JP2019123479A - Delay time analysis device, delay time analysis method, and train operation support system - Google Patents

Abstract

To provide a delay time analysis device, a delay time analysis method, and a train operation support system capable of analyzing a delay caused by a plurality of factors.
A delay time analysis device 1 includes a calculation unit 22, a generation unit 26, and a display control unit 35. The calculation unit 22 calculates an explanatory variable using at least one of the arrival time and the departure time of each train at each station of a predetermined section of the railway. The generation unit 26 generates a regression equation for the objective variable using an explanatory variable having a predetermined contribution selected from among the explanatory variables, with the delay time at the predetermined station and the predetermined train in the predetermined section of the railway as the objective variable. Do. The display control unit 35 causes the display unit 30 to display information on the delay time based on the regression equation.
[Selected figure] Figure 1

Description

  Embodiments of the present invention relate to a delay time analysis device, a delay time analysis method, and a train operation support system.

  The railway is a public transportation system, and its operation is generally controlled by a diamond. In addition, it is known that Japanese railways have little difference between the time of diamond and the actual traveling time. On the other hand, delay may occur for some reason. With regard to the delay that occurs constantly, improvements can be made at the time of diagram revision. For example, in a time zone that involves rushing to commute to school or the like, the time required for passengers to get on and off is longer than the stopping time of the diamond, causing a delay. In such a case, it is conceivable to change the schedule to delay the departure time.

  However, it is not uncommon for delays to occur due to multiple factors. For example, rather than delaying the departure time as described above, shortening the stopping time of the train at the front station may possibly lead to the elimination of the delay. In other words, the stopping time at the front station also affects the delay. The delay caused by such a plurality of factors is difficult to identify. Therefore, there is a need for a delay time analyzer that can analyze multiple delay factors.

Unexamined-Japanese-Patent No. 2017-20181

  The problem to be solved by the invention is to provide a delay time analysis device, a delay time analysis method, and a train operation support system capable of analyzing a delay caused by a plurality of factors.

  According to the present embodiment, the delay time analysis device includes the calculation unit, the generation unit, and the display control unit. The calculation unit calculates a plurality of explanatory variables using at least one of the arrival time and the departure time of each train at each station on a predetermined section of the railway. The generation unit sets the delay time at a predetermined station and a predetermined train in a predetermined section of the railway as an objective variable, selects an explanatory variable having a predetermined contribution to the objective variable from among a plurality of explanatory variables, Generate a regression equation for a variable. The display control unit causes the display unit to display information related to the delay time based on the regression equation.

FIG. 2 is a block diagram showing the configuration of a delay time analysis device. The figure which shows an example of plan schedule data and performance schedule data. The figure which shows an example of departure delay time and arrival delay time. A diagram defining the definition of time intervals. The diagram which shows an example of the diamond of row number 101 and row number 103. FIG. The figure which shows typically the range which can be strongly contributed to delay time. The figure which shows the example of the explanatory variable used for analysis of the delay time of the column number 103. FIG. The figure which shows the departure delay time of line number 101 in B station. The figure which shows the departure delay time of line number 103 in D station. The figure which calculated the value of objective variable Y at the time of changing the value of explanatory variable X. The figure which shows the example of the measure shown based on the code | symbol of the explanatory variable and the calculated coefficient. The figure which shows the example which displayed the measure by the analysis result and the measure part with respect to the result. The figure which shows the analysis result and the example to which inconsistency arises in the measure with respect to the result. The flowchart which shows the processing example of a delay time analyzer. FIG. 7 is a block diagram showing the configuration of a delay time analysis device according to a second embodiment. The block diagram which shows the structure of a train driving assistance system.

Hereinafter, a delay time analysis device, a delay time analysis method, and a train operation support system according to an embodiment of the present invention will be described in detail with reference to the drawings. The embodiments described below are examples of the embodiments of the present invention, and the present invention is not construed as being limited to these embodiments. Further, in the drawings referred to in this embodiment, the same portions or portions having similar functions may be denoted by the same reference numerals or similar reference numerals, and repeated description thereof may be omitted. Further, the dimensional ratio of the drawings may be different from the actual ratio for convenience of explanation, or a part of the configuration may be omitted from the drawings.
First Embodiment

  FIG. 1 is a block diagram showing the configuration of the delay time analysis device 1. As shown in FIG. 1, the delay time analysis device 1 according to the present embodiment is a device capable of analyzing a plurality of delay factors, and includes a storage unit 10, a processing unit 20, a display unit 30, and a display control unit. It comprises 35 and the operation part 40, and is comprised.

  The storage unit 10 is realized by, for example, a random access memory (RAM), a semiconductor memory device such as a flash memory, a hard disk, an optical disk, or the like. The storage unit 10 is a storage unit that stores data necessary for processing of the diagram data and the processing unit 20, a program, and the like.

  The processing unit 20 is a processing unit that analyzes and processes the delay factor of the train using the diamond data stored in the storage unit 10. The processing unit 20 includes, for example, a processor.

  The display unit 30 displays various types of information. The display unit 30 is configured of a liquid crystal display, a CRT (Cathode Ray Tube) display, or the like. The display control unit 35 causes the display unit 30 to display the information generated by the processing unit 20 or the like.

  The operation unit 40 receives various input operations from the operator, converts the received input operations into electric signals, and outputs the electric signals to the processing unit 20. For example, the operation unit 40 is realized by a mouse, a keyboard, a trackball, a switch, a button, a joystick, or the like.

  First, the storage unit 10 will be described in more detail. The storage unit 10 includes a plan diagram storage unit 11, a performance diagram storage unit 12, and a diagram information storage unit 13. The plan diagram storage unit 11 stores the plan diagram data, and the result diagram storage unit 12 stores the result diagram data. The diagram information storage unit 13 stores data necessary for the processing of the processing unit 20. The diagram data is data composed of an arrival time to arrive at each station and a departure time to depart from each station. The diamond data according to the present embodiment means planned diamond data and actual diamond data.

  The diagram data stored in the storage unit 10 will be described more specifically based on FIG. FIG. 2 is a view showing an example of the plan diagram data and the result diagram data. As shown in the column of the plan of FIG. 2, the plan diagram data of the row number 101 is composed of the arrival time from the A station to the G station and the departure time from the station. Similarly, the planned diagram data of row number 103 is composed of the arrival time from the A station to the G station and the departure time from the station. Thus, the plan diagram data is configured by the planned values of the arrival time and the departure time of each train to each station in a predetermined section of the railway.

  In addition, as shown in the column of the results in FIG. 2, the performance diamond data of row number 101 is composed of the arrival time from the A station to the G station and the departure time from the station, The data consists of the arrival time from the A station to the G station and the departure time from the station. As described above, the actual result diamond data is constituted by actual values of arrival time and departure time of each train to each station in a predetermined section of the railway. In addition, although the diamond data which concerns on this embodiment are demonstrated as a format shown, for example in FIG. 2, it is not limited to this.

  Next, the processing unit 20 will be described more specifically. As shown in FIG. 1, the processing unit 20 includes a data input unit 21, a calculation unit 22, a selection unit 23, a setting unit 24, an analysis unit 25, and a measure generation unit 26. The data input unit 21 inputs the plan diagram data and the result diagram data from the storage unit 10. That is, the data input unit 21 includes planned diagram data which is a planned value of arrival time and departure time of each train at each station in the predetermined section of the railway, arrival time and departure of each train at each station in the predetermined section of the railway Input actual result diamond data which is actual value of time.

  The calculation process of the calculation unit 22 will be described based on FIGS. 3 to 5. The calculation unit 22 calculates a plurality of explanatory variables using at least one of the arrival time and the departure time of each train at each station of a predetermined section of the railway. More specifically, the calculation unit 22 uses the plan diagram data and the result diagram data input from the storage unit 10 to relate arrival delay times, departure delay times, stop times, and train intervals of each station and each train. At least one of the driving time intervals is calculated as an explanatory variable.

  FIG. 3 is a diagram showing an example of departure delay time and arrival delay time. FIG. 3A is a diagram showing an example of the departure delay time of the row numbers 101 and 103 and the arrival delay time. FIG. 3 (b) is a diagram showing the vehicle ID and driver ID of April 20, 2017 of row number 101, and FIG. 3 (c) is the vehicle of April 20, 2017 of row number 103. It is a figure which shows ID and driver | operator ID. Here, the row number means a number assigned to a train. Thus, the date, the vehicle ID, and the driver ID are associated with the column numbers of the delay time data.

  As shown in FIG. 3A, the difference between the planned value and the actual value of the arrival time of the train to the station is the arrival delay time.

Formula 1 is an example of the calculation formula of arrival delay time. Here, Ta (day, Train. ID, Sta. ID) indicates arrival delay time. day is a date, Train.ID is a row number, and Sta.ID is a station number. The station number is a number attached to the station. Similarly, Taactual (day, Train. ID, Sta. ID) is the actual arrival time obtained by the result diamond data, and Taaplan (day, Train. ID, Sta. ID) is obtained by the plan diamond data. It is the planned arrival time.
For example, the arrival time of the planned diagram data at station D in FIG. 2 is 8:10, and the arrival time of the actual diagram data is 8:12. That is, as shown in FIG. 3, the arrival delay time of the train of row number 101 at station D is 2 minutes obtained by subtracting 8:10 from 8:12.

Further, as shown in FIG. 3A, the difference between the planned value and the actual value of the departure time from the train station is the departure delay time. Formula 2 is an example of a calculation formula of departure delay time. Here, Td (day, Train. ID, Sta. ID) indicates the departure delay time. Tdactual (day, Train. ID, Sta. ID) is the actual departure time obtained by actual diamond data, and Tdaplan (day, Train. ID, Sta. ID) is the planned departure time obtained by planned diamond data It is.

  For example, the departure time of the plan diagram at station D in FIG. 2 is 8:11, and the departure time of the result diagram data is 8:13. That is, as shown in FIG. 3, the departure time of the train of row number 101 at station D is 2 minutes obtained by subtracting 8:11 from 8:13. As described above, it is possible to calculate the arrival delay time and the departure delay time of each station from the difference between the plan diagram data and the performance diagram data.

Formula 3 is an example of calculation formula of stop time. Here, Ts (day, Train. ID, Sta. ID) indicates a stop time. day is a date, Train.ID is a row number, and Sta.ID is a station number. The station number is a number attached to the station.

  For example, the departure time of the result diamond data at station D in FIG. 2 is 8:13, and the arrival time of the result diamond data is 8:12. That is, the stopping time of the train with the row number 101 at the D station is one minute obtained by subtracting 8:12 from 8:13. As described above, the vehicle stopping time can be calculated by subtracting the actual arrival time from the actual departure time of the actual result diamond data. Here, although the calculation method of the explanatory variable which can be calculated from the diagram data is described, information such as a boarding rate as a variable other than time, a day of the week as qualitative data, and weather may be added.

  FIG. 4 is a diagram defining the definition of the driving interval. FIG. 5 is a diagram showing an example of diamonds of row numbers 101 and 103. The horizontal axis of FIG. 5 is time, and the vertical axis is the station name. As shown in FIG. 4 and FIG. 5, the calculation unit 22 also calculates the arrival and departure intervals, the arrival and departure intervals, the arrival and departure intervals, and the arrival and departure intervals.

  The departure time is the difference between the time when the preceding train left a certain station and the time when its own train arrived, and the departure time difference is the time difference between the time when the preceding train left a certain station and the time when its own train left There is a time difference between the time when the preceding train arrived at the station and the time when the own train arrived, and the time between arrival and departure was when the preceding train arrived at the certain station and when the own train left It is a time difference of time.

  Here, assuming that the number of row numbers in a predetermined section of a railway is N, the number of stations is S, and the time interval shown in FIG. 4 is an explanatory variable, the number of explanatory variables is 4 NS in the whole diagram. These 4NS explanatory variables can all be considered as candidates for the explanatory variables in principle. However, in an actual diagram, it is only the explanatory variable at a time before the time that affects the delay time at a certain time. For example, since the data after the E station of the row number 103 is not related to the departure delay time at the D station of the row number 103, it is possible to remove it from the explanatory variable. Furthermore, when the operation of the row number 103 is in the afternoon, the influence of the explanatory variable of the row number operated in the morning is low, so it is possible to exclude it from the explanatory variables.

  Therefore, an example of an explanatory variable used in the present embodiment will be described based on FIG. FIG. 6 is a diagram schematically showing a range that can strongly contribute to the delay time of row number 103 at station D on row diagram 101 and row number 103 in the diagram. The horizontal axis of FIG. 6 is time, and the vertical axis is the station name. The diagram of FIG. 6 can also be interpreted as a graph having two variables (time, station position). As shown in FIG. 6, in the present embodiment, for example, the range that can strongly contribute to the delay time is limited to a range that is temporally close and a range that is spatially close. For example, it is an explanatory variable in the C station of the same row number 103 that is temporally close to the arrival delay time and the departure delay time at the station D of the row number 103. Also, for example, it is an explanatory variable of the row number 101 of the preceding train at the same D station that is spatially close.

  That is, in the present embodiment, the past time within a predetermined range from the time to be analyzed is regarded as a range close in time. Here, in order to simplify the description, the range which is close in time is described as one station, but it is not limited to this. Similarly, a station within a predetermined range from a station to be analyzed is assumed to be a range which is spatially close. Here, in order to simplify the description, the spatially adjacent range is described as the same station as the station to be analyzed, but is not limited thereto.

  FIG. 7 is a diagram showing an example of explanatory variables used for analysis of the delay time of the column number 103. As shown in FIG. That is, the stop time of station C, the departure delay time of station C, and the arrival of station C, which are explanatory variables related to station C with row number 103, can be a range that can contribute more strongly to the delay time of row number 103 shown in FIG. The delay time, the stop time of the D station, the departure delay time of the D station, and the arrival delay time of the D station, which are explanatory variables for the D station with the row number 101, are selected. Furthermore, four time intervals shown in FIG. 4 are added as explanatory variables as variables representing the interaction between the column numbers 103 and 101. That is, in the regression analysis in the present embodiment, a total of ten explanatory variables are adopted. LASSO regression or the like is performed on these ten types of explanatory variables to perform variable selection and coefficient calculation.

  The selection unit 23 selects a predetermined station and a predetermined train based on an average value of delay times of each station and each train in a predetermined section of the railway. For example, the selection unit 23 selects a station and a row number whose average value of delay time is 1 minute or more. Further, the selection unit 23 selects a station and a row number in which data of the delay threshold θ or more exceeds α. More specifically, based on FIG. 8 and FIG. 9, the process example which the selection part 23 selects the station and row number whose data more than delay threshold value (theta) exceeds (alpha) is demonstrated.

  FIG. 8 is a diagram showing the departure delay time of the train number 101 at the B station. Here, the distribution of the departure delay time of the train number 101 at the B station from the previous revision of the diagram to the present time is shown. The horizontal axis indicates time, and the vertical axis indicates a value obtained by normalizing the appearance frequency. FIG. 9 is a diagram showing the departure delay time of the train number 103 at the D station. Similar to FIG. 8, the horizontal axis indicates time, the horizontal axis indicates time, and the vertical axis indicates a value obtained by normalizing the appearance frequency. As shown in FIG. 8, the average departure delay time of row number 101 at station B is 0 minutes, and as shown in FIG. 9, the average departure delay time of row number 103 at station D is 1 minute. is there.

  For example, as shown in FIGS. 8 and 9, the selection unit 23 generates a histogram of the departure delay time for each station and row number, and calculates the average value and the dispersion value of the generated histogram. The selection unit 23 also generates a histogram of arrival delay time for each station and row number, and calculates the average and the variance of the generated histogram. Then, in the generated histogram, when the data exceeds α% by more than the delay threshold θ serving as the selection reference, the selection unit 23 selects the station and the column number as the objective variable.

  The setting unit 24 sets the delay threshold values θ and α through the operation of the operation unit 40. For example, assuming that θ = 1, α = 0.1 (= 10%), if there is 10% or more of a ratio in which the delay time exceeds 1 minute as a threshold, as a candidate for the station and column number for analysis select. In FIG. 8, since the ratio of θ = 1 or more is about 1%, the delay time of row number 101 at station B is not selected as analysis symmetry. On the other hand, since the ratio of θ = 1 or more is about 40% in FIG. 9, the delay time of row number 103 at station D is selected as analysis symmetry. Thus, the data of the station ID and the row number that match the condition are selected as the analysis symmetry from among the combination of the number of stations × the row number.

  In addition, when there is a row number that passes between multiple stations such as express trains, there is a row number that departs from the station halfway, and if there is a row number that loops back at the station, the combination number is more than the number of stations × row number. Decrease. Moreover, you may change each value of (theta) and (alpha) for every row number and a station. As can be understood from these, it is possible to select the combination of the station and the row number to be analyzed by the delay threshold values θ and α set by the setting unit 24.

  The analysis unit 25 generates a regression equation for the objective variable using an explanatory variable having a predetermined contribution selected from among the explanatory variables, with the delay time at the predetermined station and the predetermined train in the predetermined section of the railway as the target variable. Do. For example, the analysis unit 25 uses the delay time of the station and column number selected by θ and α set by the setting unit 24 as target variables, and uses the variable calculated by the calculation unit 22 for each target variable as an explanatory variable. Perform multiple regression analysis.

  It is possible to use general LASSO regression (Lasso Regressor), multistage multiple regression analysis, etc. for this multiple regression analysis. In LASSO regression, for example, an explanatory variable contributing to calculation of a target variable is automatically selected from among 10 types of explanatory variables, and coefficients of the selected explanatory variable are calculated. For example, when multiple regression analysis is performed with the departure delay time of row number 103 at station D as the objective variable and the above 10 types of variables as explanatory variables, it contributes to the calculation of the objective variable among the 10 types of variables An explanatory variable is selected, and the coefficient of the explanatory variable is calculated. In this case, the coefficient of the explanatory variable that has a small contribution to the calculation of the objective variable is output as 0. Also, for example, in the multistage multiple regression analysis, the explanatory variable is selected based on the contribution ratio of the explanatory variable to the objective variable. Thus, the analysis unit 25 selects an explanatory variable related to the predetermined station which is the objective variable and the delay time of the predetermined train.

  FIG. 10 is a diagram in which the value of the objective variable Y when the value of the explanatory variable X is changed is calculated in the explanatory variable X and the coefficient thereof selected by the analysis unit 25. As shown in FIG. 10, the selected explanatory variable X has a high correlation with the target variable, for example, 0.7 or more.

  In addition, the analysis unit 25 may analyze the data for each day of the week using the data for each day of the performance diamond. More specifically, the calculation unit 22 uses the plan diagram data and the performance diagram data for each day of the week input from the storage unit 10, and the arrival delay time, departure delay time, and stop time of each station and each train for each day of the week , And calculate the driving interval. Thus, the analysis unit 25 can analyze the data for each day of the week using the data for each day of the week of the performance diamond.

  Furthermore, the analysis unit 25 may perform data analysis excluding at least one of the data on the date when the accident occurred and the date on which the temporary flight occurred. This makes it possible to perform analysis processing excluding statistically unique data. The analysis unit 25 according to the present embodiment corresponds to a generation unit.

  Based on the explanatory variable selected by the analysis unit 25, the policy generation unit 26 causes the display unit 20 to display a measure for the delay time of the station and the train (row number) selected by the selection unit 23.

  FIG. 11 is a diagram showing an example of a measure presented based on the explanatory variable and the sign of the calculated coefficient. FIG. 12 is a diagram showing an example of displaying the result of the multiple regression analysis by the analysis unit 25 and the measure by the measure generation unit 26 with respect to the result. Here, the term "measures" means contents instructing a driver, an operator, etc. what kind of traveling should be performed. In addition, the measure here is not traveling of the preceding train but shows the existing train at the corresponding station. Further, the measure plan is set in advance, and when it is desired to add a measure, the setting unit 24 can access and add the measure generation unit 26.

  For example, the stopping time of C station and the arrival / arrival time at D station with row number 101 are selected as explanatory variables of regression equation with column number 103 at D station as the objective variable, and the sign of the coefficient of stopping time of C station Is positive, and it is assumed that the sign of the arrival and arrival interval coefficient at station D is negative. In this case, the measure generation unit 26 presents on the display unit 30 that “shortening stop time” at the C station is a measure, and presents “delay the arrival at the D station” as a measure. That is, in this case, it means that the train speed between Station C and Station D is reduced. By performing analysis by statistical processing in this manner, it is also possible to obtain analysis results of a plurality of delay factors different from the feeling of a normal driver and measures that are the measures based on statistics. For example, in this case, there is a tendency that the arrival at the D station of the row number 103 tends to be too fast, and there are many cases where the vehicle can be stopped by the stop signal before the D station. For this reason, it is possible to suppress the delay as a result if the operation is performed at a reduced speed and the signal is allowed to pass without being stopped by the signal rather than being stopped by the stop signal. Thus, the user of the delay time analysis device 1 can obtain measures for traveling to each station of each row number, for example, by viewing the screen shown in FIG.

  FIG. 13 is a diagram showing an example of the result of multiple regression analysis by the analysis unit 25 and an example in which the measure for the result is inconsistent. As shown in FIG. 13, it is possible to indicate that the measure can not be determined from the analysis result of the multiple regression by displaying, for example, “Measure is indefinite” on the screen. For example, when a statistically significant explanatory variable can not be selected, "The measure is indefinite" is displayed.

  As described above, the processing unit 20 according to the present embodiment is configured of, for example, a processor. Here, the term processor means, for example, a central processing unit (CPU), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), a programmable logic device (for example, a simple programmable logic device). It means circuits such as (Simple Programmable Logic Device: SPLD), Complex Programmable Logic Device (CPLD), and Field Programmable Gate Array (FPGA) etc. The processor is in the storage unit 10. The function is realized by reading out and executing the stored program, and instead of storing the program in the storage unit 10, the program may be directly incorporated in the circuit of the processor. Processor 20 are combined to form the processing unit 20, and each processor There may be one that realizes the functions by executing a program.

  Next, a process example of the delay time analysis device 1 will be described. FIG. 14 is a flowchart showing an example of processing of the delay time analysis device 1. Here, a case where a predetermined section of the railway is determined in advance will be described.

  First, the data input unit 21 inputs planned schedule data and actual result schedule data from a predetermined section of the railway, here, from station A to station G (step S101). Also, with the plan diagram data and the actual diagram data, data from the day of the previous diagram revision to the day of the previous day is input. Subsequently, the calculation unit 22 uses arrival schedule data and performance history data as explanatory variables, such as arrival delay time, departure delay time, stopping time, arrival / departure time, departure time interval, arrival time interval, arrival time Calculate for all stations and all row numbers for time intervals

  Next, the selection unit 23 calculates the average value of the arrival delay times calculated by the calculation unit 22, and selects, for example, a station and a row number that is 1 minute or more (step S102). In addition, you may select regarding departure arrival delay time.

  Next, the analysis unit 25 performs LASSO regression using the explanatory variables calculated by the calculation unit 22 with the delay time of the station and the row number selected by the selection unit 23 as the target variable (step S103). Thereby, the explanatory variable related to the station and row number which were selected by the selection part 23 is selected, and the coefficient is calculated. The analysis unit 25 stores a series of regression equations in the ear information storage unit 13.

  Next, the policy generation unit 26 causes the display unit 30 to display the regression equation for each station and row number selected by the selection unit 23 and the policy. Thereby, the operator can grasp the delay time from the A station to the G station, the cause thereof, and the measure.

  As described above, according to the present embodiment, the analysis unit 25 uses the delay time in the predetermined station and the predetermined train in the predetermined section of the railway as the objective variable, and the explanatory variable can be used for the objective variable. Choose an explanatory variable with a predetermined contribution. As a result, an explanatory variable related to the delay time can be grasped based on statistical processing. Further, since the station name and the column number for which the explanatory variable is calculated can be grasped from the selected explanatory variable, the operator can objectively grasp the cause of the delay time.

Second Embodiment
The delay time analysis device according to the second embodiment is different from the first embodiment in that the delay time analysis device further includes a prediction unit that predicts the delay time of the train using the regression equation generated by the analysis unit. Hereinafter, differences from the first embodiment will be described.

  FIG. 15 is a block diagram showing the configuration of the delay time analysis device according to the second embodiment. FIG. 16 is a block diagram showing the configuration of a train operation support system.

  As shown in FIG. 15, the prediction unit 27 predicts delay times of a predetermined station and a predetermined train by substituting actual values for the explanatory variables of the regression equation generated by the analysis unit 25. The prediction unit 27 predicts the delay time of the train number 103 of the D station, for example, using a regression equation indicating the delay time of the train number 103 of the D station. In this case, the delay time of the train number 103 of the D station can be predicted by substituting the actual measurement value into the explanatory variable of the regression equation.

  The train operation support system 100 is a system for supporting the operation of the train 300, and includes a delay time analysis device 1 and a train operation support device 200. The train operation support device 200 is a device that performs operation support of a train using information generated by the delay time analysis device 1 and includes a transmission unit 201.

  The transmission unit 201 of the train driving support device 200 transmits the prediction result of the prediction unit 27 to the train of the row number. In this case, the result of the measure of the measure generation unit 26 is also transmitted to the train. In the train to which the information is transmitted by the transmission unit 201, the information is presented to the driver or the like by a display device such as a tablet.

  As described above, according to the present embodiment, it is possible to predict the delay time of the predetermined station and the predetermined train by substituting the actual value into the regression equation generated by the analysis unit 25. In addition, the train driving support device 200 transmits information related to the prediction result to the driver, and instructs traveling to the next stop station. As a result, the result analyzed by the delay time analysis device 1 can be immediately used for actual train operation.

  While certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the invention. These novel embodiments can be implemented in other various forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and the gist of the invention, and are included in the invention described in the claims and the equivalent scope thereof.

  1: delay time analyzer, 21: data input unit, 22: calculation unit, 23: selection unit, 25: analysis unit, 26: measure generation unit, 27: prediction unit, 30: display unit, 35: display control unit, 100: Train driving support system, 200: Train driving support device, 201: Transmission unit, 300: Train

Claims (11)

A calculation unit that calculates a plurality of explanatory variables using at least one of arrival time and departure time of each train at each station of a predetermined section of the railway;
The delay time at a predetermined station and a predetermined train in the predetermined section of the railway is set as a target variable, and an explanatory variable having a predetermined contribution to the target variable is selected from the plurality of explanatory variables, A generator that generates a regression equation for the target variable;
A display control unit that causes the display unit to display information related to the delay time based on the regression equation;
Delay time analyzer comprising: The system further includes a policy generation unit that generates a policy for the delay time at the predetermined station and the predetermined train based on the selected explanatory variable.
The delay time analysis device according to claim 1, wherein the display control unit causes the display unit to display the measure. Plan diamond data, which is a planned value of arrival time and departure time of each train at each station in the predetermined section of the railway, and actual values of arrival time and departure time of each train at each station in the predetermined section of the railway It further comprises a data input unit for inputting certain performance diamond data,
The calculation unit uses at least one of a departure delay time, an arrival delay time, a stopping time, and an operation time interval relating to a train interval of each of the stations and the trains using the plan diagram data and the result diagram data. The delay time analysis device according to claim 1 or 2, which calculates The operation time intervals are any of arrival time and arrival time, arrival time, arrival time, and arrival time,
The delay time analyzer according to claim 3, wherein the delay time in the target variable is an arrival delay time or a departure delay time.   The generation unit according to any one of claims 1 to 4, wherein the generation unit generates the regression equation by excluding at least one of data of an accident occurrence date and a temporary flight occurrence date. Latency analyzer.   The delay time analysis device according to claim 3, wherein the generation unit generates the regression equation for each day of the week using data for each day of the week of the performance diamond data.   The selection unit according to any one of claims 1 to 6, further comprising: a selection unit that selects the predetermined station and the predetermined train based on an average value of delay time of each station and each train in the predetermined section of the railway. Latency analyzer.   The delay time analysis device according to any one of claims 1 to 7, wherein the generation unit selects the explanatory variable based on a contribution ratio of the explanatory variable to the objective variable.   The prediction unit according to any one of claims 1 to 8, further comprising: a prediction unit that predicts the delay time of the predetermined train at the predetermined station by substituting a track record value into each of the explanatory variables of the regression equation. Delay time analyzer. A calculating step of calculating a plurality of explanatory variables using at least one of arrival time and departure time of each train at each station of a predetermined section of the railway;
The delay time at a predetermined station and a predetermined train in the predetermined section of the railway is set as a target variable, and an explanatory variable having a predetermined contribution to the target variable is selected from the plurality of explanatory variables, A generation step of generating a regression equation for a target variable;
A display control step of causing the display unit to display information on the delay time based on the regression equation;
A delay time analysis method comprising: The delay time analyzer according to any one of claims 1 to 9,
A transmission unit configured to transmit train information on predicted values of delay times in which actual value is substituted for each explanatory variable of the regression equation. Train operation support system.