JP2016190619A - Operation information distribution device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an operation information distribution device for enabling a receiving side to correspond properly to a change in an operation plan by enabling a receiving side to decide the reliability of a prediction value of operation information based on a change in an operation diagram.SOLUTION: On the basis of a change in the running plans of a plurality of movers, the prediction information in the operation of a movable body, and a fixedness of the prediction information is set. On the basis of the prediction information and the fixedness, the mode of the running related information is determined so that the operation related information determined is distributed to a crew section 1600 related and a plurality of stations 1700.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an operation information distribution system used for operation management work of a moving body such as a train.

  When the train operation is disturbed due to bad weather or a vehicle failure, the train operation manager changes the timetable to quickly restore the train operation. The schedule not only shows the train operation plan, but is also closely related to the work plan of the crew on board the train.Furthermore, the railway passengers move on and off the train according to the schedule. It is also necessary to look at the guidance methods for coordinating with related departments and guiding passengers accurately. A technique for distributing train operation prediction information has been proposed as a technique for performing coordination with related departments for timetable changes and passenger guidance (for example, Japanese Patent Laid-Open No. 2014-42141).

JP 2014-43141 A

  The operation information distribution system simulates how the schedule for the next several hours will change in response to changes in the schedule based on train operation delays, etc., and obtains predicted times related to operations. The related departments can take appropriate actions based on the predicted time. However, when the schedule is frequently changed due to accumulation of train operation delays, there is a problem that the receiving side cannot determine how reliable the predicted time is. Therefore, the present invention makes it possible for the receiving side to appropriately cope with the change of the operation plan by enabling the receiving side of the operation information to determine the reliability of the predicted value of the operation information based on the change of the operation schedule. The purpose is to provide an operation information distribution system capable of

  In order to achieve the object, the present invention relates to operation information distribution for distributing operation-related information related to operations of a plurality of mobile objects, and the mobile object is based on a change in the operation plan of the mobile object. Set prediction information in the operation, set a certainty degree for the prediction information, determine the mode of the operation related information based on the prediction information and the certainty degree, and relate the determined operation related information By delivering first, it becomes possible for the receiving side of the operation information to judge the reliability of the predicted value of the operation information based on the distribution information reflecting the degree of definiteness. It becomes possible to respond.

  According to the present invention, the reliability of the prediction information of the operation information can be determined by the receiving side of the operating information, and the receiving side can appropriately cope with the change of the operating plan.

FIG. 1 is a block diagram of an operation information distribution system. FIG. 2 is a flowchart showing processing of the operation information distribution system. FIG. 3 is a diagram based on the operation plan stored in the diagram table and the route data stored in the route data table. FIG. 4 is a diagram based on the result of the operation prediction calculation. FIG. 5 is a network diagram based on the train operation network table. FIG. 6 is a train operation network table. FIG. 7 is a prediction definiteness definition table. FIG. 8 is a diagram change pattern table. FIG. 9 is a flowchart showing the process of the prediction certainty calculator. FIG. 10 is a prediction certainty history table. FIG. 11 is a route proposal table. FIG. 12 is an example of operation information distributed to the crew by the prediction information transmission unit. FIG. 13 is an example of operation information distributed by the prediction information transmission unit for passenger guidance. FIG. 14 is a characteristic diagram illustrating an example of the output of the prediction certainty verification unit.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings, taking operation information distribution in a train operation system as an example. FIG. 1 is a block diagram of an operation information distribution system. The operation information distribution system is configured as a distributed computer system, and mainly includes a computer system 1 and a storage system 2.

  The computer system 1 includes a central processing unit 1100 that executes an operation information distribution control program, and the storage device 2 includes a database 1200 that stores data. Furthermore, the computer system 1 includes an input device 1300 that receives input from a user, and a display device 1400 that displays various GUIs including a diagram.

  The operation information distribution system further includes a network 1500 as a communication route, a crew ward device 1600 that transmits business information to crew members such as conductors and drivers, and a station device that obtains train times and provides information to passengers. 1700.

  By executing the operation information distribution control program, the central processing unit 1100 targets a range of a predetermined time from the present (for example, a range of several hours in the future) due to a change in the operation plan accompanying a train delay, etc. Prediction calculation unit 1110 that obtains prediction information (prediction time) related to train station arrival time and station departure time, train operation net generation unit 1120 that generates structural data representing the causal relationship between prediction times, and prediction time Prediction variation factor setting unit 1130 for setting variation factors, prediction determinism calculation unit 1140 for calculating the degree (determinism) that the prediction time can be a fixed value, and proposals for changing the crew route based on the prediction time Route proposal generation unit 1150 for generating information, prediction time, prediction fluctuation probability (determinism), prediction information transmission unit 1160 for transmitting route change proposal information to crew ward or station, prediction confirmation for verifying the accuracy of the prediction time Every time The verification unit 1170 is realized.

  The database 1200 includes a diagram data table 1210 for storing schedule data of operation plans and actual times, a route data table 1220 for storing crew operation plan data, a train operation network table 1230 for storing causal relationship data between predicted times, Prediction determinism definition table 1240 that stores fluctuation factors of prediction time and fluctuation probability data (determinism), diamond change pattern that is assumed to be implemented in the future for the prediction time, and diamond change pattern that stores execution probability data A table 1250, a prediction certainty degree history table 1260 for storing the prediction time and the certainty degree data for the prediction time, and a route proposal table 1270 for storing crew plan change plan data are recorded.

  The input device 1300 is a general-purpose input device for a computer such as a mouse or a keyboard. The display device 1400 is an output device such as a display. The crew ward device 1600 is a device provided in each crew ward, and transmits information necessary for work to the crew, such as operation information and route change information. The station device 1700 is a device provided in each station, and guides the actual time and operation information to passengers. The central processing unit 1100, the crew ward device 1600, and the station device 1700 can transmit and receive information via the wide area network 1500.

  FIG. 2 is a flowchart showing the operation of each processing unit of the central processing unit 1100. In step 2100, the prediction calculation unit 1110 acquires and sets train operation prediction times for several hours in the future based on information in the diagram data table 1210.

  FIG. 3 is a diagram illustrating the data (operation plan) in the diagram data table 1210 and the route data table 1220, and shows that the trains make a round trip between the A station and the B station. The vertical axis represents the travel section, and the horizontal axis represents time. Diagonal lines indicate train operations. For example, line segment 3111 represents a train running from station A to station B. In addition, the line segment connecting the diagonal lines represents that the train is operated by the same vehicle, and train a (3111), train b (3112), and train c (3113) represent trains of the same vehicle. ing. Similarly, train d (3121), train e (3122), and train f (3123) are the same vehicle, and train g (3131) and train h (3132) are the same vehicle.

  The thick line connecting the diagonal lines indicates the crew's operation plan, line segment 3211 indicates that train a (3111) and train e (3122) are connected by the same crew member, and line segment 3212 indicates train e (3122) And the train f (3123) are transferred by the same crew member. Therefore, the same crew member is on board the train a (3111), the train e (3122), and the train f (3123). Similarly, the same member is in charge of train d (3121), train g (3131), and train c (3113), and the same member is in charge of train b (3112) and train h (3132). The line segment 3300 represents the current time line and indicates that the current time is 9:00.

  FIG. 4 is a diagram in which the diagram shown in FIG. 3 is changed based on the prediction time obtained by the prediction calculation unit 1110. A situation is shown in which the departure time of station A of train a (4111) is delayed by 30 minutes until 9:00, and the departure time of station B of train e (4122) is delayed by 9:15. Note that the delay of the train a is a delay that appears as a running record, and the delay of the train e is a delay that the operator inputs to the central processing unit 1100 because a delay is expected.

  In response to the delay of train a and train e, the other trains are also delayed. For example, train b (4112) is delayed due to the delay of train e (4122), which is the preceding departure train, and train c (4113) receives the delay of train b (4112) that runs on the same vehicle. There is a delay. The predicted time also indicates the influence on the flight path of the crew member, that is, the presence or absence of the obstacle to the crew crew task. For example, the route 4211 indicates that the same crew member is connected to the train a (4111) and the train e (4122), but the connection time is not sufficient, and the route 4231 is The same crew should transfer from train g (4131) to train c (4113), but train c (4113) will depart before arrival of train g (4131), indicating that connection cannot be made .

  In step 2200 of FIG. 2, the train operation network generation unit 1120 generates a train operation network. The train operation net generation unit 1120 displays the causal relationship existing between the predicted times on the display device 1400 as graph data. FIG. 5 is a diagram showing a causal relationship between predicted times as a network graph. A node indicated by a circle indicates a predicted time, and a link indicated by an arrow indicates a causal relationship between the predicted times. Symbols in the nodes indicate trains and stations. FIG. 5 shows the prediction calculation result shown in FIG. 4 as a network graph. For example, the node 5111 represents the predicted time of the A station train a (4111). Link 5202 indicates that there is a causal relationship between the predicted time of A station train a (4111) and the predicted time of B station train a (4111) indicated by node 5112, that is, train a (4111) is at A station. It shows the causal relationship regarding the travel time from to the station B.

  There are two types of links, a normal line and a bold line, and the link indicated by the bold line indicates that it contributed to the prediction time determination. For example, for the predicted time of the B station train b (4122) indicated by the node 5113, the dependency link 5203 between the arrival time (node 5112) of the B station of the train a (4111) operated by the same vehicle, Two of the dependency links 5205 to the departure time (node 5131) of train e (4122), which is the preceding train train, are connected, but it contributes to the prediction time determination of B station train b (4112) It is shown that this is due to the causal relationship (dependency link 5205) from the departure time of station B of train e (4122) indicated by a bold line.

  FIG. 6 shows the configuration of the train operation network table 1230, which corresponds to the network graph of FIG. One link corresponds to a record in each column. The record in each column includes a start node 6110 that represents a node on the link start side, an end node 6120 that represents a node on the end point, and a link type 6130 that represents a link type. The branch constant 6140 indicating the predicted time difference to be protected between the start node and the end node, and the decision value flag 6150 indicating whether or not the link has contributed to the predicted time determination of the end node. Yes. For example, data 6202 represents link data indicating the causal relationship existing between the predicted time of the A station train a (4111) and the predicted time of the B station train a (4111), and the causal relationship between the two predicted times ( (Link type) is the time between stations, and requires a time difference of 10 minutes, indicating that it contributes to the determination of the predicted time of station B train a (4111).

  In step 2300 of FIG. 2, the prediction variation factor setting unit 1130 determines whether or not there is a factor that causes the prediction time acquired in step 2100 to fluctuate. If the determination is affirmative, the variation factor is set. There are two types of fluctuation factors: those that depend on the factors that determine the predicted time, and those that depend on changes in the diagram. The prediction determinism definition table 1240 (FIG. 7) stores a variation factor that depends on the determination factor of the prediction time, and the diagram change pattern table 1250 (FIG. 8) stores a variation factor that depends on a diagram change pattern that may occur in the future. .

  FIG. 7 shows a table configuration of the prediction deterministic definition table 1240, which is composed of records of a prediction time determination factor 7110 and a prediction determinism 7120 indicating the accuracy of the prediction time set for the determination factor. Yes. For example, the data 7210 indicates that the delay of the performance is a determining factor, and the prediction definiteness, in other words, the expected value of accuracy with respect to the predicted time is 100%, that is, the predicted time is correct. On the other hand, the data 7220 indicates that the accuracy of the predicted time is reduced to 90% when the determination factor of the predicted time is an estimation (probability) by the operator.

  FIG. 8 is a diagram showing a table configuration of the diagram change pattern table 1250. The record is composed of a record of a change rule 8110 indicating a pattern change pattern and a prediction determination degree 8120 indicating a probability that the diamond change according to the pattern is executed by the operator. For example, data 8210 indicates that even if the previous train is delayed by 20 minutes or more, the probability that the next train using the same vehicle will be operated is 40%. If the previous operation train is delayed for more than 20 minutes, the schedule change of canceling the next train is executed with a probability of 60%. The meaning of the data 8220 is also as shown in FIG.

  In step 2300 of FIG. 2, the prediction variation factor setting unit 1130 sets the prediction certainty defined in the data of each table based on the prediction certainty definition table 1240 and the diagram change pattern table 1250 for the prediction time. In the prediction results shown in FIG. 4, since the actual delay is the determining factor for the A station train a (4111) and the expected delay is the determining factor for the B station train e (4122), the predictive variation factor setting unit 1130 From the data of the definition level definition table 1240, the prediction accuracy is set to 100% and 90% for each prediction time.

  Furthermore, since the train a (4111) is delayed by 20 minutes or more, the prediction variation factor setting unit 1130 is configured to train the train b (4112), which is the next train traveling on the same vehicle, from the diagram change definition table 1250. The prediction determinism is set to 40% with respect to the prediction time of b (4112). Similarly, since train f (4123) is delayed due to the influence of train h (4132), which is the preceding train, the prediction certainty factor is set to 50% with respect to the prediction time of train f (4123).

  In step 2400 of FIG. 2, the prediction certainty calculation unit 1140 calculates the certainty of the prediction time for all the prediction times based on the prediction variation factor set in step 2300. The calculation of the degree of definiteness of the predicted time is executed by propagating the set degree of predicted definiteness to other predicted times based on the causal relationship between the predicted times using the data of the train operation network table 1230.

  FIG. 9 is a flowchart showing the operation of the predicted time determination degree calculation unit 1140. In step 9100, the predicted time determination degree calculation unit 1140 registers all the nodes (FIG. 5) in the queue. In step 9200, it is determined whether the queue is an empty set. If the queue is an empty set, it is determined that the prediction determinism has been set for all prediction times, and the process is terminated.

  If the predicted time determination degree calculation unit 1140 determines that the set is not an empty set, it proceeds to step 9300 and acquires one node from the queue (referred to as node N). In Step 9400, link data whose determination value flag is “Yes” is acquired from the train operation network table 1230 among the links whose acquired node is the starting node.

  In step 9500, the predicted time determination degree calculation unit 1140 refers to the table to obtain the node n that is the end point of the link data. In step 9600, the predicted time determination degree calculation unit 1140 compares the prediction determination degrees of the node N and the node n. If the prediction determination degree of the node N is equal to or higher than the prediction determination degree of the node n, the prediction determination of the node n is performed. It is determined that the update is unnecessary, and the process returns to step 9200. When the predicted determination degree of the node N is smaller, the process proceeds to Step 9700 to update the predicted determination degree of the node n.

  In step 9700, the predicted time determination degree calculation unit 1140 updates the prediction determination degree of the node n to the value of the prediction determination degree of the node N. In step 9800, the node n is registered in the queue in order to recalculate the definiteness of the predicted time having a causal relationship with the node n. Accordingly, since the degree of determinism can be obtained based on the causal relationship between the nodes, the degree of determinism can be set appropriately and comprehensively among a plurality of nodes. In addition, although the magnitude relationship between prediction time was utilized for calculation of a prediction definite degree, this is an example. When the predicted time determination degree calculation unit 1140 executes the predicted time determination degree calculation process of FIG. 9 and updates the prediction time determination degree, the determination degree of the prediction time based on FIG. 4 is determined as follows.

  A station train a (4111): 100%, B station train a (4111): 100%, B station train e (4122): 90%, B station train b (4112): 60%, A station train e (4122) ): 90%, B station train g (4131): 60%, A station train g (4131): 60%, A station train b (4112): 60%, A station train c (4113): 60%, A Station train h (4132): 60%, Station A train f (4123): 50%

  The predicted time determination degree calculation unit 1140 stores the calculated calculation history of the prediction determination degree in the prediction determination degree history table 1260. FIG. 10 shows a table configuration of the prediction certainty history table 1260. The prediction execution date and time 10110, which is the date and time when the prediction calculation is acquired by the prediction calculating unit 1110, the prediction target train 10120, the station 10130, the prediction time 10140, and the prediction certainty 10150 , Each record of the determination factor 10160 of the predicted time is included.

  In step 2500 of FIG. 2, the route proposal generating unit 1150 generates crew member operation proposal information (route proposal table) based on the predicted time and the predicted definiteness. FIG. 11 shows a table configuration of the route proposal table 1270, and one proposal information is indicated by a record in each column. The table shows the proposed content 11110 indicating the proposed content, the related column number 11120 indicating the train number related to the proposal, the station 11130 indicating the station related to the proposal, and the definiteness of the predicted time based on which the proposed content was generated , And a record of proposal priority 11150 indicating whether the proposal should be a priority or awaiting treatment.

  Even if the operator is prompted to propose a change in crew operation based on the predicted time associated with the schedule change, the proposal may be wasted. Therefore, the system can relate the certainty of the predicted time to the content of the proposal for changing the crew operation. The record 11210 in FIG. 11 is based on the predicted time shown in FIG. 4 `` Insufficient transit time '' when transit time is insufficient for connecting a crew member from train a (4211) to train e (4122) The relevant column numbers are train a (4211), train e (4122), the transit station is B station, the prediction accuracy of the predicted time used for determining the transit time is 90%, and priority is given to handling as the proposal priority It indicates “priority”. The prediction accuracy is 90% of the smaller accuracy of the prediction accuracy 100% of train a and the prediction accuracy 90% of train e obtained in step 2400.

  When the prediction accuracy is greater than or equal to a predetermined value, for example, 90%, the route proposal generation unit 1150 sets the proposal priority to “priority”, assuming that the schedule is operated almost at the predicted time. “Priority” indicates that it is not wasteful to arrange another crew member on the train e (4122) at station B. Therefore, the crew can be efficiently operated without waste in accordance with the schedule change.

  Record 11220, like record 11210, is a proposal for lack of transit, but based on the fact that the prediction accuracy of train g (4131) and train c (4113) is as low as 60%, the route proposal generator 1150 determines the schedule. The proposal priority is set to “waiting” so that it is necessary to deal with crew arrangements after waiting for this.

  In step 2600 of FIG. 2, the predicted time transmission unit 1160 performs transmission by changing the content of the degree of prediction determination for each target. FIG. 12 shows an example of the predicted information transmission content to the crew ward office device 1600. The predicted time transmission unit 1160 transmits the predicted time of the main location and the degree of determination of the predicted time based on the diagram (FIG. 4) showing the crew's operation plan. FIG. 12 shows the crew of a train traveling between the A station and the B station, and the definiteness of the predicted time and the predicted time is given to the arrival time and the departure time of the A station and the B station.

  FIG. 13 shows an example of the content of the prediction information transmitted to the station device 1700. This information is assumed to be information for passenger guidance at the station, and includes the train number and predicted time of the up / down train at station A. If a forecast time with a low forecast accuracy is presented to the passenger, the guided forecast time may change frequently, and only a forecast time with a forecast certainty level higher than a predetermined value is sent to avoid confusion for the passenger. The

  FIG. 14 illustrates an example in which the deterministic degree can be verified by the prediction deterministic degree verifying unit 1170. The vertical axis represents the degree of prediction determination, and the horizontal axis represents the error between the predicted time and the actual time that is the actual train travel result. Each of the plurality of curves indicates characteristics for each of a plurality of prediction determination factors (factors for which prediction time is determined: results, delay expectation, travel between stations (see FIG. 7)). The prediction determinism verification unit 1170 obtains the determinism, the prediction time, and the actual time by actual train travel results from the history data, and performs regression analysis of these values to obtain the function characteristics as shown in FIG. Create and graph. Regardless of the factor, it can be seen that the degree of determinism obtained as described above is smaller in error as the degree of determinism is higher, and the predicted time almost matches the actual time. According to the plurality of functions shown in FIG. 14, it can be seen that if the degree of definiteness is equal to or greater than the value at the inflection point, the train operation can be reliable. Thus, by evaluating and verifying the determinism from the characteristics of the variation pattern of the determinism, the operation manager can verify the validity of the determinism for each prediction determination factor. It is also possible to set the predicted time from the degree of definiteness by a function that evaluates the correlation between the degree of definiteness and the error.

  According to the operation information distribution system shown in FIG. 1, based on the distribution information reflecting the degree of definiteness, the operation information receiving side can determine the reliability of the predicted value of the operation information, and the receiving side (station, crew, If the passenger) is more than the predetermined value, even if it immediately responds to the change in the operation plan, there is no need to react again. The receiver side can appropriately respond to the prediction information, such as avoiding re-correspondence by refraining from responding until the certainty level is increased. Furthermore, the operation information system can make it easier for the receiver side to make a decision by changing the mode of the distribution information in accordance with the needs of the receiver side.

  In the above-described embodiment, the determinism is set from the table of the prediction determinism definition table 1240 and the table of the diagram change pattern table 1250. It can also be obtained by calculation with each parameter.

1100 ··· Central processing unit, 1200 · · Database, 1300 · · Input device, 1400 · · Output device, 1500 · · Network, 1600 · · Crew member device, 1700 · · Station device, 1110 · · Prediction calculation unit, 1120-Train operation net generation unit, 1130-Prediction fluctuation factor setting unit, 1140-Prediction determinism calculation unit, 1150-Route proposal generation unit, 1160-Prediction information transmission unit, 1170-Prediction determinism verification Part, 1210 ・ ・ Diagram data table, 1220 ・ ・ Route data table, 1230 ・ ・ Train service network table, 1240 ・ ・ Prediction definition definition table, 1250 ・ ・ Diagram change pattern table, 1260 ... ・ Prediction decision history table, 1270 ・ ・ Route proposal table

Claims (9)

In the operation information distribution system that distributes operation-related information related to the operation of multiple mobile units,
A recording device for recording an operation plan of the plurality of moving bodies;
A processing device for executing information processing based on the operation plan;
With
The processor is
Based on the change in the operation plan of the mobile body, set prediction information in the operation of the mobile body,
Set a certainty for the prediction information,
Based on the prediction information and the degree of determination, determine the mode of the operation related information,
Deliver the determined operation-related information to related parties,
Operation information distribution system. The processor is
Determining the factors that cause the prediction information to fluctuate;
Based on the determined factor, setting a certainty for the prediction information,
Operation information distribution system. The moving body is a train;
The prediction information includes a plurality of prediction times related to the operation of each of a plurality of trains,
The processor is
Setting the degree of definiteness for each of the plurality of predicted times;
Based on the combination of the predicted time and the degree of determination for each predicted time, determine the mode of the operation related information,
The operation information distribution system according to claim 2. The operation related information includes a plurality of operation information,
The processing device sets a priority for each of the plurality of operation information based on a degree of determination for each predicted time,
Deliver the operation related information together with the priority to related parties,
The operation information distribution system according to claim 3. The processor is
Based on the causal relationship between the plurality of prediction times, set the certainty for each of the plurality of prediction times,
The operation information distribution system according to claim 3. The processor is
Based on a combination of the predicted time and the degree of definiteness for each predicted time, set the crew's operation plan between multiple trains,
Based on the operation plan, to determine the mode of the operation related information,
The operation information distribution system according to claim 3. The processor is
Compare the certainty level with a predetermined value, and select a predicted time when a certain degree of definite degree or more is set,
Determining the mode of the operation-related information based on the selected predicted time;
The operation information distribution system according to claim 3. The processor is
In accordance with the occurrence factor of the predicted time, the degree of definiteness of the predicted time is displayed on the display device based on the predicted time and the actual time of train operation.
The operation information distribution system according to claim 3. In an operation information distribution method for distributing operation-related information related to the operation of a plurality of mobile objects,
The method is executed by information processing based on the operation plan of the processing device,
The processor is
Based on the change in the operation plan of the mobile body, set prediction information in the operation of the mobile body,
Determine the factors that change the forecast information,
Based on the determined factor, set a certainty for the prediction information,
Based on the prediction information and the degree of determination, determine the mode of the operation related information,
Deliver the determined operation-related information to related parties,
Operation information distribution system method.

Images