Driving arrangement proposal system and driving arrangement proposal method

Abstract

To propose a suitable driving arrangement content for the current event.SOLUTION: A driving arrangement proposal system maintains driving arrangement information that indicates a plurality of past events, the driving arrangement content corresponding to each of the plurality of past events, and the values of a plurality of types of attributes, and the plurality of types of attributes comprises event communication attributes, operation arrangement attributes, and equipment sensor attributes, and obtains the current event, the values of the plurality of types of attributes corresponding to the current event, and the weights corresponding to each of the plurality of types of attributes, and calculate the weighted similarity between the current event and each of the plurality of past events on the basis of the values of the plurality of attributes corresponding to the current event and the plurality of past events and the corresponding weights, and output the driving arrangement content corresponding to the driving arrangement information for the past event with the high similarity.SELECTED DRAWING: Figure 1A

Description

[Technical field]

[0001] The present invention relates to a timetable replanning proposal system and a timetable replanning proposal method. [Background technology]

[0002] Conventional traffic control systems are equipped with a function that allows dispatchers to input traffic rescheduling details based on train operation status. However, the recovery time from operational problems caused by traffic rescheduling based on the content input by the dispatcher can vary greatly depending on the experience of the dispatcher. Furthermore, since the situation, location, and time in which operational problems occur are not constant, it takes a long time for dispatchers to gain experience in inputting traffic rescheduling details.

[0003] Furthermore, due to the declining birthrate and aging population, it is expected that the number of dispatchers who can input effective train rescheduling will decrease year by year. Therefore, in order to minimize disruptions to the schedule due to operational problems, it is expected that there will be a demand for the development of a system that supports dispatchers' decision-making by proposing optimal train rescheduling after considering various factors such as passengers, crew members, and vehicle facilities.

[0004] As a background technology of this technical field, there is JP 2021-107189 A (Patent Document 1). This publication states that "The operation management system of the embodiment includes a storage unit, an acquisition unit, and an extraction unit. The storage unit stores vehicle identification information for identifying a first vehicle, the position of the first vehicle, operation reorganization information to be performed for the first vehicle, and related information that corresponds to operation reorganization information to be performed for a second vehicle different from the first vehicle, in association with each other. The acquisition unit acquires first vehicle identification information for identifying the delayed first vehicle and position information indicating the position of the delayed first vehicle when a delay occurs in a first vehicle traveling on a route based on predetermined time information. The extraction unit extracts a combination of operation reorganization information indicating operation reorganization to be performed for the first vehicle and operation reorganization information indicating operation reorganization to be performed for the second vehicle associated with the related information based on the first vehicle identification information and the position information. " (See abstract). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2021-107189 A Summary of the Invention [Problem to be solved by the invention]

[0006] The technology described in Patent Document 1 extracts the details of traffic rescheduling based on the train number and the distance between the train's stations, but since the likelihood of a similar operational disruption occurring on the same train and at the same station location is low, there is a risk that appropriate traffic rescheduling details may not be output. Therefore, one aspect of the present invention proposes appropriate traffic rescheduling details for the current event. [Means for solving the problem]

[0007] In order to solve the above problem, one aspect of the present invention employs the following configuration: A timetable rescheduling proposal system includes a processor and a storage device, the storage device holds timetable rescheduling information, which is structured data indicating a plurality of past occurrence events that caused a request for timetable rescheduling for a train, the content of timetable rescheduling implemented for each of the plurality of past occurrence events, and values ​​of a plurality of types of attributes corresponding to each of the plurality of past occurrence events, the plurality of types of attributes including event notification attributes related to the event, traffic management attributes related to traffic management of the train, and equipment sensor attributes related to the measurement results by sensors of the state of a predetermined facility and / or a predetermined environment related to the train, and the processor holds the timetable rescheduling information for the train, which is structured data indicating the content of timetable rescheduling implemented for each of the plurality of past occurrence events, the plurality of types of attributes including event notification attributes related to the event, traffic management attributes related to traffic management of the train, and equipment sensor attributes related to the measurement results by sensors of the state of a predetermined facility and / or a predetermined environment related to the train, The system obtains a current occurrence event, which is a factor requiring processing, and the values ​​of the multiple types of attributes corresponding to the current occurrence event, obtains weights corresponding to each of the multiple types of attributes, calculates a weighted similarity between the current occurrence event and each of the multiple past occurrence events based on the values ​​of the multiple types of attributes corresponding to the current occurrence event and the values ​​of the multiple types of attributes corresponding to each of the multiple past occurrence events, extracts from the multiple past occurrence events those past occurrence events for which the calculated similarity is high based on predetermined conditions, and outputs the timetable reorganization content corresponding to the extracted past occurrence event in the timetable reorganization information. Effect of the Invention

[0008] According to one aspect of the present invention, it is possible to propose appropriate operation schedule adjustments in response to the current event.

[0009] Problems, configurations and effects other than those described above will become apparent from the following description of the embodiments. [Brief description of the drawings]

[0010] [Figure 1A] FIG. 1 is a block diagram showing an example of the overall configuration of a system including a timetable replanning proposal system in a first embodiment. [Figure 1B] FIG. 2 is a block diagram showing an example of a hardware configuration of the timetable rescheduling proposal system according to the first embodiment. [Diagram 2]2 is a diagram showing an example of a data configuration of structured data stored in a train rescheduling information DWH (Data WareHouse) in the first embodiment. FIG. [Diagram 3] FIG. 13 is a diagram showing an example of a screen configuration of a dashboard that displays information generated by a breaking news notification processing unit and a timetable replanning proposal processing unit in the first embodiment. [Figure 4] 4 is a flowchart showing an example of the overall processing by the timetable replanning proposal system in the first embodiment. [Diagram 5] 11 is a flowchart showing an example of a timetable replanning proposal process in the first embodiment. [Figure 6] FIG. 13 is a diagram illustrating an example of a screen configuration of a parameter tuning area in the first embodiment. [Figure 7A] FIG. 13 is a diagram illustrating an example of a screen configuration of a heat map display area in the first embodiment. [Figure 7B] FIG. 13 is a diagram showing an example of a screen configuration of a proposed operation reorganization group display area in the first embodiment. [Figure 8] FIG. 13 is a diagram showing an example of a screen configuration of a data analysis screen in the first embodiment. [Figure 9] FIG. 13 is a diagram showing an example of a screen configuration of a data analysis screen in the first embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. In this embodiment, the same components are generally designated by the same reference numerals, and repeated explanations are omitted. Note that this embodiment is merely an example for realizing the present invention, and does not limit the technical scope of the present invention. EXAMPLES

[0012] 1A is a block diagram showing an example of the overall configuration of a system including a timetable replanning proposal system. The timetable replanning proposal system 101 includes, for example, a data reception processing unit 105, an ETL (Extract Transform Load) tool unit 107, a news notification processing unit 109, and a timetable replanning proposal processing unit 110, which are all functional units. The timetable replanning proposal system 101 also holds an unstructured DL (Data Lake) 106 and a timetable replanning information DWH (Data WareHouse) 108.

[0013] The data reception processing unit 105 receives, for example, traffic management information 102, event notification information 103, and facility sensor information 104 from a system external to the timetable replanning proposal system 101 at any timing, and identifies the type of information received.

[0014] Traffic management information 102 includes data handled by the traffic management system, such as various timetables such as planned timetables, implemented timetables, and actual timetables, operation rescheduling history, and crew operation plans and history. The planned timetable refers to a timetable that is planned in advance. The implemented timetable is a timetable that reflects changes made to the planned timetable by dispatchers, etc. in real time when delays, accidents, etc. occur. The actual timetable is a timetable that shows the history of actual train operations.

[0015] The event notification information 103 includes data such as the type of event that has occurred and information on the location where the event has occurred. Note that an event in this embodiment is an event that, when it occurs, requires train rescheduling. For example, an event that causes a train service disruption is an example of an event in this embodiment.

[0016] The equipment sensor information 104 includes sensor data indicating the measurement results of the status of specific equipment related to the train, such as automatically measured train speed, air temperature, and the status of vehicle parts, as well as sensor data indicating the measurement results of the specific environment to which the train and the equipment belong.

[0017] The data reception processing unit 105 determines that a new event and an associated operation disruption have occurred upon receipt of the event notification information 103. When the data reception processing unit 105 receives the event notification information 103, the breaking news notification processing unit 109 notifies the dispatcher of an event occurrence breaking news on the display device 111, while simultaneously performing linked processing and data storage in the order of the unstructured DL 106, the ETL tool unit 107, the traffic rescheduling information DWH 108, and the traffic rescheduling proposal processing unit 110.

[0018] The data receiving processor 105 stores the received unstructured data of the operation management information 102, the event reporting information 103, and the facility sensor information 104 as unstructured data in the unstructured DL 106. In other words, data that may need to be referenced in the future is stored in the unstructured DL 106 for a long period of time in the state in which it was received by the data receiving processor 105.

[0019] The ETL tool unit 107 extracts information necessary for proposing timetable re-arrangements from the unstructured data stored in the unstructured DL 106, processes the extracted information into structured data that is easy to refer to, and stores the information in the timetable re-arrangement information DWH 108. For example, not all information among the planned timetable, actual timetable, and actual timetable contained in the traffic management information 102 is necessarily used for proposing timetable re-arrangements. In this case, the ETL tool unit 107 extracts only information to be stored in the timetable re-arrangement information DWH 108 from the unstructured DL 106.

[0020] The traffic rescheduling information DWH 108 holds structured data that is referenced when the traffic rescheduling proposal processor 110 proposes a group of traffic rescheduling proposals. The traffic rescheduling information DWH 108 manages events that occur during operation disruptions, the details of traffic rescheduling for the events that occur, and attributes such as the type of the event and the distance of occurrence. Input data required to configure the tables of the traffic rescheduling information DWH 108 is obtained from the traffic management information 102, event notification information 103, and facility sensor information 104.

[0021] The timetable replanning proposal processing unit 110 extracts one or more past occurrences that are similar to the current occurrence from the timetable replanning information DWH108, and outputs the contents of the timetable replanning carried out at the time of the extracted past occurrences to the display device 111 as a group of timetable replanning proposals.

[0022] The display on the display device 111 by the breaking news notification processing unit 109 and the timetable replanning proposal processing unit 110 is realized, for example, by using a dashboard application. The dispatcher operates on the dashboard, the timetable replanning proposal processing unit 110 performs data analysis according to the operation and displays the analysis results on the display device 111, and the dispatcher refers to the contents displayed on the display device 111 to determine the optimal timetable replanning group proposal.

[0023] 1B is a block diagram showing an example of the hardware configuration of a timetable rescheduling proposal system 101. The timetable rescheduling proposal system 101 is configured by a computer having, for example, a CPU (Central Processing Unit) 1001, a memory 1002, an auxiliary storage device 1003, an input I / F (Interface) 1004, an output I / F 1005, and a communication device 1006.

[0024] The CPU 1001 includes a processor and executes programs stored in the memory 1002. The memory 1002 includes a ROM (Read Only Memory), which is a non-volatile storage element, and a RAM (Random Access Memory), which is a volatile storage element. The ROM stores immutable programs (e.g., a Basic Input / Output System (BIOS)). The RAM is a high-speed, volatile storage element such as a DRAM (Dynamic Random Access Memory), and temporarily stores programs executed by the CPU 1001 and data used when the programs are executed.

[0025] The auxiliary storage device 1003 is a large-capacity non-volatile storage device such as a magnetic storage device (HDD (Hard Disk Drive)) or a flash memory (SSD (Solid State Drive)), and stores programs executed by the CPU 1001 and data used when the programs are executed. That is, the programs are read from the auxiliary storage device 1003, loaded into the memory 1002, and executed by the CPU 1001.

[0026] The input I / F 1004 is connected to input devices such as a keyboard and a mouse, and receives input from an operator. The output I / F 1005 is connected to a display device 111, a printer, and the like, and outputs the results of program execution in a format that can be viewed by the operator.

[0027] The communication device 1006 is a network interface device that controls communication with other devices according to a predetermined protocol, and may include a serial interface such as a Universal Serial Bus (USB).

[0028] A part or all of the programs executed by the CPU 1001 may be provided to the timetable replanning proposal system 101 from a removable medium (CD-ROM, flash memory, etc.) which is a non-transitory storage medium, or from an external computer equipped with a non-transitory storage device via a network, and stored in a non-volatile auxiliary storage device 1003 which is a non-transitory storage medium. For this reason, the timetable replanning proposal system 101 may have an interface for reading data from removable media.

[0029] The timetable rescheduling proposal system 101 is a computer system configured on a single physical computer, or on multiple logically or physically configured computers, and may operate in separate threads on the same computer, or may operate on a virtual computer constructed on multiple physical computer resources.

[0030] The CPU 1001 includes the above-mentioned functional units, that is, the data reception processing unit 105, the ETL tool unit 107, the breaking news processing unit 109, and the train schedule rescheduling proposal processing unit 110. For example, the CPU 1001 functions as the data reception processing unit 105 by operating according to a data reception processing program loaded into the memory 1002, and functions as the ETL tool unit 107 by operating according to an ETL tool program loaded into the memory 1002. The relationships between programs and functional units are similar for the other functional units included in the CPU 1001.

[0031] Some or all of the functions of the functional units included in the CPU 1001 may be realized by hardware such as an Application Specific Integrated Circuit (ASIC) or a Field Programmable Gate Array (FPGA).

[0032] The auxiliary storage device 1003 holds the unstructured DL 106 and the timetable replanning information DWH 108. Note that some or all of the information stored in the auxiliary storage device 1003 may be stored in the memory 1002, or in an external database connected to the timetable replanning proposal system 101.

[0033] In this embodiment, the information used by the timetable rescheduling proposal system 101 may be expressed in any data structure regardless of the data structure. For example, the information may be stored in a data structure appropriately selected from a table, a list, a database, or a queue.

[0034] Fig. 2 is a diagram showing an example of the data configuration of structured data stored in the train schedule rescheduling information DWH 108. In the example of Fig. 2, the structured data is represented by a table 201. Each row of the table 201 stores one occurring event (event subject), the details of train schedule rescheduling corresponding to the occurring event, and N types of attributes related to the occurring event.

[0035] The top row of table 201 contains one current occurrence f, the operation rescheduling content (to be determined) to be performed for the current occurrence f, and N types of attributes f corresponding to the current occurrence f. n (n=1, . . . , N). In addition, the second and subsequent rows of table 201 each store information indicating a past occurrence event g m (m=1, ,M) and the operation schedule changes made for past events g m,1 (m=1, ,M) and N types of attributes g m,n (m=1, ,M, n=1, ,N) is stored.

[0036] The timetable replanning proposal processing unit 110 calculates the similarity between the current occurrence event and the past occurrence event using the attribute values ​​of the current occurrence event and the past occurrence event in the table 201. The timetable replanning proposal processing unit 110 realizes the proposal of a group of timetable replanning proposals for the current occurrence event by evaluation using the similarity.

[0037] Since the content of possible train rescheduling varies depending on the event, the location, and the time of occurrence, it is desirable to store these details as attributes in table 201. Furthermore, even for the same event, the policy for train rescheduling may differ depending on the train operation situation. For example, if an event occurs during rush hour, priority is given to ensuring the number of trains in operation rather than eliminating delays in order to prevent congestion from worsening.

[0038] In this way, table 201 stores attributes based on the characteristics of train operation management, making it possible to propose operation rescheduling plans that take into account the operation situation.

[0039] 2, the event subject, type of event that occurred, distance in kilometers from section start point, distance in kilometers from section end point, date, and time are information obtained from event notification information 103 (all of which except for the event subject are examples of event notification attributes), the temperature is information obtained from facility sensor information 104 (an example of a facility sensor attribute), and the number of canceled trains and operation rescheduling details are information obtained from traffic management information 102. Note that the maintenance of the number of trains, whether or not there will be splitting or merging, whether or not the last train will be affected, and whether or not there will be a rush hour in the example of table 201 in FIG. 2 may be information entered by a dispatcher via an input device when an event occurs (for example, when an event occurrence alert is displayed as described later) (in this case, it is preferable to provide an area for entering this information in the event occurrence alert display area 301 as described later), or may be information obtained from traffic management information 102.

[0040] After the current event is resolved, the traffic replanning proposal processing unit 110 stores the traffic replanning content decided by the dispatcher for the current event in the traffic replanning information DWH 108, and then converts the current event f in the traffic replanning information DWH 108 into a past event g. m The process of changing it to one of the following is performed.

[0041] 3 is a diagram showing an example of a screen configuration of a dashboard that displays information generated by the breaking news processing unit 109 and the timetable replanning proposal processing unit 110. A dashboard 300 is displayed on the display device 111, and a dispatcher performs operations on the dashboard 300.

[0042] The dashboard 300 includes, for example, an event occurrence breaking news display area 301, a parameter tuning area 302, a heat map display area 303, and a traffic reorganization proposal display area 304. The event occurrence breaking news display area 301 is drawn by the breaking news notification processing unit 109, and the parameter tuning area 302, the heat map display area 303, and the traffic reorganization proposal display area 304 are drawn by the traffic reorganization proposal processing unit 110.

[0043] A display for notifying a commander that an event has occurred is displayed in the event occurrence quick report display area 301. When the data reception processing unit 105 receives the event communication information 103 related to the currently occurring event, the quick report processing unit 109 displays information related to the currently occurring event in the event occurrence quick report display area 301. The information displayed in the event occurrence quick report display area 301 includes, for example, the reporter of the currently occurring event, the type of the occurring event, and the date and time of occurrence.

[0044] The parameters set in the parameter tuning area 302 indicate weights used in calculating the similarity for extracting past occurrences similar to the current occurrence.

[0045] When proposing a rescheduling of train operations, in order to reflect the dispatcher's policy on rescheduling of train operations when extracting past events similar to the current event, the dispatcher can set in the parameter tuning area 302 which attribute of table 201 of the train operation rescheduling information DWH108 should be emphasized when extracting similar events.

[0046] A heat map of the similarity between the current event and past events that are similar to the current event is displayed in the heat map display area 303. In the example of Fig. 3, the time of the event and the distance in kilometers where the event occurred are used as axes in the heat map.

[0047] A list of past events that are similar to the current event (e.g., event subject, similarity, and attributes) is displayed in the traffic reorganization plan display area 304. The traffic reorganization plan display area 304 is a display for determining traffic reorganization plans that are in line with the dispatcher's traffic reorganization policy.

[0048] For example, when a record on the traffic reorganization group proposal display area 304 is selected (in FIG. 3, the topmost record filled with dots is selected), a point 3031 on the heat map in the heat map display area 303 indicating the similarity corresponding to the selected record is highlighted. Conversely, when an arbitrary point indicating the similarity on the heat map is selected, a record corresponding to the selected point on the traffic reorganization group proposal display area 304 may be highlighted.

[0049] Details of the parameter tuning area 302 will be described using FIG. 6, details of the heat map display area 303 will be described using FIG. 7A, and details of the proposed operation schedule reorganization group display area 304 will be described using FIG. 7B.

[0050] 4 is a flowchart showing an example of the overall processing by the timetable replanning proposal system 101. The data reception processing unit 105 receives data including the traffic management information 102, the event notification information 103, and the facility sensor information 104 (S401).

[0051] The data reception processing unit 105 judges whether the train operation status is an event occurring or a normal train operation without an event occurring (S402). Specifically, for example, the data reception processing unit 105 judges that an event is occurring when a record of the currently occurring event with undetermined operation schedule details is registered in the operation schedule information DWH 108, and judges that the train is operating normally when the record is not registered.

[0052] When the data reception processing unit 105 determines that the train operation status is normal operation (S402: normal operation), it determines whether the event notification information 103 is included in the data received in the most recent step S401 (S403). When the data reception processing unit 105 determines that the event notification information 103 is not included in the data received in the most recent step S401 (step S403: no), it stores the data received in the most recent step S401 (without processing) in the unstructured DL 106 (S404), and ends the entire process. By the process of step S404, the latest information is accumulated in the unstructured DL 106.

[0053] If the data reception processing unit 105 determines that the data received in the most recent step S401 contains the event notification information 103 (step S403: Yes), the breaking news processing unit 109 executes a process for drawing an event breaking news (S405). Specifically, the breaking news processing unit 109 executes drawing of the event breaking news display area 301 in FIG. 3.

[0054] The breaking news notification processing unit 109 may display, in addition to the information contained in the event contact information 103, facility sensor information 104 (e.g., temperature, etc.) related to the currently occurring event and operation management information 102 (e.g., whether there is a rush hour or not) related to the currently occurring event in the event occurrence breaking news display area 301.

[0055] When the facility sensor information 104 and the operation control information 102 are displayed in the event occurrence report display area 301, for example, the data reception processing unit 105 may request an external system, which is the information source of the information, to transmit the information. The process of step S405 allows the commander to be notified of the occurrence of an event.

[0056] Next, the data reception processing unit 105 stores (without processing) the data received in the most recent step S401 in the unstructured DL 106 (S406). If the data reception processing unit 105 determines that the train operation status is in an event (S402: event occurring), it transitions to step S406.

[0057] The ETL tool unit 107 structures the data by extracting, processing, and loading the data stored in the unstructured DL 106, and stores the structured data in a record related to the current event in table 201 of the train rescheduling information DWH 108 (S407).

[0058] The timetable replanning proposal processing unit 110 determines whether the timetable replanning proposal processing is being executed in step S409 described later (S408). When the timetable replanning proposal processing unit 110 determines that the timetable replanning proposal processing is being executed (S408: Executing), it ends the entire processing (the currently executing timetable replanning proposal processing continues to be executed as is).

[0059] When the timetable reorganization proposal processing unit 110 determines that the timetable reorganization proposal processing is not being executed (S408: Not Executing), it executes the timetable reorganization proposal processing (S409). In step S409, the timetable reorganization proposal processing unit 110 displays the contents of timetable reorganizations implemented for past occurrences that have a high similarity to the current occurrence in the timetable reorganization proposal group display area 304, thereby proposing a timetable reorganization group for the current occurrence event to the dispatcher.

[0060] In step S409, the timetable replanning proposal processing unit 110 draws a heat map of the similarity of the group of timetable replanning proposals in the heat map display area 303. The dispatcher determines the most suitable one from the group of timetable replanning proposals proposed on the dashboard 300. Details of the timetable replanning proposal processing in step S409 will be described later with reference to FIG.

[0061] It is desirable that the timetable replanning proposal processing unit 110 starts the timetable replanning proposal processing in step S409 after all attribute values ​​of the records related to the current event in table 201 of the timetable replanning information DWH 108 have been stored. If the timetable replanning proposal processing unit 110 determines immediately before the start of step S409 that there are missing attribute values ​​of the records related to the current event, for example, the data reception processing unit 105 may request an external system, which is the information source of the attributes, to transmit the values ​​of these attributes.

[0062] Next, when the dispatcher inputs the contents of the traffic replanning for the current event on a traffic replanning screen (not shown) displayed on a screen separate from the dashboard 300, the traffic replanning proposal processing unit 110 inputs the data of the record related to the current event f in the table 201 of the traffic replanning information DWH 108 (including the contents of the traffic replanning for the current event) to the past event g. m Then, the editing process is executed (S410), and the overall process is terminated.

[0063] 5 is a flowchart showing an example of the timetable replanning proposal process in step S409. The timetable replanning proposal processor 110 extracts a record related to the current occurrence from the table 201 of the timetable replanning information DWH 108 (S501).

[0064] The timetable rescheduling proposal processor 110 executes a statistical data preparation process (S502). In the statistical data preparation process in step S502, past statistics of weights to be displayed in the parameter tuning area 302 are calculated. Details of the past statistics of weights will be described later with reference to FIG.

[0065] The timetable rescheduling proposal processing unit 110 draws the parameter tuning area 302 based on the statistical data prepared in step S502, and determines the weight (parameter) of each attribute according to the operation by the dispatcher on the parameter tuning area 302 (S503).

[0066] The timetable rescheduling proposal processing unit 110 is configured to process M past occurrence events g indicated by the timetable rescheduling information DWH 108. m The process of step S505 is executed for each of m=1, . . . , M (S504). m Similarity S(f,g m ) is calculated (S505). m ) is calculated, for example, according to the following formula (1).

[0067]

number

[0068] Similarity S(f,g m ) is an example of a weighted similarity where weights are assigned to attributes. Here, w 1 ,···,w N are the attributes f 1 ,f N is the weight corresponding to the value input by the dispatcher in step S503, and 0≦w n ≦1(n=1, . . . , N).

[0069] In accordance with the dispatcher's traffic rescheduling policy, a large weight is assigned to an attribute with high importance, and a small weight is assigned to an attribute with low importance. The possible range of similarity defined by equation (1) is 0≦S(f,g m ) ≦ 1 (m = 1, , M), and the smaller the value, the smaller the f and g m are not similar, and the larger the value, the more similar f and g are. m Among the attribute values ​​stored in table 201, values ​​defined as character strings are converted into numeric strings according to a predetermined algorithm before the similarity is calculated.

[0070] The timetable rescheduling proposal processing unit 110 calculates the similarity S(f, g m ) is extracted from the M past events in descending order of similarity (the degree of similarity is high based on a predetermined condition) (S506). Specifically, for example, the timetable replanning proposal processing unit 110 may extract a predetermined number of past events in descending order of similarity from the M past events, or may extract a predetermined ratio of past events in descending order of similarity, or may extract past events whose similarity is equal to or greater than a predetermined value.

[0071] The timetable replanning proposal processing unit 110 executes a drawing process for data analysis of the past occurrence events extracted in step S506 (S507). In step S507, the timetable replanning proposal processing unit 110 draws the heat map display area 303 and the timetable replanning group proposal display area 304. Also in step S507, the timetable replanning proposal processing unit 110 displays a data analysis screen, which will be described later with reference to Figures 8 and 9, on the display device 111.

[0072] The dispatcher checks the proposed timetable re-arrangement plans by the timetable re-arrangement proposal system 101 on the dashboard 300, while performing data analysis on the screens of Figures 8 and 9 described below, and confirms which of the proposed timetable re-arrangement plans is optimal. This enables the dispatcher to further narrow down the proposed timetable re-arrangement plans, and prevents the dispatcher from selecting an inappropriate timetable re-arrangement plan.

[0073] The timetable replanning proposal processing unit 110 determines a group of timetable replanning proposals that are optimal for the current event in accordance with the input by the dispatcher (S508), and ends the timetable replanning proposal processing.

[0074] As described above, the timetable replanning proposal processing unit 110 extracts past occurrence events with a high degree of similarity to the current occurrence event, and can quickly determine a group of timetable replanning proposals because the amount of calculation required to calculate the similarity is small. In addition, since the dispatcher can select weights based on statistics of past weights, similar past occurrence events can be extracted with high accuracy. Furthermore, if a certain number of past occurrence events have been accumulated, there is a high possibility that a past occurrence event with the same attribute value as the current occurrence event does not exist, but there is a high possibility that a past occurrence event with a high degree of similarity exists. Therefore, the timetable replanning proposal processing unit 110 can extract past occurrence events and can therefore propose timetable replanning contents to the dispatcher with a high probability.

[0075] 6 is a diagram showing an example of the screen configuration of the parameter tuning area 302 displayed in step S503. The parameter tuning area 302 includes, for example, a past statistics display area 601, a weight selection area 602, a representative attribute display area 604, and an enter button 605.

[0076] The representative attribute display area 604 displays information indicating a representative attribute included in the attributes indicated in the table 201 of the train schedule reconciliation information DWH 108 and the value of the representative attribute. The value of the representative attribute displayed in the representative attribute display area 604 is, for example, a value that matches (or is similar according to a predetermined condition) the value of the representative attribute of the currently occurring event.

[0077] In the case of train schedule disruptions, if it is possible to identify what event occurred, when, and where it occurred, it is highly likely that the direction of operation rescheduling for that event can be narrowed down. For this reason, the type of event that occurred, the line, and the time are suitable as representative attributes, and in the example of Figure 6, the type of event that occurred, the line, and the time are displayed as representative attributes.

[0078] In addition, attributes other than the type of event, route, and time may be included in the representative attributes, and the representative attributes may be specified by the dispatcher. For example, the representative attributes may include a mileage indicating the location where the event occurred (for example, the kilometerage of the section start point, the kilometerage of the section end point, or the center position between the kilometerage of the section start point and the kilometerage of the section end point).

[0079] The past statistics display area 601 displays statistical information on weights (for each attribute, not limited to the representative attribute) adopted in the process of suggesting train schedule reorganization for past events that have the same value as the value of the representative attribute shown in the representative attribute display area 604 or a similar value according to predetermined conditions. In the example of Fig. 6, the past statistics display area 601 displays the average value of the weights as statistical information, but other statistical information on the weights, such as the mode value and standard deviation value of the weights, may also be displayed.

[0080] Statistical information to be displayed in the past statistics display area 601, which corresponds to the representative attribute indicated in the representative attribute display area 604, is generated in step S502.

[0081] The example in Figure 6 shows that the average weighting of time adopted in the operation schedule rescheduling proposal process for a past event in which the event type is "traffic signal failure", the route is "α line", and the time is "14:00" is in the range of 1 to 0.9, and the average weighting of the obstruction location is in the range of 0.5 to 0.6.

[0082] By displaying the past statistics display area 601, even an inexperienced commander can select an appropriate weighting while referring to past statistical information on weighting, and thus can extract past occurrences similar to the current occurrence with high accuracy.

[0083] The weight selection area 602 is an area for selecting the weight of each attribute. When a pull-down corresponding to each attribute in the weight selection area 602 is selected, a pull-down list 603 for selecting the weight for the attribute is displayed. In the example of Fig. 6, five levels of priority ("highest", "high", "medium", "low", and "ignore") are displayed in the pull-down list 603.

[0084] For example, when "highest" is selected as the priority in the pull-down list 603, the weight of the attribute is set to w n = 1 is set as the weight of the attribute for which "high" is selected. n = 0.75 is set, and when "Medium" is selected, the weight of the attribute is set to w n = 0.5 is set, and when "Low" is selected, the weight is w n = 0.25, and when "ignore" is selected, the weight of the attribute is set to w n =0 is set.

[0085] The number of weighting levels for each attribute is not limited to five, and the weighting values ​​corresponding to each level are not limited to the above example. The commander may directly input the weighting value to specify any value between 0 and 1. The initial weighting value in the weighting selection area 602 may be a value that is closest to the statistic shown in the past statistics display area 601 (the average value in the example of FIG. 6).

[0086] In the weight selection area 602, the weight w for each attribute is n After the weight w for each attribute is set, when the decision button 605 is selected, the process of step S503 ends and the process proceeds to step S504. n After the weights are set, when the decision button 605 is selected, the weights corresponding to each attribute of the currently occurring event are saved (for example, the weights are managed in the train schedule rescheduling information DWH108 (for example, a table other than table 201)), and the weights set for the currently occurring event are reflected in the statistical data preparation process in the next step S502.

[0087] Fig. 7A is a diagram showing an example of the screen configuration of the heat map display area 303 displayed in step S507. Fig. 7B is a diagram showing an example of the screen configuration of the operation reorganization group proposal display area 304 displayed in step S507.

[0088] In the heat map display area 303, for example, a heat map 701, a data analysis mode switching button 702, and a decision button 703 are displayed. The heat map 701 shows the similarity S(f,g m In the heat map 701, the higher the similarity of a past occurrence event to the current occurrence event f, the higher the brightness of the display.

[0089] In the case of railway operation disruptions, if a highly similar past event occurred at a similar time and location, it is highly likely that a similar rescheduling plan can be applied to the current event. Therefore, in the example of Figure 7A, in order to help the dispatcher find the optimal rescheduling plan for the current event, kilometers (e.g., the kilometers from the start of the section, the kilometers from the end of the section, or the center position between the kilometers from the start of the section and the kilometers from the end of the section) are used as the horizontal axis of the heat map 701, and time is used as the vertical axis of the heat map 701.

[0090] Therefore, in the example of Figure 7A, the similarity S(f,g m When and where a past occurrence event with a high score occurred is visually displayed in an easy-to-understand manner. Note that other attributes may be used for the horizontal and vertical axes of the heat map 701, and the attributes of the horizontal and vertical axes may be specified by the commander.

[0091] For example, as mentioned above, in the case of a train schedule disruption, if it is possible to identify what event occurred, when, and where it occurred, it is highly likely that the direction of operation rescheduling in response to the event can be narrowed down, so the type of event that occurred and the line may be used as attributes for the horizontal and vertical axes.

[0092] The event subject of the past event extracted in step S506, the similarity between the current event and the past event calculated in step S505, and the attribute value corresponding to the past event are displayed in the operation reorganization group proposal display area 304. The operation reorganization group proposal display area 304 may display all or some of the attributes in the table 201 of the operation reorganization information DWH108.

[0093] As described in FIG. 3, for example, when a point on the heat map 701 is selected, the record corresponding to the point on the traffic reorganization plan display area 304 is highlighted. Here, consider a case where there are multiple traffic reorganization plans with the same degree of similarity on the heat map 701. In such a case, in consideration of the above characteristics, the past occurrence event g that occurred in a section and at a time closer to the current occurrence event f is selected. m However, since it is highly likely that this is a more appropriate group of train rescheduling plans for the current event f, an analysis using the example in Figure 7A is effective.

[0094] When the data analysis mode switching button 702 is selected, the screen transitions to the data analysis screen shown in Figures 8 and 9. When the decision button 703 is selected with a past occurrence event selected on the heat map 701 or on the traffic reorganization group proposal display area 304, the screen transitions to step S508.

[0095] In step S508, the operation reorganization content corresponding to the selected previously occurring event may be automatically selected as the operation reorganization content for the current event, or the dispatcher may manually determine the operation reorganization content by referring to the operation reorganization content corresponding to the selected previously occurring event.

[0096] 8 and 9 are diagrams showing examples of the screen configuration of the data analysis screen. The data analysis screen is a screen for the dispatcher to analyze the group of timetable rescheduling proposals proposed by the timetable rescheduling proposal processing unit 110 from the viewpoint of various attributes.

[0097] On the data analysis screen, the current occurrence event f and past occurrence event g m Similarity S(f,g m ) for each past occurrence g m This provides command staff with a means to color-code the data and analyze it using various attributes as the analytical axis.

[0098] The data analysis method used is, for example, OLAP (Online Analytical Processing) analysis, which is represented by dicing, slicing, drilling down, and drilling up on the heat map mentioned above. These analysis results are displayed graphically on the data analysis screen, allowing the dispatcher to visually analyze the proposed rescheduling plans in an easy-to-understand manner.

[0099] As an example, let us assume that the current event is a curved track caused by high temperatures on a hot day, making it impossible for trains to run. In this case, the value of "track fault" is input as the event type of the current event.

[0100] Here, the traffic rescheduling proposal system 101 proposes a group of traffic rescheduling plans based on similar past occurrences, but there is a possibility that the proposed plans may include a traffic rescheduling plan for a track fault caused by a factor other than high temperature, such as a track circuit fault. Therefore, there is a risk that the dispatcher will carry out an incorrect traffic rescheduling. In this way, when an inappropriate traffic rescheduling plan is mixed in with the multiple traffic rescheduling plans, OLAP analysis using the data analysis screen is effective.

[0101] Screen 801 displays a three-dimensional heat map of the similarities of the past occurrence events extracted in step S506 (for convenience of illustration, cubes are shown in the examples of screen 801 in FIG. 8 and screen 901 in FIG. 9 described below, but in reality, the higher the similarity of a past occurrence event with the current occurrence event f, the brighter the dot displayed).

[0102] In the three-dimensional heat map on the screen 801, the Z axis represents the type of event that occurred, the Y axis represents time, and the X axis represents kilometers. Note that the commander may be able to select any attribute as an attribute of the three-dimensional heat map.

[0103] When the dispatcher selects a value for an equipment failure (corresponding to the current event, for example) from the Z-axis (occurring event type) on screen 801, the timetable rescheduling proposal processing unit 110 slices a two-dimensional graph with the Y-axis (time) and X-axis (kilometers) with the selected equipment failure value fixed, and plots it on screen 802. The group of timetable rescheduling proposals shown in the two-dimensional graph on screen 802 are distributed over various time periods, and trends in similar past occurrences cannot be seen.

[0104] Here, for example, if the dispatcher inputs an input to change the Y-axis attribute from time to date, the timetable replanning proposal processing unit 110 redraws a two-dimensional graph on the screen 803 with the Y-axis changed to date by drilling up the Y-axis. On the screen 803, the dispatcher can visually determine that there is only one timetable replanning proposal that falls outside the period of extremely hot days (i.e., one that is inappropriate as a timetable replanning to be implemented for the current event), and therefore the dispatcher can avoid selecting a timetable replanning proposal that falls outside the period of extremely hot days. This is particularly useful for dispatchers with little experience.

[0105] Another example of data analysis will be described. In the three-dimensional heat map on screen 901 in Fig. 9, the Z axis represents temperature, the Y axis represents time, and the X axis represents kilometers. Here, since track bending events frequently occur on extremely hot days, it is assumed that the dispatcher selects data on the Z axis (temperature) on screen 901 that is 30 degrees Celsius or higher.

[0106] At this time, the timetable replanning proposal processing unit 110 slices a two-dimensional graph on the Y-axis (time) and X-axis (kilometers) for data with temperatures of 30 degrees Celsius or higher, and draws it on screen 902. Furthermore, the timetable replanning proposal processing unit 110 slices a two-dimensional graph on the Y-axis (time) and X-axis (kilometers) for data with temperatures of less than 30 degrees Celsius, and draws it on screen 903.

[0107] By comparing the sliced ​​two-dimensional graphs, it can be visually determined that most of the proposed traffic rescheduling plans are distributed on the two-dimensional graph displayed on screen 902, but only one is on the two-dimensional graph displayed on screen 903. By using such an analysis method, it is possible to prevent the dispatcher from making erroneous traffic rescheduling plans. This is particularly useful for dispatchers with little experience.

[0108] In addition, the operation schedule reorganization plan display area 304 and the OK button 703 may be displayed on each screen in FIG. 8 and FIG.

[0109] The present invention is not limited to the above-mentioned embodiment, and various modifications are included. For example, the above-mentioned embodiment has been described in detail to clearly explain the present invention, and is not necessarily limited to those having all the configurations described. It is also possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.

[0110] In addition, the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. In addition, the above-mentioned configurations, functions, etc. may be realized in software by a processor interpreting and executing a program that realizes each function. Information such as the program, table, file, etc. that realizes each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

[0111] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are connected to each other. [Explanation of symbols]

[0112] 101 Traffic rescheduling proposal system, 102 Traffic management information, 103 Event notification information, 104 Facility sensor information, 105 Data reception processing unit, 106 Unstructured DL, 107 ETL tool unit, 108 Traffic rescheduling information DWH, 110 Traffic rescheduling proposal processing unit, 300 Dashboard, 302 Parameter tuning area, 303 Heat map display area, 304 Traffic rescheduling group proposal display area, 1001 CPU, 1002 Memory, 1003 Auxiliary storage device, 1004 Input I / F, 1005 Output I / F, 1006 Communication device

Claims

1. A train rescheduling proposal system, A processor and a storage device are provided. The storage device holds train rescheduling information, which is structured data indicating a plurality of past occurrence events that caused a request for train rescheduling, details of train rescheduling implemented for each of the plurality of past occurrence events, and values ​​of a plurality of types of attributes corresponding to each of the plurality of past occurrence events; The multiple types of attributes include an event communication attribute related to an event, a traffic management attribute related to traffic management of the train, and an equipment sensor attribute related to a measurement result by a sensor of a state of a predetermined equipment and / or a predetermined environment related to the train, The processor, obtain a current occurrence event that is a cause for requesting the train rescheduling and values ​​of the plurality of types of attributes corresponding to the current occurrence event; Obtaining a weight corresponding to each of the plurality of types of attributes; calculating a weighted similarity between the currently occurring event and each of the plurality of previously occurring events based on the values ​​of the plurality of types of attributes corresponding to the currently occurring event and the values ​​of the plurality of types of attributes corresponding to each of the plurality of previously occurring events, and the weights; extracting from the plurality of past occurrence events a past occurrence event having a high degree of similarity calculated based on a predetermined condition; A timetable rescheduling proposal system that outputs timetable rescheduling content that corresponds to the extracted past occurrence event in the timetable rescheduling information.

2. The system for suggesting timetable rescheduling according to claim 1, connected to an input device and a display device; the storage device holds, for each of the plurality of past occurrence events, a history of the weightings adopted when the train schedule replanning was proposed for the past occurrence event; The processor, calculating a predetermined statistic of the weights adopted when proposing the train schedule replanning for the plurality of past occurrence events based on the history of the weights; displaying the calculated predetermined statistical quantity on the display device; A train schedule rescheduling proposal system that obtains the weights for calculating the weighted similarity through input to the input device.

3. The system for suggesting timetable rescheduling according to claim 2, The plurality of types of attributes include a predetermined representative attribute, The processor, referring to the train schedule rescheduling information, identifying a past occurrence event whose representative attribute value matches the current occurrence event; A timetable replanning proposal system that calculates the specified statistics of the weighting adopted when proposing the timetable replanning for the identified past occurrence event based on the weighting history.

4. The system for suggesting timetable rescheduling according to claim 3, A train schedule rescheduling proposal system, wherein the representative attributes include at least one of the type of the event, the route on which the event occurred, the distance traveled by the event, and the time at which the event occurred.

5. The system for suggesting timetable rescheduling according to claim 1, The storage device maintains a data lake that stores unstructured data; The processor, receiving event notification information indicating the event notification attribute, traffic management information indicating the traffic management attribute, and equipment sensor information indicating the equipment sensor attribute, all of which are the unstructured data; The event contact information, the operation management information, and the facility sensor information are stored as the unstructured data in the data lake; A timetable rescheduling proposal system that processes the event notification information, the operation management information, and the facility sensor information into the structured data and stores values ​​of the multiple types of attributes in the timetable rescheduling information.

6. The system for suggesting timetable rescheduling according to claim 1, connected to a display device, The processor of this system for suggesting train schedule rescheduling generates a heat map of the similarity corresponding to the extracted past events, the heat map being centered on a specified attribute included in the multiple types of attributes, and displays the heat map on the display device.

7. The system for suggesting timetable rescheduling according to claim 6, A train schedule rescheduling suggestion system, wherein the specified attributes include at least one of the type of the event, the route on which the event occurred, the distance traveled by the event, and the time at which the event occurred.

8. The system for suggesting timetable rescheduling according to claim 6, the processor performs an analysis on the heat map to perform at least one of dicing, slicing, drilling down, and drilling up; A train schedule rescheduling proposal system that displays the results of the analysis on the display device.

9. A method for proposing traffic replanning using a traffic replanning proposal system, comprising: The timetable rescheduling proposal system includes a processor and a storage device, The storage device holds train rescheduling information, which is structured data indicating a plurality of past occurrence events that caused a request for train rescheduling, details of train rescheduling implemented for each of the plurality of past occurrence events, and values ​​of a plurality of types of attributes corresponding to each of the plurality of past occurrence events; The multiple types of attributes include an event communication attribute related to an event, a traffic management attribute related to traffic management of the train, and an equipment sensor attribute related to a measurement result by a sensor of a state of a predetermined equipment and / or a predetermined environment related to the train, The method for proposing train rescheduling comprises: The processor acquires a current occurrence event that is a cause for requesting the train rescheduling and values ​​of the multiple types of attributes corresponding to the current occurrence event; The processor obtains a weight corresponding to each of the plurality of types of attributes; the processor calculates a weighted similarity between the currently occurring event and each of the plurality of previously occurring events based on the values ​​of the plurality of types of attributes corresponding to the currently occurring event, the values ​​of the plurality of types of attributes corresponding to each of the plurality of previously occurring events, and the weights; The processor extracts, from the plurality of past occurrence events, a past occurrence event having a high degree of similarity calculated based on a predetermined condition; A method for proposing timetable replanning, in which the processor outputs timetable replanning content corresponding to the extracted past occurrence event in the timetable replanning information.

Images