JP2005259052A - Program and vehicle operation program creation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To automatically create a vehicle operation program surely allocating a test to be carried out on a vehicle for satisfying a test period and an operation condition and capable of efficiently operating the vehicle. <P>SOLUTION: This vehicle operation program creation part 120 assumes a frequency of check-up tests minimally necessary for satisfying the test period. The vehicle operation program creation part 120 selects an arc capable of executing a vehicle test during an interval of the connection of a train represented by the arc among arcs constituting a train operation network for allocating the respective vehicle tests to the vehicle operation network. A circuit path search part 124 searches for a circuit path, which passes all nodes only once and includes all the selected checkable arcs, and outputs it as a vehicle operation program. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a program for creating a vehicle operation plan and a vehicle operation plan creation device.

  In the railway, a plan called a vehicle operation plan is defined in order to realize the created operation schedule (train schedule). This vehicle operation plan is a plan related to vehicle operation including vehicle use order (train connection order) and maintenance management for the vehicle, and is created in consideration of various constraints.

  For example, vehicles are obliged to carry out inspections according to their travel time and distance for maintenance, but they are called work inspections and alternation inspections, especially when creating a vehicle operation plan. Vehicle inspection must be considered. Each inspection defines an inspection cycle (inspection cycle) and the content of the inspection. Specifically, in the work inspection, visual inspection of the vehicle is performed in about one hour, and in the alternating inspection, for example, inspection regarding the function of the vehicle such as a brake device, an electric device, and a traveling device is performed for about half a day. Is done. In addition, each inspection is usually performed during the daytime in a vehicle base equipped with facilities necessary for the inspection and installed in a predetermined station or the like. Note that the inspection content of the alternating inspection satisfies the inspection content of the industrial inspection, and the industrial inspection is performed simultaneously with the alternating inspection.

  For this reason, the vehicle operation plan must be created in consideration so that the work inspection and the alternating inspection are performed on the vehicle without exceeding the set inspection cycle. Furthermore, at this time, it is necessary to set the execution position of each inspection in consideration of the place where the inspection can be performed, the time zone, the required time, etc. (hereinafter referred to as “execution conditions”).

  In addition to this, it is necessary to consider various restrictions such as predetermined connection conditions when connecting trains and securing passenger transportation.

  On the other hand, since the operation of the vehicle affects the cost related to the railway operation, there has been a demand for cost reduction for the created vehicle operation meter. Specifically, for example, it is desirable to create a vehicle operation plan so that the number of alternating inspections and work inspections is minimized, and to use the vehicle efficiently.

By the way, a technology for automatically creating a vehicle operation plan that satisfies these constraints has been developed, and the applicant of the present application also connects the vehicle operation plan to all trains that make up the train diagram cyclically. Considering the creation of a vehicle operation plan as a traveling salesman problem with a single connection pattern, the vehicle operation plan is represented by a circuit that is best evaluated in accordance with the evaluation criteria set in advance based on the above-mentioned constraints. Is used to automatically create a vehicle operation plan (see Patent Document 1).
JP 2003-154939 A

  However, when a vehicle operation plan is created using the technique of Patent Document 1 described above, there are the following problems. That is, in Patent Document 1, a vehicle operation plan in which each inspection is assigned by changing the inspection circuit and the execution condition of the inspection inspection in advance by assigning the execution position of the alternating inspection in advance and changing the searched circuit to satisfy the inspection cycle and execution conditions of the work inspection. However, depending on the original train schedule, there is a case where it is difficult to obtain a proper vehicle operation plan because the work inspection may not be properly assigned to the circuit.

  Therefore, the present invention is a vehicle operation plan in which an inspection to be performed on a vehicle is surely assigned so as to satisfy the inspection cycle and execution conditions, and the vehicle operation plan can efficiently operate the vehicle. The purpose is to create automatically.

A first invention for solving the above-described problems is a computer,
Based on a train diagram having a plurality of trains and arrival / departure time data of each train, a vehicle operation network in which nodes representing each train constituting the train diagram are connected by an arc representing that the trains are connected to each other. Generating means (for example, the vehicle operation network generating unit 122 shown in FIG. 3),
For each arc, cost setting means (for example, in FIG. 3) sets the cost based on at least the arrival time of the train corresponding to the original node of the arc and the departure time of the train corresponding to the destination node of the arc. Vehicle operation network generator 122),
Inspection arc selection means (for example, the vehicle operation plan creation unit 120 shown in FIG. 3) that selects an arc for performing vehicle inspection between the trains of the original node and the destination node of the arc among the arcs.
A circuit that takes a given node as an initial point from among the nodes of the vehicle operation network and returns to the initial point by passing through all the nodes once by sequentially selecting arcs. Means for searching for a tour including the selected arc (for example, a tour searching unit 124 shown in FIG. 3);
Creating means for creating a vehicle operation plan based on the circuit searched by the route search means (for example, the vehicle operation plan creating unit 120 shown in FIG. 3);
It is a program to make it function as.

Moreover, the vehicle operation plan creation device of the seventh invention is
Based on a train diagram having a plurality of trains and arrival / departure time data of each train, a vehicle operation network in which nodes representing each train constituting the train diagram are connected by an arc representing that the trains are connected to each other. Generating means for generating
For each arc, at least, a cost setting means for setting the cost based on the arrival time of the train corresponding to the original node of the arc, and the departure time of the train corresponding to the destination node of the arc;
Among the arcs, inspection arc selection means for selecting an arc for performing vehicle inspection between the trains of the original node and the destination node of the arc,
A circuit that takes a given node as an initial point from among the nodes of the vehicle operation network and returns to the initial point by passing through all the nodes once by sequentially selecting arcs. A circuit search means for searching for a circuit including the selected arc;
Creating means for creating a vehicle operation plan based on the circuit searched by the circuit search means;
It is characterized by having.

  According to the first or seventh aspect of the invention, the vehicle inspection is performed during the connection between the trains of the original node and the destination node from among the arcs constituting the vehicle operation network generated based on the train schedule. A vehicle operation plan can be created by selecting a possible arc and searching for a circuit including the selected arc. This makes it possible to create a vehicle operation plan that reliably assigns vehicle inspections that must be performed on the vehicle. In addition, if an arc that can be inspected between trains is selected so that the number of vehicle inspections is minimized, the vehicle operation plan created on the basis of the searched circuit will make the vehicle more efficient. It can be operated.

The second invention is the program of the first invention,
The circuit search means includes
Among the selected arcs, there is a path selection means between selected arcs that determines a path of an arc that connects from the previous node of the i-th arc (i is an integer) to the original node of the i + 1-th arc,
For causing the computer to function to search for a circuit including all of the selected arcs by sequentially changing the value of i from 0 and causing the selected inter-arc path determining means to determine a path. It is a program.

  According to the second aspect of the present invention, the arc path connecting from the previous node of the i-th arc, which is one of the selected arcs, to the original node of the i + 1-th arc is determined, and the value of i is sequentially changed from 0 Then, by determining a path connecting from the i-th arc previous node to the i + 1-th arc original node, it is possible to search for a circuit including all the selected arcs.

The third invention is the program of the second invention,
The circuit search means cuts and cuts one of the arcs included in the determined path based on the operation required time when the vehicle is operated on the path determined by the selected inter-arc path determining means. Path update means for re-searching a path connecting from the original node of the arc to the original node of the (i + 1) -th arc and updating a path connecting from the previous node of the i-th arc to the original node of the (i + 1) -th arc. Is a program for causing the computer to function.

  According to the third aspect of the present invention, one of arcs included in the path is cut based on the required operation time (for example, the number of days may be used) when the vehicle is operated on the determined path. The path connecting the original node of the arc to the original node of the (i + 1) th arc can be re-searched to update the path connecting from the previous node of the i-th arc to the original node of the (i + 1) th arc. According to this, the path can be updated so that the operation required time when the vehicle is operated with the path from the node before the i-th arc to the original node of the i + 1-th arc satisfies the inspection cycle of the vehicle inspection.

The fourth invention is the program of the third invention,
The path updating means is a program for causing the computer to function so as to have a cutting arc selection means for selecting an arc for connecting the train of the original node and the train of the destination node across the day as an arc to be cut.

  According to the fourth aspect of the present invention, when an arc that forms a path connecting from the i-th arc previous node to the i + 1-th arc original node includes an arc that spans the day between connections of trains. In addition, by cutting the arc, the path is updated so that the operation time required when the vehicle is operated in the path from the node before the i-th arc to the original node of the i + 1-th arc satisfies the inspection cycle of the vehicle inspection. can do.

The fifth invention is the program of the third or fourth invention,
The path update means includes a path that connects nodes that can operate on the same day as the train represented by the original node of the cut arc, and includes a path that connects nodes whose arrival time on the same day is a predetermined time condition. , A program for causing the computer to function to search again for a path connecting the original node of the cut arc to the original node of the i + 1-th arc.

  According to the fifth aspect of the present invention, when the arc path connecting the original node of the cut arc to the original node of the (i + 1) th arc is re-searched, the train can be operated on the same day as the train represented by the original node of the cut arc. Paths that connect nodes, and include paths that connect nodes that have a predetermined arrival time on the same day, can be searched again. According to this, a vehicle operation plan is created so that the number of trains operated on the same day by the vehicle increases as much as possible by operating the vehicle on the path from the node before the i-th arc to the original node of the i + 1-th arc. be able to. Therefore, a vehicle operation plan that can efficiently use the vehicle can be created.

A sixth invention is a program according to any one of the first to fifth inventions,
When the search for the tour route by the tour route search unit fails, the inspection arc selection unit increases the number of arcs to be inspected for the vehicle and selects again, and the tour route search unit searches for the tour route. This is a program for further causing the computer to function as control means (for example, a vehicle operation plan creation unit 120 shown in FIG. 3) that performs control to be executed again.

  According to the sixth aspect of the present invention, when the search for the tour route fails, the control is performed such that the number of arcs to be inspected for the vehicle is increased and selected again, and the search for the tour route is executed again. it can. As a result, among the nodes of the vehicle operation network, the circuit that passes through all the nodes only once and returns to the initial point with the given node as the initial point, and reliably includes the selected arc including the selected arc Can be explored.

  According to the present invention, an arc capable of performing a vehicle inspection during connection between trains of a source node and a destination node is selected from arcs constituting a vehicle operation network generated based on a train schedule. In addition, a vehicle operation plan can be created by searching for a circuit including the selected arc. This makes it possible to automatically create a vehicle operation plan that reliably assigns vehicle inspections that must be performed on the vehicle. In addition, if an arc that can be inspected between trains is selected so that the number of vehicle inspections is minimized, the vehicle operation plan created on the basis of the searched circuit will make the vehicle more efficient. It can be operated.

  Hereinafter, an embodiment of a vehicle operation plan creation device according to the present invention will be described in detail with reference to the drawings. In the following description, it is assumed that the vehicle operation plan creation device is configured by one personal computer (hereinafter referred to as a personal computer), but the present invention is not limited to this. In addition, explanations will be made using simple expressions such as “connecting nodes”, “connecting nodes”, and “marking”, but these expressions are metaphors for the contents of information processing, and computer processing That is the reality.

[Overview]
As described above, the vehicle operation plan defines a vehicle operation plan including the vehicle use order and maintenance management for the vehicle, and is created based on the train schedule planned at the time of schedule revision, etc. .

  An example of a train diagram is shown in FIG. In FIG. 1A, a timetable is represented on the horizontal axis and stations AA, BB, and CC are represented on the vertical axis, and a train diagram is represented by “train lines” representing train operations. In addition, each “train line” has a corresponding train number. In the vehicle operation plan, in order to realize this train diagram, a vehicle operation schedule is created by allocating vehicles to all trains constituting the train diagram.

  FIG. 1B shows an example of a vehicle operation plan created based on the train diagram shown in FIG. In FIG. 1B, three routes indicated by a route 1, a route 2, and a route 3 with the horizontal axis direction as time are shown. Here, the route is an operation plan for a certain vehicle for one day, and the operation of the vehicle on a certain day is expressed by “horizontal bars”. “Finish” shown in FIG. 1B indicates that the above-described work inspection is performed.

  For example, in the route 1, the vehicle is operated as the train 2 from the CC station to the BB station, and then the operation inspection is performed at the BB station, and then the train 1 is operated as the train 1 to the CC station. Means driving to BB station. Further, the “horizontal bars” constituting each route correspond to any “train streaks” shown in FIG. 1A, and in the vehicle operation plan shown in FIG. Are operated in accordance with each of the route 1 to the route 3, respectively, thereby realizing the train diagram shown in FIG.

  Further, the three routes shown in FIG. 1 (b) are operated in rotation. That is, a vehicle assigned to the route 1 and operated as described above is operated in a three-day cycle such that the vehicle is driven according to the route 2 on the next day of the driving day, the route 3 on the next day, and returned to the route 1. Is done. That is, the vehicle operation plan is a set of routes (alternating numbers) created by dividing connection patterns in which all trains constituting a train diagram are connected cyclically.

  As described above, the vehicle operation plan is created so as to satisfy the restrictions regarding the alternating inspection and the work inspection (vehicle inspection) and the restrictions when connecting the train, and to meet the demand for cost reduction. For example, there are the following restrictions and requests.

(1) Restrictions on inspection execution conditions As described above, in the alternating box, the execution positions of the alternating inspection and the work inspection must be set so as to satisfy the inspection cycle and the execution conditions.

(2) Restrictions when connecting trains First, in the trains before and after connecting, the terminal station of the front train and the starting station of the rear train must be the same station. For example, if the terminal station of the X train and the starting station of the Y train are not the same station, it is impossible to connect the trains. When creating a vehicle operation plan, a forwarding train is set as appropriate in order to satisfy this restriction. In addition, at this time, in order to ensure a minimum stop time including getting on / off and cleaning of passengers / crew at the connecting station, a predetermined time interval is set between the arrival time of the previous train and the departure time of the rear train. It is necessary to open.

  In addition, the arrival time of the previous train and the departure time of the subsequent train must not be reversed except when connecting to a train that is operated the next day. For example, when the first train departure time of the Y train precedes the last train arrival time of the X train, it is impossible to connect the X train and the Y train.

  In the police box, the end station and the start station of two consecutive routes must be the same. This is because the vehicle that has finished the first route at the end station starts driving from the start station of the second route the next day, and this constraint is met when creating a vehicle operation plan. In addition, a forwarding train is set as appropriate.

(3) Request for cost reduction A vehicle operation plan capable of efficiently using vehicles is desirable. For this reason, for example, a request to minimize the number of times of alternating inspections and work inspections is used, and it is used to realize a one-day train schedule so that the number of trains operated on a daily basis by each vehicle is increased as much as possible. There is a demand to reduce the number of vehicles (that is, the number of routes) as much as possible, and a demand to reduce the number of forward trains and the travel distance of the number of trains as much as possible.

  In this embodiment, the vehicle operation plan is represented by one connection pattern (tour circuit) for cyclically connecting all the trains constituting the train diagram, and is represented by the best circuit that satisfies the above-described restrictions and requests. By adopting the vehicle operation plan draft, a vehicle operation plan is automatically created.

  Specifically, the connection pattern between trains constituting the train diagram is expressed by a network (hereinafter referred to as “vehicle operation network”) in which nodes representing each train are connected by an arc representing the connection between trains. In the vehicle operation network, a candidate solution is generated by searching for a circuit that passes through all the nodes only once and returns to the initial point. Then, the process of searching for a circuit is repeated to search for a better circuit (the best solution), and the candidate solution generation process is repeated a specified number of times, and the train of the train indicated by the circuit indicated by the best solution is displayed. The connection is a vehicle operation plan.

  FIG. 2A shows an example of a vehicle operation network. As shown to Fig.2 (a), a vehicle operation network represents each node which represents each train (train 1-train 7 in Fig.2 (a)) which comprises a train diagram, and the connection of the connectable trains. It consists of a directed arc. For example, among the arcs connecting the node N10 representing the train 7 and the node N12 representing the train 6, the arc A10 from the node N10 to the node N12 is operated as the train 6 after the vehicle is operated as the train 7. It represents that. Whether trains can be connected is determined as follows. For example, when the time required for train connection is insufficient, such as a stop time that must be secured at a minimum at a station to be connected, it is determined that the trains cannot be connected. In addition, if the number of vehicles that can stay at the station at the same time is determined in advance due to restrictions due to the number of lines, etc., and connection candidates between trains at the station are limited, Connection of trains other than connection candidates is judged impossible.

  Specifically, between the nodes, the four types of arcs shown in the following (a) to (d) are based on the restriction (2) when connecting the train described as the restriction when creating the vehicle operation plan. Are connected by any of the above.

(A) Same-day connection arc Nodes representing trains that can be connected during the day are connected by the same-day connection arc. In FIG. 2A, the connection arc on that day is indicated by a thick solid arrow. For example, in FIG. 2A, the day connection arc A <b> 12 from the node N <b> 14 to the node N <b> 16 represents that the vehicle is operated as the train 1 after the vehicle is operated as the train 2.

(B) Next-day connection arc Nodes representing trains connectable on the next day are connected by a next-day connection arc. In FIG. 2A, this next day connection arc is indicated by a thin solid arrow. For example, in FIG. 2A, the next day connection arc A14 from the node N16 to the node N14 represents that the vehicle is operated as the train 2 the next day after the vehicle is operated as the train 1.

(C) Connected arc on the same day when the trains arrive at different stations, and the nodes representing trains that can be connected during the day by operating as a forward train are connected by the connected arc on the same day. The In FIG. 2A, the connection arc on the same day is indicated by a thick dotted arrow. For example, in FIG. 2A, the day-to-day connection arc A16 from the node N18 to the node N12 is routed from the terminal station of the train 4 to the first station of the train 6 during the day after the vehicle is operated as the train 4. It represents driving as a train and driving as a train 6.

(D) Next-day connection arc for rerouting The nodes representing trains that can be connected to the next day by operating as a forward train are connected by this connection arc for the next-day transmission. The In FIG. 2 (a), the next day connection arc is indicated by a thin dotted arrow. For example, in FIG. 2 (a), the next day connection arc A18 from the node N12 to the node N19 is transmitted from the terminal station of the train 6 to the starting station of the train 4 during the day or the next day after the vehicle is operated as the train 6. It is operated as a forward train until it is operated as the train 4 on the next day.

  In each node, the time (total operation time) from when the corresponding train leaves the first station to the last station is set as a cost, and each arc is represented by the original node. A time interval (hereinafter referred to as “departure time interval”) between the train end time and the train start time represented by the previous node is set as a cost.

  Further, among the arcs constituting the vehicle operation network, a mark indicating that each inspection can be performed is attached to an arc that satisfies the vehicle inspection execution condition. Specifically, the arc is an arc in which the starting station of the train represented by the previous node is a station having a vehicle base, or an arc in which the terminal station of the train represented by the former node is a station having a vehicle base. An arc that is determined to satisfy the conditions related to the time zone in which the vehicle inspection can be performed based on the start time of the train represented by the previous node or the end time of the train represented by the former node. In addition, there is a mark indicating that each inspection can be performed on condition that the set cost (departure time interval) is an arc that satisfies the execution conditions regarding the time required for vehicle inspection. The

  For example, in FIG. 2 (a), the next day connection arc A18 is marked with an alternate inspection possible mark M10, and alternate inspection and work inspection are carried out between the train 6 and the train 5 for the vehicle. It shows that it is possible. Further, in FIG. 2 (a), the inspection arc M12 is attached to the connection arc A12 on the day, and it is possible to carry out the operation inspection between the connection of the train 2 and the train 1 to the vehicle. Show. Hereinafter, an arc with an alternating inspection possible mark is referred to as an “alternate inspection possible arc”, and an arc with an operational inspection possible mark is referred to as an “inspection possible arc”. .

  FIG.2 (b) is a figure which shows an example of the circuit searched based on the vehicle operation network shown to (a). Although details will be described later, in the present embodiment, the minimum number of inspections required to satisfy the inspection cycle is assumed, and the number of inspectable arcs based on the number of inspections is first selected. A circuit that passes through all the nodes only once and includes all the selected arcs that can be inspected is searched for and output as a vehicle operation plan.

  Further, when the search for the tour route has failed, the number of inspections described above is increased to select an inspectable arc, and the tour route is searched again. FIG. 2 (b) shows an example of a circuit when it is assumed that the number of work inspections required to satisfy the inspection cycle is one. In FIG. 2 (b), the selected inspectable arc (alternate inspection) is shown. It shows a circuit that includes the connection arc A18 of the next day transmission that is a possible arc). In addition, this circuit corresponds to the vehicle operation plan shown in FIG.

[Function configuration]
Next, the functional configuration of the vehicle operation plan creation device 10 will be described. FIG. 3 is a block diagram illustrating an example of a functional configuration of the vehicle operation plan creation device 10. As illustrated in FIG. 3, the vehicle operation plan creation device 10 includes a CPU 100, a ROM 200, an input device 300, a display device 400, a communication device 500, a RAM 600, a storage device 700, and a storage medium 800, and each unit other than the storage medium 800. Are connected by a bus 900.

  The CPU 100 executes processing based on a predetermined program in accordance with an input instruction, performs an instruction to each functional unit, data transfer, and the like, and comprehensively controls the vehicle operation plan creation apparatus 10. Specifically, the CPU 100 reads out various programs stored in the storage medium 800 included in the ROM 200 or the storage device 700 in accordance with an operation signal or the like input via the input device 300, expands the program in the RAM 600, and Execute the process according to the program. Then, the processing result is stored in the RAM 600, and a display signal for displaying the processing result is output to the display device 400. Further, part or all of the processing results stored in the RAM 600 are stored in the storage medium 800.

  Various functions are realized by the CPU 100. In particular, a function unit that creates a vehicle operation plan is called a vehicle operation plan creation unit 120. In the vehicle operation plan creation unit 120, a function unit that creates a vehicle operation network is called a function unit. The vehicle operation network generation unit 122 and a function unit that searches for a tour are called a tour search unit 124 and will be described below.

  The vehicle operation plan creation unit 120 causes the vehicle operation network generation unit 122 (to be described later) to execute a vehicle operation network generation process, and assigns vehicle inspections according to the inspection cycle to the inspectable arcs that make up the generated vehicle operation network. In the present embodiment, the upper limit of the inspection cycle for the work inspection is set to 3 days, and the inspection cycle is satisfied by performing the alternating inspection once. Then, the vehicle operation plan creation unit 120 performs control to cause the traveling route search unit 124 to repeatedly execute the traveling route search process a predetermined number of times, and the evaluation value is the best among the traveling routes searched by the traveling route search unit 124. The patrol circuit is output as a vehicle operation plan.

  The vehicle operation network generation unit 122 generates the vehicle operation network described with reference to FIG. 2 based on the train schedule data 820 stored in the storage medium 800.

  Specifically, the vehicle operation network generation unit 122 first sets the total operation time as the cost of each node representing each train constituting the train diagram. In addition, the vehicle operation network generation unit 122 sets the type of arc (arc type) that connects between nodes representing connectable trains, and sets the departure and arrival time intervals as the cost of each arc.

  Next, the vehicle operation network generation unit 122 has an arc in which the first station of the train represented by the previous node is a station having a vehicle base, or a station in which the last station of the train represented by the former node has a vehicle base. A list of arcs that have been determined that vehicle inspection is possible between the trains of the source node and the destination node based on the arrival and departure time intervals set as the costs as described above (inspectable arc list) ). Specifically, the vehicle operation network generation unit 122 creates a list of arcs that can be subjected to an alternating inspection and a list of arcs that can be subjected to a work inspection.

  The vehicle operation network generation unit 122 connects each node with an arc of the set arc type, and indicates a mark (alternate inspection is possible) indicating that the vehicle inspection can be performed on the corresponding arc according to the inspectable arc list. A vehicle operation network is generated with a mark or a work inspection possible mark).

  The tour searching unit 124 searches for a tour of the vehicle operation network generated by the vehicle operation network generating unit 122. Specifically, the tour searching unit 124 is a tour that passes through all the nodes constituting the vehicle operation network only once, and includes a tour that includes all the inspectable arcs assigned for vehicle inspection. Explore.

  The ROM 200 stores initial programs for performing various initial settings, hardware configurations, loading of necessary programs, and the like.

  The input device 300 is realized by a keyboard having a cursor key, a numeric keypad, various function keys, etc., and a pointing device such as a mouse. When these are operated, an operation signal corresponding to the operation is provided. Is output to the CPU 100.

  The display device 400 is a display device such as an LCD (Liquid Crystal Display) or an ELD (Electronic Luminescent Display), and displays various screens based on display signals input from the CPU 100.

  The communication device 500 is a device for exchanging information used inside the device with the outside via a communication line, and performs control for communicating with other devices.

  The RAM 600 is a semiconductor memory used as a working memory for the CPU 100, and includes a memory area that temporarily stores programs executed by the CPU 100, data related to the execution of these programs, and the like. In particular, in order to realize the present embodiment, a candidate solution storage area 620 that holds a candidate solution of the tour route and its evaluation value, a best solution storage region 640 that holds the best solution of the tour route and its evaluation value, and vehicle operation Vehicle operation network data storage area 660 for holding network data, and in particular, cost data 662 and inspectable arc list 664 are stored in the vehicle operation network data storage area 660.

  The cost data 662 stores the cost set for each arc and each node by the vehicle operation network generation unit 122 as described above. FIG. 4 is a diagram illustrating an example of the cost data 662. As shown in FIG. 4, the cost data 662 stores node cost data 662a and arc cost data 662b. In the node cost data 662a, a cost is set in association with a node number indicating a node representing each train. In the arc cost data 662b, an arc type and a cost are set in association with an arc number indicating an arc representing a connection between connectable trains.

  The inspectable arc list 664 stores the inspectable arc list generated by the vehicle operation network generation unit 122 as described above. FIG. 5 is a diagram illustrating an example of the inspectable arc list 664. As shown in FIG. 5, the inspectable arc list 664 stores an alternating inspectable arc list 664 a and a work inspectable arc list 664 b. The alternate inspection possible arc list 664a stores a list of arc numbers of arcs representing connections between trains capable of performing alternate inspection during the connection. In the work inspection possible arc list 664b, a list of arc numbers of arcs representing connections between trains capable of performing the work inspection in the middle of connection is stored.

  The storage device 700 includes a storage medium 800 configured by a magnetic or optical storage medium, a semiconductor memory, or the like. The storage medium 800 is fixedly attached to the storage device 700 or is detachably mounted, and stores various processing programs, data processed by these processing programs, and the like. In particular, train diagram data 820, vehicle operation plan data 840, and a vehicle operation plan creation program 860 are stored to realize this embodiment.

  In the train diagram data 820, train information of each train constituting the train diagram is accumulated. The train information includes, for example, information such as the train number of the train, the train type, and the arrival and departure times at each station where the train arrives and departs. Further, the vehicle operation plan data 840 stores the best solution of the tour route created and confirmed by the vehicle operation plan creation device 10 as vehicle operation plan data.

  The vehicle operation plan creation program 860 is a program for causing the CPU 100 to function as the vehicle operation plan creation unit 120. The vehicle operation plan creation program 860 causes the vehicle operation plan creation unit 120 to function as the vehicle operation network creation unit 122. A vehicle operation network generation program 862 that is a program for the above and a tour search program 864 that is a program for causing the vehicle operation plan creation unit 120 to function as the tour search unit 124.

[Process flow]
Next, the flow of the vehicle operation plan creation process in this embodiment will be described. FIG. 6 is a diagram illustrating an example of a process flow of the vehicle operation plan creation unit 120 related to execution of the vehicle operation plan creation process. This vehicle operation plan creation process is realized by the vehicle operation plan creation unit 120 reading and executing the vehicle operation plan creation program 860.

  First, the vehicle operation plan creation unit 120 performs initial setting processing such as initialization of the best solution storage area 640 in which the number of generations of candidate solutions (the number of loops of loop A described later) and the best solution and its evaluation value are stored, for example. Is executed (step S10).

  Subsequently, the vehicle operation network generation unit 122 executes a vehicle operation network generation process (step S15). FIG. 7 is a diagram illustrating an example of a process flow of the vehicle operation network generation unit 122 related to execution of the vehicle operation network generation process. The vehicle operation network generation process is realized by the vehicle operation network generation unit 122 reading and executing the vehicle operation network generation program 862.

  The vehicle operation network generation unit 122 first determines whether or not each train constituting the train diagram can be connected, and between all the nodes constituting the train diagram and nodes representing connectable trains. Cost data 662 is created for the arcs to be connected (step S152).

  Specifically, the vehicle operation network generation unit 122 sets the total operation time as the cost of the node representing each train based on the train diagram data 820, and stores it in the cost data 662 as the node cost data 662a. To do.

  Moreover, the vehicle operation network production | generation part 122 sets the kind (arc kind) of each arc. Specifically, the vehicle operation network generation unit 122 sets the current day connection arc as the arc type between nodes representing trains that can be connected during the day, and sets the next day as the arc type between nodes representing trains that can be connected the next day. Set the connection arc. In addition, the vehicle operation network generation unit 122 sets the forward connection day connection arc to the arc type between nodes representing trains that can be connected during the day by operating as a forward train when the departure stations between the trains are different. Then, the next day connection arc is set to the arc type between nodes representing the train that can be connected the next day by being operated as a forward train. Then, the vehicle operation network generation unit 122 sets the departure / arrival time interval as the cost of each arc set as described above, and stores it in the cost data 662 as the arc cost data 662b together with the set arc type.

  Next, the vehicle operation network generation unit 122 creates an inspectable arc list 664 that satisfies the vehicle inspection implementation conditions (step S154).

  Specifically, the vehicle operation network generation unit 122 is represented by an arc or a former node where the starting station of the train represented by the previous node is a station having a vehicle base based on the train schedule data 820. An arc is extracted where the terminal station of the train is a station with a depot.

  And the vehicle operation network production | generation part 122 can implement an alternating inspection from the extracted arc based on the train start time represented by the previous node or the train end time represented by the former node. An arc that is determined to satisfy the implementation conditions related to the time zone, and that can be checked alternately between the trains of the source node and the destination node according to the arrival time interval set as the cost as described above. Extracted and stored in the inspectable arc list 664 as an alternating inspection inspectable arc list 664a. Similarly, arcs that can be inspected in operation are extracted and stored in the inspectable arc list 664 in the inspectable arc list 664b.

  And vehicle operation network production | generation part 122 produces | generates a vehicle operation network with reference to the vehicle operation network data storage area 660 (step S156). At this time, based on the arc cost data stored in the cost data 662, the vehicle operation network generation unit 122 connects the connectable nodes with an arc of the corresponding arc type. Further, at this time, the vehicle operation network generation unit 122 attaches the alternating inspection possible mark to the corresponding arc based on the alternating inspection possible arc list 664a stored in the inspectable arc list 664, and also performs the work inspection possible arc list 664b. Based on the above, mark the applicable arc for work inspection.

  Returning to FIG. 6, when the vehicle operation network generation process executed by the vehicle operation network generation unit 122 is completed, the vehicle operation plan generation unit 120 refers to the train diagram data 820 to at least realize the train diagram. Obtain the required number of vehicles (required minimum number of vehicles), assume the minimum number of work inspections required to satisfy the inspection cycle based on this minimum number of vehicles, and subtract 1 from this number Temporarily determined as the estimated number of times of work inspection (step S20). Note that the maximum number of trains operated in the same time zone in the train diagram is acquired as the minimum required number of vehicles.

  Next, the vehicle operation plan creation unit 120 assigns “0” to the execution order and assigns an alternating inspection to the generated vehicle operation network (step S25). Specifically, the vehicle operation plan creation unit 120 refers to the inspectable arc list 664, selects one arc number randomly from the alternating inspection inspectable arc list 664a, for example, by generating a random number, and uses the corresponding arc. “0” is assigned to the execution order as performing the alternating inspection at the connection between the represented trains.

  Subsequently, the vehicle operation plan creation unit 120 executes the process of loop A a predetermined number of times (steps S30 to S55).

  In loop A, the vehicle operation plan creation unit 120 first assigns a work inspection by assigning a number after “1” as the execution order to the generated vehicle operation network (step S35). Specifically, the vehicle operation plan creation unit 120 refers to the inspectable arc list 664, selects, for example, the number of arcs for the estimated number of times of the work inspection from the work inspectable arc list 664b, and displays the arc number by the corresponding arc. As the work inspection is carried out between the connected trains, numbers after “1” are assigned in the selected order. It should be noted that the arc number selected when assigning the work inspection must be selected so as not to overlap with the arc number selected as the alternating inspection execution position, and the vehicle operation plan creation unit 120 is selected in step S25. If an arc number that overlaps the selected arc number is selected, the arc number is selected again. In addition, in order to search for a circuit that finally passes through all the nodes only once and returns to the initial point, the source node and the destination node of each inspectable arc selected in step S25 and step S35 do not overlap each other. Each arc number must be selected. Therefore, the vehicle operation plan creation unit 120 sets the arc number again even when the source node or the destination node of the selected inspectable arc overlaps with either the source node or the destination node of the already selected inspectable arc. Select.

  Subsequently, the traveling route search unit 124 performs a traveling route search process (step S40). FIG. 8 is a diagram illustrating an example of a processing flow of the tour searching unit 124 related to execution of the tour searching process. This traveling route search process is realized by the traveling route search unit 124 reading and executing the traveling route search program 864.

  The tour searching unit 124 first initializes by assigning “0” to the target execution number k (step S400). Further, at this time, a definite path P0 used for work when searching for a circuit is initialized.

  Subsequently, the traveling route search unit 124 sets the starting point node S and the ending point node E based on the target execution number k, and determines between the selected arcs (step S410). Specifically, the traveling route search unit 124 sets the destination node of the inspectable arc assigned the value of the target execution number k as the execution number as the starting node S, and the inspectable arc assigned the value of k + 1 as the execution number. The original node is the end point node E, and the distance between the start point node S and the end point node E is determined between the selected arcs.

  For example, when the initial value “0” is set for the target execution number k, the interval between the selected arcs is the destination node of the inspectable arc (alternate inspection possible arc) assigned “0” for the execution number. It is set between the start node S and the end node E which is the original node of the inspectable arc (work inspection inspectable arc) assigned “1” as the execution number. When the target execution number k matches the estimated number of work inspections, between the selected arcs, the start node S that is the destination node of the inspectable arc assigned “k” as the execution number and “0” It is set between the end node E, which is the original node of the allocated checkable arc (alternate checkable arc).

  Subsequently, the tour searching unit 124 refers to the cost data 662 and searches for the shortest path from the start node S to the end node E (the path having the lowest cost of the entire path connected by the arc) (step S420). ). In this embodiment, this shortest path is searched using a technique called Dijkstra method, which is a known algorithm for solving the shortest path problem. The path search method from the start node S to the end node E is not limited to the Dijkstra method. For example, by using a technique called greedy method, the destination node is sequentially selected by selecting the arc with the lowest cost among the arcs connected to the source node, and the path from the start node S to the end node E It is good also as searching. At this time, a path formed by adding the searched shortest path to the confirmed path P0 is held in the candidate solution storage area 620 as a candidate solution.

  FIG. 9 is a diagram showing an example of the shortest path searched by the process of step S420. FIG. 9 shows a part of the nodes constituting the generated vehicle operation network. The shortest path shown in FIG. 9 is, for example, a starting node N100 that is a destination node of the alternating inspectable arc A100 that is arbitrarily selected from the inspectable arcs constituting the vehicle operation network and assigned the vehicle inspection, and can be inspected. The shortest path from the start node S to the end node E, which is searched when the distance from the end node N110 that is the original node of the arc A110 is set as the target selection arc, is shown.

  Subsequently, the traveling route searching unit 124 counts the number of arcs that are straddling among the arcs included in the path between the target selection arcs, that is, the next day connection arc or the next day connection arc (step S430). . Then, the traveling route search unit 124 determines whether or not the counted number of arcs matches the work inspection cycle (3 days), and if it does not match (step S440: NO), the path search processing is performed. (Step S450).

  FIG. 10 is a flowchart illustrating an example of the path addition process. As shown in FIG. 10, the traveling route search unit 124 first cuts the arc closest to the start node S among the arcs that have been straddled (step S452). Subsequently, the traveling route search unit 124 can connect to the train represented by the original node of the disconnected arc from the unreached nodes (nodes not selected as paths) on the same day, and the arrival time Are sequentially selected to generate an additional path (step S454). At this time, the additional path is added to the final path, and the final path P0 is updated. The traveling route search unit 124 can connect to the train represented by the original node of the disconnected arc on the same day, and the unreachable node whose arrival time is the latest is the end node E. Secondly, arcs that arrive at the node with the second latest arrival time are sequentially selected to generate an additional path. Then, the traveling route search unit 124 updates the starting node S to the earliest node that the additional path can reach (step S456).

  In the path addition process, in step S452, the arc closest to the starting node S among the arcs that are straddled is cut, but other methods may be used. In other words, for example, when a plurality of arcs that span between the objects are included in the path between the target selection arcs, the arcs that are to be cut may be randomly determined using a random number. Also, for each straddling arc included in the path between the target selection arcs, an additional path is generated on the assumption that the straddling arc is cut, and based on the number of arcs and nodes included in the additional path It is good also as determining the arc of the day straddling to cut. For example, it may be determined as an arc that cuts a cross-day arc determined to generate an additional path that includes the largest number of arcs or nodes by cutting.

  FIG. 11 is a diagram for explaining a process of generating a path between target selection arcs, and illustrates how the path between target selection arcs shown in FIG. 9 is corrected. In FIG. 11 (a), among the arcs constituting the path between the target selection arcs, the arc that is straddling and that is closest to the starting node S (the next day connection arc A200 indicated by a thin solid line) Is disconnected. Of the unreachable nodes connected to the former node N200 of the next-day connection arc A200 that has been disconnected, arcs that reach the node that arrives on the same day and has the latest arrival time are sequentially selected. Additional paths are generated.

  For example, in FIG. 11B, a path from the node N200 to the node N300 is generated as an additional path, and the node N300 is changed to the starting node S.

  Returning to FIG. 8, when the path search unit 124 finishes the path addition processing in step S450, the traveling circuit search unit 124 proceeds to step S420 and determines in step S440 that the number of arcs in which the day span has occurred coincides with the work inspection cycle. Until the above process is repeated.

  FIG. 11C illustrates an example of a path between target selection arcs that has been corrected by searching for the shortest path from the starting node S changed to the ending node E as illustrated in FIG. FIG. Note that there are two arcs included in this path that have crossed days (next-day connection arcs A400 and A410), which do not match the inspection cycle (3 days) of the work inspection. The scissor arc (next day connection arc A400) is cut, and the above-described processing is executed again. Further, when the number of straddling arcs constituting the path between the target selection arcs coincides with the inspection period of the work inspection, the path between the target selection arcs is determined. And between the object selection arcs is changed according to the execution order of the inspectable arcs to which the vehicle inspection is assigned, and the paths between the object selection arcs are sequentially determined. That is, this makes it possible to correct the path between the target selection arcs, which are between the inspectable arcs to which vehicle inspection is assigned in advance, so that the vehicle operates as many trains as possible within the inspection cycle.

  Then, in step S440 shown in FIG. 8, when it is determined that the number of arcs that have passed over the day coincides with the work inspection cycle, the traveling route search unit 124 subsequently assumes that the target execution number k is the work inspection assumption. It is determined whether or not it matches the number of times, and if it does not match (step S460: NO), after the target execution number k is incremented (step S470), the process proceeds to step S410 and the above-described processing is repeatedly executed. . Thereby, as described above, the distance between the selected arcs is sequentially changed, and the path between the selected arcs is sequentially generated, thereby completing the circuit.

尚、ｘ_iを車両運用計画を評価する基準となる種々の項目に対応するパラメータ値、ｐ_iをそれぞれの評価基準に対する重みとする。例えば、ｎ＝３とし、ｘ１を使用車両数、ｘ２を回送列車の本数、ｘ３を回送列車の走行距離、そしてｐ１〜ｐ３をｘ１〜ｘ３それぞれに対応する重みとする。 If it is determined in step S460 that the estimated number of work inspections is satisfied, the traveling route search unit 124 calculates the evaluation value of the candidate solution according to the following equation (1) (step S480).

Note that x _i is a parameter value corresponding to various items that serve as a criterion for evaluating the vehicle operation plan, and p _i is a weight for each evaluation criterion. For example, n = 3, x1 is the number of vehicles used, x2 is the number of forward trains, x3 is the travel distance of the forward trains, and p1 to p3 are weights corresponding to x1 to x3, respectively.

  Thereby, for example, the traveling route search unit 124 sets the evaluation value of the candidate solution using the items such as the number of vehicles (the number of routes), the number of traveling trains, and the travel distance used in the vehicle operation plan represented by the candidate solution as evaluation criteria. Can be calculated.

  Returning to FIG. 6, when the traveling route search process executed by the traveling route search unit 124 is completed, the vehicle operation plan creation unit 120 determines the best solution evaluation value stored in the best solution storage region 640 and the candidate solution storage region. The evaluation value of the candidate solution stored in 620 is compared (step S45), and the best solution storage area 640 is updated in the candidate solution storage area 620 when the newly searched candidate solution has a higher evaluation than the best solution. (Step S50).

  As described above, the candidate solution is a circuit having an initial point at the destination node of the selected alternating inspection possible arc, and satisfies the inspection period of the operation inspection in the vehicle operation plan represented by the obtained circuit. At the same time, the correction of the circuit is repeated so that as many trains as possible are operated by the vehicle within this inspection cycle, and the final generated circuit is evaluated based on the request for the vehicle operation plan. Created.

  When the loop A ends, the vehicle operation plan creation unit 120 determines whether or not the best solution circuit is a circuit that passes through all the nodes only once and returns to the initial point. It is determined whether the best solution is complete or incomplete (step S60). If the vehicle operation plan creation unit 120 determines that the obtained best solution is incomplete, the vehicle operation plan creation unit 120 increments the estimated number of work inspections by 1 (step S65), returns to step S25, and is described above. Repeat the process.

  As a result, if the obtained best solution is incomplete, it is determined that a vehicle operation plan in which the number of roads is the required minimum number of vehicles acquired in step S20 cannot be created. And the best solution creation process can be redone.

  If it is determined in step S60 that the obtained best solution is complete, the vehicle operation plan represented by the best solution is stored in the storage medium 800 as the vehicle operation plan data 840, for example, on the display device 400. This is output (step S70) and the process is terminated.

  As described above, according to the present embodiment, the candidate solution searches for the shortest path between the selected arcs to which vehicle inspection is assigned (the path with the lowest cost of the entire path connected by the arc) and each inspection. The process of repeating the correction of the shortest path so that the vehicle operates as many trains as possible within the inspection cycle is performed between each selected arc, thereby generating a circuit that satisfies the conditions for vehicle inspection and vehicle operation. It is created by performing an evaluation based on the request for the plan. Therefore, it is possible to create a vehicle operation plan that observes the restrictions relating to the creation of the vehicle operation plan described above and reflects requests for the vehicle operation plan as much as possible.

  In the above-described embodiment, for each arc, the time interval (departure time interval) between the train end time represented by the original node and the train start time represented by the previous node is set as a cost. However, it may be as follows.

  That is, it is good also as adjusting the cost set with respect to the arc which connects between the nodes showing the train which can be connected on the next day. Specifically, for these arcs, a time interval from the train arrival time represented by the original node to 3 am, for example, may be set as a cost. This is because the cost cannot be handled in the same manner between the time zone from 3 am to the first departure time and the daytime time zone.

  In addition, a value obtained by adding a predetermined value to the arrival / departure time interval for the sending day connection arc and the next day connection arc representing the connection between trains that can be connected by driving the vehicle as a forwarding train. It is good also as setting as. At this time, the predetermined value to be added may be changed in consideration of the traveling time of the forward train. Since the path with the lowest cost of the entire path connected by the arc is searched as the shortest path (step S420 in FIG. 8), when the connection between the trains corresponding to the arc connecting the nodes uses the forward train This is because, by setting a large value for the cost set for the arc, consideration can be given so that the corresponding nodes are not connected as much as possible. Thereby, the vehicle operation plan which does not drive a forwarding train as much as possible can be created.

  Moreover, it is good also as performing the vehicle operation plan preparation process shown and demonstrated in FIG. 6 with respect to a several different alternating inspection plan. According to this, it is possible to search for the best circuit.

  Further, although the train diagram has been described as being stored in the storage medium 800, for example, data communication is performed with a database in which train diagram data is stored via the communication device 500 and the communication line, and the corresponding train diagram is obtained. Of course, you can download the data.

(A) shows an example of a train schedule, and (b) shows an example of a vehicle operation plan based on the train schedule shown in (a). (A) is an example of a vehicle operation network, (b) is a figure which shows an example of a circuit of the vehicle operation network shown to (a). The block diagram which shows an example of a function structure of a vehicle operation plan preparation apparatus. The figure which shows an example of cost data. The figure which shows an example of an arc list | wrist which can be test | inspected. The flowchart which shows an example of a vehicle operation plan creation process. The flowchart which shows an example of a vehicle operation network production | generation process. The flowchart which shows an example of a tour search process. The figure which shows an example of the shortest path between object selection arcs. The flowchart which shows an example of a path addition process. The figure for demonstrating the production | generation process of the path | pass between object selection arcs.

Explanation of symbols

10 vehicle operation plan creation device 100 CPU
110 Vehicle operation plan creation unit 112 Vehicle operation network generation unit 114 Travel route search unit 200 ROM
300 input device 400 display device 500 communication device 600 RAM
620 Candidate solution storage area 640 Best solution storage area 660 Vehicle operation network data storage area 662 Checkable arc list 664 Cost data 700 Storage device 800 Storage medium 820 Train diagram data 840 Vehicle operation plan data 860 Vehicle operation plan creation program 862 Vehicle operation network Generation program 864 tour search program

Claims (7)

Computer
Based on a train diagram having a plurality of trains and arrival / departure time data of each train, a vehicle operation network in which nodes representing each train constituting the train diagram are connected by an arc representing that the trains are connected to each other. Generating means for generating
Cost setting means for setting the cost based on at least the arrival time of the train corresponding to the original node of the arc and the departure time of the train corresponding to the destination node of the arc for each arc.
Inspection arc selection means for selecting an arc for performing vehicle inspection between the trains of the original node and the destination node of the arc among the arcs,
A circuit that takes a given node as an initial point from among the nodes of the vehicle operation network and returns to the initial point by passing through all the nodes once by sequentially selecting arcs. A circuit search means for searching for a circuit including the selected arc;
Creating means for creating a vehicle operation plan based on the circuit searched by the circuit search means;
Program to function as. The circuit search means includes
Among the selected arcs, there is a path selection means between selected arcs that determines a path of an arc that connects from the previous node of the i-th arc (i is an integer) to the original node of the i + 1-th arc,
For causing the computer to function to search for a circuit including all of the selected arcs by sequentially changing the value of i from 0 and causing the selected inter-arc path determining means to determine a path. The program according to claim 1.   The circuit search means cuts and cuts one of the arcs included in the determined path based on the operation required time when the vehicle is operated on the path determined by the selected inter-arc path determining means. Path update means for re-searching a path connecting from an arc origin node to the i + 1 arc origin node and updating a path connecting the i th arc destination node to the i + 1 arc origin node. The program according to claim 2 for causing said computer to function.   4. The computer according to claim 3, wherein the path update means causes the computer to function as cutting arc selection means for selecting an arc for connecting the train of the original node and the train of the destination node across the day as an arc to be cut. The program described.   The path update means includes a path that connects nodes that can operate on the same day as the train represented by the original node of the cut arc, and includes a path that connects nodes whose arrival time on the same day is a predetermined time condition. The program according to claim 3 or 4, for causing the computer to function to search again for a path connecting the original node of the cut arc to the original node of the i + 1-th arc.   When the search for the tour route by the tour route search unit fails, the inspection arc selection unit increases the number of arcs to be inspected for the vehicle again and selects it again, and the tour route search unit searches for the tour route. The program as described in any one of Claims 1-5 for making the said computer further function as a control means which performs control performed again. Based on a train diagram having a plurality of trains and arrival / departure time data of each train, a vehicle operation network in which nodes representing each train constituting the train diagram are connected by an arc representing that the trains are connected to each other. Generating means for generating
For each arc, at least, a cost setting means for setting the cost based on the arrival time of the train corresponding to the original node of the arc, and the departure time of the train corresponding to the destination node of the arc;
Among the arcs, inspection arc selection means for selecting an arc for performing vehicle inspection between the trains of the original node and the destination node of the arc,
A circuit that takes a given node as an initial point from among the nodes of the vehicle operation network and returns to the initial point by passing through all the nodes once by sequentially selecting arcs. A circuit search means for searching for a circuit including the selected arc;
Creating means for creating a vehicle operation plan based on the circuit searched by the circuit search means;
A vehicle operation plan creation device comprising:

Images