the transportation plan creation device 70. The operation plan creation means 71 creates a train operation plan based on the planning conditions (step S71). The vehicle operation plan creation means 72 creates a vehicle operation plan based on the train operation plan (step S72). The crew operation plan creation means 73 creates a crew operation plan based on the vehicle operation plan (step S73). The adjustment means 74 adjusts the train operation plan, the vehicle operation plan, and the crew operation plan based on the user's preferences (step S74).

[0071] According to the above-described transportation plan creation device 70, it is possible to optimize each plan as a whole by adjusting individual plans.

[0072] Some or all of the above embodiments may also be described as follows, but are not limited to the following:

[0073] (Note 1) A transport planning device comprising: a means for creating a train operation plan based on planning conditions; a means for creating a vehicle operation plan based on the train operation plan; a means for creating a crew operation plan based on the vehicle operation plan; and an adjustment means for adjusting the train operation plan, the vehicle operation plan, and the crew operation plan based on user preferences.

[0074] (Note 2) The transportation planning device according to Note 1, wherein the adjustment means modifies at least one of the train operation plan, the vehicle operation plan, and the crew operation plan at least partially, and modifies the other plans based on the modified plan.

[0075] (Note 3) The transportation planning device according to Note 2, wherein the adjustment means modifies the at least one plan with respect to at least one of the satisfaction level of the plan, the priority of items to be considered in creating the plan, and the plan conditions.

[0076] (Note 4) The crew operation plan creation means is a transport plan creation device according to any one of Notes 1 to 3 that creates the crew operation plan using a simulator created in advance by reinforcement learning.

[0077] (Note 5) The crew operation plan creation means extracts a decision timing for determining whether or not a crew state transition is necessary based on the train operation plan and the vehicle operation plan, and reinforces the simulator to determine an action to transition the crew state at the decision timing. (Note 4)

[0078] (Note 6) The transportation planning device according to Note 5, wherein the crew operation plan creation means extracts the timing when the train arrives at a station where crew members can board or alight as the decision timing, based on the train operation plan and the vehicle operation plan, and reinforces the simulator so that at the decision timing, the boarding or alighting of crew members at the station where crew members can board or alight is determined as the action.

[0079] (Note 7) The operation plan creation means is a transportation plan creation device according to any one of Notes 1 to 6 that creates an operation plan for the train in order to reduce the degree of congestion in each time period and each section.

[0080] (Note 8) The transportation planning device according to any one of Notes 1 to 6, wherein the adjustment means adjusts the train operation plan to improve the parts of the vehicle operation plan in which the operation of the vehicles is inefficient.

[0081] (Note 9) A method for creating a transportation plan performed by a computer, comprising: creating a train operation plan based on planning conditions; creating a vehicle operation plan based on the train operation plan; creating a crew operation plan based on the vehicle operation plan; and adjusting the train operation plan, the vehicle operation plan and the crew operation plan based on the user's preferences.

[0082] (Note 10) A recording medium that records a program that causes a computer to perform the following processes: create a train operation plan based on planning conditions; create a vehicle operation plan based on the train operation plan; create a crew operation plan based on the vehicle operation plan; and adjust the train operation plan, the vehicle operation plan, and the crew operation plan based on user preferences.

[0083] Furthermore, some or all of the configurations described in Appendices 2 to 8, which are subordinate to Appendice 1 above, may also be subordinate to Appendices 9 and 10 in the same way as those described in Appendices 2 to 8. Moreover, not limited to Appendices 1, 9, and 10, some or all of the configurations described as appendices may also be subordinate to various hardware, software, various recording means for recording software, or systems, without departing from the embodiments described above.

[0084] Although the present disclosure has been described above with reference to embodiments and examples, the present disclosure is not limited to the above embodiments and examples. Various modifications to the structure and details of the present disclosure can be understood by those skilled in the art within the scope of the present disclosure.

[0085] 1 Transportation planning system 2 User terminal 11 Processor 22 Integrated control unit 23 Route DB 24 Vehicle DB 25 Crew DB 30 Train timetable planning unit 40 Vehicle operation planning unit 50 Crew operation planning unit 100 Transportation planning device

Claims

1. A transportation planning device comprising: operation plan creation means for creating a train operation plan based on planning conditions; vehicle operation plan creation means for creating a vehicle operation plan based on the train operation plan; crew operation plan creation means for creating a crew operation plan based on the vehicle operation plan; and adjustment means for adjusting the operation plan, the vehicle operation plan, and the crew operation plan based on user preferences.

2. The transportation planning device according to claim 1, wherein the adjustment means modifies at least one of the train operation plan, the vehicle operation plan, and the crew operation plan at least partially, and modifies the other plans based on the modified plan.

3. The transportation planning device according to claim 2, wherein the adjustment means modifies the at least one of the following: satisfaction with the plan, priority of items to be considered in creating the plan, and the plan conditions.

4. The transportation planning device according to claim 1, wherein the crew operation plan creation means creates the crew operation plan using a simulator created in advance by reinforcement learning.

5. The transportation planning device according to claim 4, wherein the crew operation plan creation means extracts a decision timing for determining whether or not a crew member's state transition is necessary based on the train operation plan and the vehicle operation plan, and reinforces the simulator to determine an action to transition the crew member's state at the decision timing.

6. The transportation planning device according to claim 5, wherein the crew operation plan creation means extracts the timing at which the train arrives at a station where crew members can board or alight as the decision timing, based on the train operation plan and the vehicle operation plan, and reinforces the simulator to determine the boarding or alighting of crew members at the station where crew members can board or alight as the action at the decision timing.

7. The transportation planning device according to claim 1, wherein the means for creating the operation plan creates an operation plan for the train in order to reduce the degree of congestion in each time period and each section.

8. The transportation planning device according to claim 1, wherein the adjustment means adjusts the train operation plan to improve any inefficient parts of the vehicle operation plan.

9. A method for creating a transportation plan performed by a computer, comprising: creating a train operation plan based on planning conditions; creating a vehicle operation plan based on the train operation plan; creating a crew operation plan based on the vehicle operation plan; and adjusting the train operation plan, the vehicle operation plan, and the crew operation plan based on user preferences.

10. A recording medium that contains a program that causes a computer to perform the following processes: create a train operation plan based on planning conditions; create a vehicle operation plan based on the train operation plan; create a crew operation plan based on the vehicle operation plan; and adjust the train operation plan, the vehicle operation plan, and the crew operation plan based on user preferences.

Images