Patent Application
20190318253
The present application claims priority from Japanese application JP 2018-078764, filed on Apr. 16, 2018, the contents of which is hereby incorporated by reference into this application.
The present invention: relates to a schedule analysis support device and method; and is preferably applicable to a schedule analysis support device to support analysis work of a train schedule, for example.
A schedule of a train has heretofore been produced manually in consideration of the convenience of a passenger and the profitability of a railway company while multifaceted decision factors such as urban development or decline, an economic situation, and the like are used as input sources. In a railway transport system for travelling urban suburbs in particular, a schedule is produced from the viewpoint of how to transport passengers without delay during rush hours. This is because, once a delay occurs, a delay generated secondarily (hereunder referred to as a secondary delay) by an original delay, including the increase of time required for getting on and off, the increase of an accident risk caused by congestion at a station platform, and the delay of a following train accompanying the delay of a preceding train, expands.
Further, a railway company: spends a long time to collect actual traveling schedules of trains (hereunder referred to as actual schedules), on-site voices, and opinions of passengers; and analyzes and specifies a place in a schedule where a delay of a train is likely to occur in order to improve the delay by the revision of a schedule or the change of operation. The “change of operation” here means special operation, such as the omission of transfer announcement, not shown in a schedule. There are some actual cases of being able to reduce the disturbance of a schedule by shortening and omitting announcement. Even now, an experienced person in charge addresses the analysis work of such a schedule while relying on a wide variety of data, lots of effort and time, and a sense.
As a technology related to such a schedule analysis, Japanese Patent Application No. 2008-114779 discloses an invention of being able to visually acquire the trends of departure and arrival delays by drawing delay times obtained from a reference schedule and a comparative schedule on a graph of frequency distribution/cumulative relative frequency distribution.
The invention disclosed in Japanese Patent Application No. 2008-114779 however has the problem of not being able to analyze delays for each factor although it has means for analyzing delays for each type. Further, the invention disclosed in Japanese Patent Application No. 2008-114779 cannot eliminate the influence of a secondary delay and also has the problem of not being able to correctly analyze the delay of a train alone.
The present invention: has been established in consideration of the above situation; and proposes a schedule analysis support device and a method that can more reduce work time and facilitate work in the analysis of a schedule.
In the present invention, in order to solve the above problem, a schedule analysis support device to support analysis work of schedules of mobile bodies in a transport system is configured to include: a model generation unit to generate a statistical model of a delay time of each of the mobile bodies to a plan schedule that is a transport schedule preplanned for the mobile body in each predetermined section on the basis of the plan schedule, an actual schedule that is a daily travel result of the mobile body, and actual values of respective delay factors in the section in chronological order; an analysis unit to execute at least either of past result analysis processing of classifying a first delay time that is a delay time of the mobile body in the section into second delay times that are delay times of the delay factors in the section of the mobile body on the basis of the generated statistical model and the actual values of the delay factors and future prediction analysis processing of predicting a third delay time that is a delay time of a requested mobile body in the section on the basis of the statistical model and assumed values of the delay factors with respect to a new plan schedule; and an output unit to display a processing result of the past result analysis processing or the future prediction analysis processing in a predetermined representation form.
Further, in the present invention, a schedule analysis support method executed by a schedule analysis support device to support analysis work of schedules of mobile bodies in a transport system is configured to include: a first step of generating a statistical model of a delay time of each of the mobile bodies to a plan schedule that is a transport schedule preplanned for the mobile body in each predetermined section on the basis of the plan schedule, an actual schedule that is a daily travel result of the mobile body, and actual values of respective delay factors in the section in chronological order; a second step of executing at least either of past result analysis processing of classifying a first delay time that is a delay time of the mobile body in the section into second delay times that are delay times of the delay factors in the section of the mobile body on the basis of the generated statistical model and the actual values of the delay factors and future prediction analysis processing of predicting a third delay time that is a delay time of a requested mobile body in the section on the basis of the statistical model and assumed values of the delay factors with respect to a new plan schedule; and a third step of displaying a processing result of the past result analysis processing or the future prediction analysis processing in a predetermined representation form.
By a schedule analysis support device and a method according to the present invention, a user can easily know the influence of each delay factor on a delay time of each train in each section between stations.
The present invention makes it easier to reduce work time and facilitate work in the analysis of a schedule.
An embodiment according to the present invention is explained hereunder in detail in reference to the drawings.
In
The CPU 3 is a processor to control the operations of the entire schedule analysis support device 1. Further, the memory 4: includes a volatile semiconductor memory such as a DRAM (Dynamic RAM (Random Access Memory)) or an SRAM (Static RAM) for example; and is used as a work memory of the CPU 3.
The auxiliary storage device 5: includes a nonvolatile large-capacity storage device such as a hard disk device or an SSD (Solid-State Drive) for example; and is used for retaining a program and necessary data for a long period of time. A program stored in the auxiliary storage device 5 is read out to the memory 4 at the start of the schedule analysis support device 1 or when necessary, the CPU 3 executes the program, and thus various kinds of processing that will be described later is executed.
The communication device 6: includes an NIC (Network Interface Card) for example; and executes protocol control during communication with an external device through a communication network 9 such as the Internet. Further, the input device 7: includes a keyboard or a mouse for example; and is used when a user applies various operations to the schedule analysis support device 1. Furthermore, the display device 8: includes a liquid crystal display or an organic EL (Electro Luminescence) display for example; and is used for displaying a necessary screen or information.
Successively, a schedule analysis support function installed in the schedule analysis support device 1 is explained hereunder. A schedule analysis support function to support the analysis work of an operation schedule of a train is installed in the schedule analysis support device 1.
Actually, the schedule analysis support device 1, as schedule analysis processing based on such a schedule analysis support function, generates a delay model of a delay time of each train to a transport schedule preplanned for the train (hereunder referred to as a plan schedule) in each section between stations (the section is hereunder referred to as a station section) on the basis of the plan schedule, an actual schedule that is a daily travel result of the train, and actual values of factors that can cause a train to be delayed (hereunder referred to as delay factors) in the station section in a chronological order.
As “delay factors” here, precipitation (rain), wind, snowfall, a train occupancy by passengers, the number of passengers on platforms (the number of passengers staying on platforms) at stations on the departure side and the arrival side of a corresponding station section, presence or absence of a large-scale event around a station, and the like are named. Further, a “delay model” means a statistical model of a delay time of a train in a station section produced on the basis of past actual values of delay factors in the station section of the train and an actual delay time in the station section. The delay model is defined as a function that has: delay factors as variables; and the degree (weight) of influence of each delay factor on the delay of a train as a coefficient of the corresponding variable.
Then the schedule analysis support device 1, when a designated analysis mode is a past analysis mode of analyzing past results of train delay, executes past result analysis processing of classifying a delay time of each train in each station section into delay times of respective delay factors on the basis of a generated delay model in the station section of the train and actual values of the delay factors in the station section during the travel of the train. Further, the schedule analysis support device 1, when a designated analysis mode is a future prediction mode of predicting a delay time of each train in each station section in a new plan schedule, executes future prediction analysis processing of predicting the delay times of respective delay factors in each station section of each train on the basis of a delay model and assumed values of the delay factors with regard to the new plan schedule.
Then the schedule analysis support device 1, during a past analysis mode, calculates a delay time of each train in each station section by: multiplying delay times of respective delay factors in the station section of the train by magnifications (0 to 1 in the present embodiment) of the delay factors designated by a user in a visualization parameter designation screen 50 that will be described later in reference to
By setting the magnification of a desired delay factor at “1” and setting the magnifications of all the other delay factors at “0” therefore, an estimated value of a delay time caused only by the delay factor the magnification of which is set at “1” included in a delay time of each train in each station section can be obtained in the station section of the train and, in the same way as this, for each of the delay factors, an estimated value of a delay time caused only by a designated delay factor can be obtained in each station section of each train. Further, by adjusting the values of the magnifications of delay factors, an estimated value of a delay time caused by the combination of arbitrary two or more delay factors can also be obtained.
Further, the schedule analysis support device 1 displays a schedule analysis result display screen 60 that will be described later in reference to
Actually, in the schedule analysis result display screen 60, in each station section of each train, an estimated value of a delay time caused by one or more delay factors calculated as stated above: is represented as a line segment 62A having a color tone corresponding to the magnitude of the estimated value as it will be described later in reference to
A user therefore can visually more intuitively acquire the magnitude of a delay time caused only by one delay factor or the magnitude of a delay time caused by multiple delay factors in each station section of each train on the basis of a color tone of a line segment 62A displayed in the schedule analysis result display screen 60 by setting the magnifications of delay factors designated in the visualization parameter designation screen 50 at desired values respectively.
Further, the schedule analysis support device 1, during the future prediction mode, in each station section of each train in a new plan schedule: calculates delay times of respective delay factors in the station section of the train; multiplies the delay times by magnifications of the delay factors designated by a user in the visualization parameter designation screen 50 (
By setting the magnification of a desired delay factor at “1” and setting the magnifications of all the other delay factors at “0” therefore, a predicted value of a delay time caused only by the delay factor the magnification of which is set at “1” included in a delay time of each train in each station section can be obtained in the station section of the train and, in the same way as this, a predicted value of a delay time caused only by a designated delay factor can be obtained in each station section of each train. Further, by adjusting the values of the magnifications of delay factors, a predicted value of a delay time caused by the combination of arbitrary two or more delay factors can also be obtained.
Then the schedule analysis support device 1 displays a predicted value of a delay time of each train in each station section caused by one or more delay factors calculated in this way in the schedule analysis result display screen 60 similarly to the past result analysis processing.
A user therefore can visually more intuitively acquire the magnitude of a delay time caused only by one delay factor or the magnitude of a delay time caused by multiple delay factors in each station section of each station on the basis of a color tone of a line segment 62A displayed in the schedule analysis result display screen 60 with regard to a new plan schedule by setting the magnifications of delay factors designated in the visualization parameter designation screen 50 at desired values respectively.
As means for materializing such a schedule analysis support function according to the present embodiment, a plan schedule management table 10, an actual schedule management table 11, a train delay time management table 12, a schedule restriction management table 13, a delay factor management table 14, a post-refinement train delay time management table 15, a delay-factor-specific result management table 16, an analysis method management table 17, and a delay model management table 18 are stored in an auxiliary storage device 5 of the schedule analysis support device 1.
The plan schedule management table 10 is a table used for managing an existing plan schedule of a railway to be an analysis target (the railway is hereunder referred to as a target railway) and a new plan schedule made for a new railway. The plan schedule management table 10 includes a train number column 10A, a train type column 10B, an up-or-down classification column 10C, and an arrival and departure time column 10D as shown in
Then in the train number column 10A, an identification number (train number) intrinsic to a train given to each train traveling along a target railway is stored. In the case of the present embodiment here, when an identical train shuttles once or more than once a day, the train is managed as different trains for each one-way and different numbers are given to the different trains.
Further, a train type, such as “local”, “rapid”, or “express”, of the train of a train number is stored in the train type column 10B and the up or down, such as “inbound” or “outbound”, of a train is stored in the up-or-down classification column 10C.
The arrival and departure time column 10D is classified into a plurality of station arrival and departure time columns 10DA so as to correspond to respective stations on a target railway. Further, each of the station arrival PREDICTION and departure time columns 10DA is classified into an arrival column 10DAA and a departure column 10DAB, and an estimated time when a train arrives at a station is stored in an arrival column 10DAA and an estimated time when a train departs from a station is stored in a departure column 10DAB.
The case of the example in
Further, the actual schedule management table 11: is a table used for managing an actual schedule of each train that has travelled along a target railway; and is made every day. The actual schedule management table 11 includes a train number column 11A, a train type column 11B, an up-or-down classification column 11C, and an arrival and departure time column 11D as shown in
Then in the train number column 11A, the train type column 11B, and the up-or-down classification column 11C, information similar to the corresponding columns (the train number column 10A, the train type column 10B, and the up-or-down classification column 10C) in the plan schedule management table 10 is stored, respectively.
Further, the arrival and departure time column 11D is, similarly to the arrival and departure time column 10D in the plan schedule management table 10, classified into a plurality of station arrival and departure time columns 11DA so as to correspond to respective stations on a target railway. Then, each of the station arrival and departure time columns 11DA is classified into an arrival column 11DAA and a departure column 11DAB, and an actual time when a train arrives at a station is stored in an arrival column 11DAA and an actual time when a train departs from a station is stored in a departure column 11DAB.
The case of the example in
The train delay time management table 12: is a table used for managing a delay time of each train in each station section; and is made every day. The train delay time management table 12 includes a train number column 12A, a train type column 12B, an up-or-down classification column 12C, and a section-specific delay time column 12D as shown in
Then in the train number column 12A, the train type column 12B, and the up-or-down classification column 12C, information similar to the corresponding columns (the train number column 10A, the train type column 10B, and the up-or-down classification column 10C) in the plan schedule management table 10 is stored, respectively.
Further, the section-specific delay time column 12D is classified into a plurality of station section columns 12DA so as to correspond to respective stations on a target railway and each of the station section columns 12DA is classified into an arrival delay time column 12DAA and a train stop delay time column 12DAB. Then, a delay time from a plan schedule of a time when a train arrives at a station on the arrival side of a station section is stored in the arrival delay time column 12DAA and a difference between a stop time of a train at a station on the arrival side of a station section and a stop time of the train at the station in the plan schedule is stored as a train stop delay time in the train stop delay time column 12DAB.
The case of the example in
The schedule restriction management table 13: is a table used for managing the restriction of a train during travelling specified for each train (the restriction is hereunder referred to as schedule restriction); and includes a train number column 13A and a restriction content column 13B as shown in
Then the train number of a train to which a schedule restriction is applied is stored in the train number column 13A and a specific content of the schedule restriction is stored in the restriction content column 13B. The case of the example in
The delay factor management table 14: is a table used for managing a delay factor of preselected train delay; and includes a factor number column 14A and a factor name column 14B as shown in
The case of the example in
The post-refinement train delay time management table 15: is a table used for managing a delay time of each train in each station section after the influence of secondary delay (the influence of delay of a preceding train, the influence of delay of the relevant train in a previous station section, and the like) is excluded while schedule restrictions are taken into consideration and; is made every day. The post-refinement train delay time management table 15 includes a train number column 15A, a train type column 15B, an up-or-down classification column 15C, and a section-specific delay time column 15D as shown in
Then in the train number column 15A, the train type column 15B, and the up-or-down classification column 15C, information similar to the corresponding columns (the train number column 12A, the train type column 12B, and the up-or-down classification column 12C) in the train delay time management table 12 (
The section-specific delay time column 15D is classified into a plurality of station section columns 15DA so as to correspond to respective station sections on a target railway. Further, each of the station section columns 15DA is classified into an arrival delay time column 15DAA and a train stop delay time column 15DAB.
Then in the arrival delay time column 15DAA, a delay time from a plan schedule of a predicted arrival time when a train arrives at a station on the arrival side of a station section when the influence of secondary delay is excluded while schedule restrictions are taken into consideration is stored. Further, in the train stop delay time column 15DAB, a difference (train stop delay time) between a stop time of a train at a station on the arrival side of a station section and a stop time of the train at the station in the plan schedule when the influence of secondary delay is excluded while schedule restrictions are taken into consideration is stored.
In the case of the example in
The delay-factor-specific result management table 16: is a table used for managing an actual value of each of delay factors in each station section during the operation of each train; and is made every day. The delay-factor-specific result management table 16 includes a train number column 16A, a factor number column 16B, and a delay factor result column 16C as shown in
Then a train number of a train is stored in the train number column 16A and a factor number of a delay factor is stored in the factor number column 16B. Further, the delay factor result column 16C is classified into a plurality of station section columns 16CA so as to correspond to respective station sections on a target railway and, in each of the station section columns 16CA, an actual value of a delay factor when a train travels in a station section is stored.
In the case of the example in
The analysis method management table 17: is a table used for managing an analysis method in each of predetermined analysis modes; and includes a method number column 17A, an analysis mode column 17B, and a method name column 17C as shown in
Meanwhile, in the case of the present embodiment, as analysis modes, there are two analysis modes of a past analysis mode of analyzing a past result of train delay and a future prediction mode of predicting a delay time of each train in each station section in a new plan schedule, as stated above.
Then an identification number (method number) given to each combination of an analysis mode and an analysis method is stored in the method number column 17A and information representing whether the analysis mode in a combination is a past analysis mode or a future prediction mode (“past analysis” in the case of a past analysis mode and “future prediction” in the case of a future prediction mode) is stored in the analysis mode column 17B.
Further, the name of an analysis method used in a combination is stored in the method name column 17C. In the present embodiment, as such an analysis method, a multiple regression analysis method, a quantification I class method, a principal component analysis method, a factor analysis method, or the like is used and hence a name (“multiple regression analysis”, “quantification I class”, “principal component analysis”, “factor analysis”, or the like) of such an analysis method is stored in the method name column 17C.
Meanwhile, in the case of the present embodiment, the “factor analysis” method is set as the default of an analysis method used in the past analysis mode and the “multiple regression analysis” method is set as the default of an analysis method used in the future prediction mode.
The delay model management table 18 is a table used for managing a delay model generated in each station section of each train. The delay model management table 18 includes an analysis number column 18A, a method number column 18B, a train number column 18C, a station section column 18D, and a delay factor coefficient column 18E as shown in
Then an identification number (analysis number) intrinsic to a delay model given to the delay model is stored in the analysis number column 18A and a method number of an analysis method used for generating the delay model is stored in the method number column 18B. Further, a train number of a train with which the delay model is associated is stored in the train number column 18C and identification information of a station section with which the delay model is associated is stored in the station section column 18D.
Furthermore, the delay factor coefficient column 18E is classified into a plurality of coefficient columns 18EA, which are associated with respective delay factors registered in the delay factor management table 14, and a constant column 18EB. Then a value of a coefficient of a delay factor corresponding to a factor number in a delay model (function) calculated by using a corresponding analysis method is stored in each of the coefficient columns 18EA and a value of a constant in the delay model (function) is stored in the constant column 18EB.
In the case of the example in
Meanwhile, as means for materializing such schedule analysis support function as stated above according to the present embodiment, in the memory 4 of the schedule analysis support device 1, an analysis program 20 including a mode setting unit 21, a plan schedule read unit 22, an actual data read unit 23, a delay time calculation unit 24, a delay time refinement unit 25, a delay model generation unit 26, a past result analysis unit 27, a future prediction analysis unit 28, and an analysis result output unit 29 is stored as shown in
The mode setting unit 21 is a module having the function of: displaying an analysis mode designation screen 30 that will be described later in reference to
Further, the plan schedule read unit 22 is a module having the function of acquiring a plan schedule of an analysis target from the plan schedule management table 10 (
The delay time calculation unit 24 is a module having the function of generating the train delay time management table 12 stated above in reference to
The delay time refinement unit 25 is a module having the function of calculating a delay time of each train in each station section during an analysis target period excluding the influence of secondary delay while the schedule restrictions of the train are taken into consideration. Actually, the delay time refinement unit 25 calculates a delay time of each train in each station section during an analysis target period excluding the influence of secondary delay while the schedule restrictions of the train registered in the schedule restriction management table 13 (
The delay model generation unit 26 is a module having the function of generating a delay model on a delay time of each train in each station section in a target plan schedule by using an analysis method designated by a user.
For example, the delay model generation unit 26, when a multiple regression analysis method is designated by a user as an analysis method, by defining the values of delay factors as x1, x2, x3, . . . , the coefficients of those delay factors as a1, a2, a3, . . . , and the constant as b, a delay model on a delay time f of each train in each station section is generated as a function shown by the expression (1).
f=a
1
x
1
+a
2
x
2
+a
3
x
3
+ . . . +b (1)
Here, in the expression (1), the terms “a1x1”, “a2x2”, “a3x3”, . . . represent delay times included in the delay time f of a certain train in a certain station section and caused only by corresponding delay factors respectively. Further, the “b” represents a delay time in the case of being influenced by none of the delay factors, namely a delay time generated regularly by being caused by a problem of a plan schedule itself or a facility itself. It can therefore be said that, in the delay model, a past delay time f is calculated by being classified into the delay times of respective delay factors (in the case of a past analysis mode) or a delay time f is calculated by adding up predicted values of the delay times of respective delay factors (in the case of a future prediction mode). Then the delay model generation unit 26 registers coefficients of respective delay factors in a generated delay model to the delay model management table 18 (
The past result analysis unit 27 is a module, during a past analysis mode, having the function of executing processing (past result analysis processing) of, in each station section of each train, calculating an estimated value of a delay time of the train in the station section by classifying a delay time of the train in the station section into delay times of respective delay factors and multiplying the classified delay times of the delay factors by magnifications given to the corresponding delay factors in the station section of the train on the basis of a delay model, an actual schedule of the train, and actual values of the delay factors of the train on a target railway.
Further, the future prediction analysis unit 28 is a module, during a future prediction mode, having the function of executing processing (future prediction analysis processing) of, in each station section of each train in a new plan schedule, calculating a predicted value of a delay time of the train in the station section by calculating delay times of respective delay factors in the station section of the train and multiplying the calculated delay times of the delay factors by magnifications given to the corresponding delay factors.
Furthermore, the analysis result output unit 29 is a module having the function of: generating a schedule analysis result display screen 60 representing the processing result of past result analysis processing executed by the past result analysis unit 27 and the processing result of future prediction analysis processing executed by the future prediction analysis unit 28 in a representation form that will be described later in reference to
The specific processing contents of the mode setting unit 21, the plan schedule read unit 22, the actual data read unit 23, the delay time calculation unit 24, the delay time refinement unit 25, the delay model generation unit 26, the past result analysis unit 27, the future prediction analysis unit 28, and the analysis result output unit 29 will be described later.
In the analysis mode/analysis method designation region 31, character strings 40A and 40B of “past analysis” and “future prediction” and check boxes 41A and 41B formed so as to correspond to the character strings 40A and 40B respectively are displayed. Then in the analysis mode designation screen 30, a past analysis mode can be designated as the analysis mode of schedule analysis processing executed by clicking the check box 41A corresponding to the character string 40A of “past analysis” and displaying a check mark (not shown in the figure) in the check box 41A. Further, in the analysis mode designation screen 30, a future prediction mode can be designated as the analysis mode of schedule analysis executed by clicking the check box 41B corresponding to the character string 40B of “future prediction” and displaying a check mark (not shown in the figure) in the check box 41B.
Further, text boxes 42A and 42B and pull-down buttons 43A and 43B are displayed respectively on the right sides of the character strings 40A and 40B of “past analysis” and “future prediction” in the analysis mode/analysis method designation region 31. Then by clicking the pull-down button 43A or 43B, a pull-down menu (not shown in the figure) in which the method names of all the analysis methods usable in a corresponding analysis mode are described can be displayed. In this way, a user selects the name of a desired analysis method from the pull-down menu and hence can select the analysis method as an analysis method used during the analysis mode. The name of an analysis method selected at that time is displayed in a text box 42A or 42B displayed on the left side of the pull-down button 43A or 43B.
In the analysis target period designation region 32, text boxes 44A and 44B and pull-down buttons 45A and 45B are displayed so as to correspond to a start date and an end date in an analysis target period, respectively. Then by clicking the pull-down button 45A or 45B, a pull-down menu (not shown in the figure) in which a calendar is described can be displayed. In this way, a user selects a desired date in the calendar displayed in the pull-down menu and hence can designate the date as a start date or an end date in the analysis target period. The start date or the designated date designated at that time is displayed in the text box 44A or 44B displayed on the left side of the pull-down button 45A or 45B.
Then in the analysis mode designation screen 30, it is possible to make the schedule analysis support device 1 execute schedule analysis processing of an analysis mode, an analysis method, and an analysis target period designated by a user at that time by: designating either the past analysis mode or the future prediction mode as the analysis mode to be executed at that time and designating the analysis method to be used in the designated analysis mode in the analysis mode/analysis method designation region 31; further designating the start date and the end date in the analysis target period respectively in the analysis target period designation region 32, as stated above; and then clicking the execution button 33.
Further, in the analysis mode designation screen 30, by clicking the cancel button 34, it is possible to close the analysis mode designation screen 30 without making the schedule analysis support device 1 execute such schedule analysis processing.
In the visualization parameter designation screen 50, character strings 51 representing the factor names of the delay factors registered in the delay factor management table 14 (
Then a user can change the magnification of a delay factor corresponding to a slider 52 to a magnification corresponding to the position of the slider 52 at that time in a delay model produced by an analysis method designated in the analysis mode designation screen 30 (
In the case of the present embodiment, a value (magnification) of a visualization parameter can be set steplessly in the range from “0” to “1”. Specifically, a value (magnification) of a visualization parameter: can be set at “0” by locating a slider 52 at a position on the left end of the movable range; and can be set at “1” by locating the slider 52 at a position on the right end of the movable range. Further, by locating the slider 52 at an arbitrary position in the movable range, a value (magnification) of a visualization parameter can be set at a value corresponding to the position of the slider 52 in the movable range of the slider 52.
Meanwhile,
Then in the schedule analysis result display screen 60, a magnitude of a delay time of each train in each station section calculated on the basis of a corresponding delay model, values (magnifications) of visualization parameters of respective delay factors designated by a user by using the visualization parameter designation screen 50 (
In the diagram 61, a second line segment 62A is a line segment having a length identical to a first line segment 62 in a corresponding station section and, as shown in the remark region 63, a color tone is displayed so as to deepen as a delay time increases in a station section of a train. Here, although the case of displaying a delay time of a train in a station section by five steps of color tones is shown in
When the values (magnifications) of all the visualization parameters of all the delay factors are set at “1” in the visualization parameter designation screen 50 as shown in
In contrast, when the value (magnification) of the visualization parameter of only one delay factor is set at “1” and the values (magnifications) of the visualization parameters of all the other delay factors are set at “0” as shown in
In the case of the example in
Further, when the values (magnifications) of the visualization parameters of two delay factors are set at “1” and the values (magnifications) of the visualization parameters of all the other delay factors are set at “0” as shown in
Here, in the case of the example in
Furthermore, when the values (magnifications) of the visualization parameters of all the delay factors are set at “0” as shown in
Meanwhile, a finish button 64 is installed on the lower right of the screen in the schedule analysis result display screen 60 and, by clicking the finish button 64, it is possible to finish the schedule analysis processing stated above and close the visualization parameter designation screen 50 and the schedule analysis result display screen 60.
Meanwhile,
Then in the assumed environment designation region 71, character strings 80 representing the names of respective delay factors registered on the delay factor management table 14 (
Then in the assumed environment designation region 71, assumed values of the delay factors can be set in the corresponding assumed value setting sections 82 respectively by using the input device 7 (
Further, the target train and station section setting region 72 is a region for setting a train and a station section to which the assumed values of the delay factors set in the assumed environment designation region 71 are to be applied. In the target train and station section setting region 72, a character string 72A representing the combination of the train umber of one train and one station section read from a plan schedule targeted at that time, a next button 72C, and a previous button 72B are displayed.
Then in the target train and station section setting region 72, it is possible to: switch the combination of a train number and a station section represented by a character string 72A at that time to the combination of a next train number and a next station section read from a plan schedule by clicking the next button 72C; and switch the combination of a train number and a station section represented by a character string 72A at that time to the combination of a last train number and a last station section read from a plan schedule by clicking the previous button 72B.
As a result, a user, in the simulation setting screen 70: displays a character string 72A representing the combination of one train number and one station section in the target train and station section setting region 72; successively sets assumed values of delay factors respectively in the assumed environment designation region 71; hence can set the assumed values of the delay factors in the station section of the train of the train number; successively sets assumed values of delay factors for all other trains and station sections respectively while the next button 72B and the previous button 72C are operated; and thus can set a simulation condition of a delay time of each of the trains in each of the station sections in a new plan schedule.
Further, in the simulation setting screen 70, by clicking the assumed environment read button 73, it is possible to: read a file of the previously produced assumed values of delay factors; and apply the assumed values of the delay factors prescribed in the file to the combination of a train number and a station section displayed at that time in the target train and station section setting region 72. On this occasion, the assumed values of the delay factors prescribed in the file are displayed as the assumed values of the delay factors in the assumed environment designation region 71.
Furthermore, in the simulation setting screen 70, by clicking the collective setting button 74, it is also possible to: read a file, which has previously been produced and registered, of assumed values of delay factors in each station section of each train in a plan schedule targeted at that time; and collectively set up the assumed values of the delay factors in each of the station sections of each of the trains prescribed in the file as the simulation condition in the simulation this time.
Then, in the simulation setting screen 70, by clicking the execution button 75 after setting up the assumed values in each of the station sections of each of the trains in the new plan schedule targeted at that time as a simulation condition as stated above, it is possible to make the schedule analysis support device 1 execute a simulation of predicting a delay time of the train in the station section when the trains are operated under the new plan schedule on the basis of the simulation condition. Further, on this occasion, a user can set up the value (magnification) of a visualization parameter of each of the delay factors by using the visualization parameter designation screen 50 stated above in reference to
Moreover, in the simulation setting screen 70, by clicking the cancel button 76, it is possible to: discard various settings set up heretofore by using the simulation setting screen 70; and close the simulation setting screen 70.
Successively, the plan schedule read unit 22 (
Successively, the delay time calculation unit 24 calculates a delay time of the train in the station section in the plan schedule on the basis of the plan schedule that is read from the plan schedule management table 10 by the plan schedule read unit 22, an actual schedule of the train during the analysis target period read from the actual schedule management table 11 by the actual data read unit 23, and the actual values of the delay factors in the station section of the train during the analysis target period read from the delay-factor-specific result management table 16 by the actual data read unit 23 (S4).
Further, the delay time refinement unit 25 (
Successively, the delay model generation unit 26 (
Then when the analysis mode designated by the user in the analysis mode designation screen 30 is a past analysis mode, successively the past result analysis unit 27 (
Otherwise, when the analysis mode designated by the user in the analysis mode designation screen 30 is a future prediction mode, the future prediction analysis unit 28 (
Then, when the processing of the step S7 by the past result analysis unit 27 or the processing of the step S8 by the future prediction analysis unit 28 is finished, the analysis result output unit 29 (
Successively, the processing returns to the step S1 when the user designates a next plan schedule and designates the execution of new schedule analysis processing (NO at S10) and then the processing of the step S1 and later is executed similarly to the above. On the other hand, when the user designates the finish of the schedule analysis processing after the finish of the step S9 (YES at S10), the schedule analysis processing finishes.
Successively, the mode setting unit 21 waits for either of a past analysis mode and a future prediction mode being selected as the analysis mode of the schedule analysis processing executed at that time in the analysis mode designation screen 30 (S21, S22).
Then, the mode setting unit 21, when the past analysis mode is selected as the analysis mode soon (YES at S21), sets up the analysis mode of the schedule analysis support device 1 to the past analysis mode (S22), successively calls the plan schedule read unit 22, and then finishes the mode setting processing.
On the other hand, the mode setting unit, when the future prediction mode is selected as the analysis mode (YES at S23), sets up the analysis mode of the schedule analysis support device 1 to the future prediction mode (S24), successively calls the plan schedule read unit 22, and then finishes the mode setting processing.
Actually, the plan schedule read unit 22, when called by the mode setting unit 21, starts plan schedule read processing shown in
Successively, the plan schedule read unit 22 reads the data of the plan schedule selected at the step S30 from the plan schedule management table 10 (S31), successively calls the actual data read unit 23, and then finishes the plan schedule read processing.
Actually, the actual data read unit 23, when called by the plan schedule read unit 22 as stated above, starts the actual data read processing shown in
Successively, the actual data read unit 23 reads all the data of all the trains in the actual schedule during an analysis target period designated by the user by using the analysis mode designation screen 30 in the actual schedule selected at the step S40 from the actual schedule management table 11 (S41).
Further, the actual data read unit 23 reads the data of the actual values of all the delay factors during the analysis target period from the delay-factor-specific result management table 16 (S42). Then successively, the actual data read unit 23 calls the delay time calculation unit 24 and then finishes the actual data read processing.
Actually, the delay time calculation unit 24, when called by the actual data read unit 23, starts the delay time calculation processing shown in
Successively, the delay time calculation unit 24 extracts an arrival time at the station selected at the step S52 and a departure time from the station (those are collectively referred to as arrival and departure times hereunder) of the train selected at the step S51 on the day selected at the step S50 from a plan schedule acquired by the plan schedule read unit 22 at the step S2 in the schedule analysis processing (
Here, when the type of the train selected at the step S51 is a special train such as “rapid” or “express” and the train does not stop at the station selected at the step S52 for example, a negative result (NO at S54) is given at the step S54. When the schedule of a train does not exist originally in a plan schedule too like when a train selected at the step S51 is an extra train, a negative result (NO at S54) is given at the step S54. Thus, the delay time calculation unit 24 returns to the step S52 at that time.
On the other hand, the delay time calculation unit 24, when it receives a positive result (YES at S54) in the judgment at the step S54, calculates a difference between an arrival time acquired at the step S53 and an arrival time at the station of the train specified in the plan schedule as an arrival delay time (S55). Further, the delay time calculation unit 24 calculates a difference between a train stop time at the station of the train calculated as a difference between an arrival time and a departure time at the station of the train acquired at the step S53 and a train stop time at the station of the train in the plan schedule as a train stop delay time (S56).
Moreover, the delay time calculation unit 24 stores an arrival delay time calculated at the step S55 and a train stop delay time calculated at the step S56 into the corresponding arrival delay time column 12DAA (
Successively, the delay time calculation unit 24, with regard to the train selected at the step S51 on the day selected at the step S50, judges whether or not the execution of the processing of the step S53 and later to all the stations on the target railway is finished (S58). Then the delay time calculation unit 24, when it receives a negative result (NO at S58) in the judgment, returns to the step S52 and then repeats the processing of the steps S52 to S58 while the station selected at the step S52 is switched to another unprocessed station sequentially.
Then the delay time calculation unit 24, when it receives a positive result at the step S58 (YES at S58) by finishing the execution of the processing of the step S53 and later to all the stations on the target railway with regard to the train selected at the step S51 soon, judges whether or not the execution of the processing of the steps S52 to S58 to all the trains to which the train numbers are given with regard to the day selected at the step S50 is finished (S59).
The delay time calculation unit 24, when it receives a negative result (NO at S59) in the judgment, returns to the step S51 and then repeats the processing of the steps S51 to S59 while the train selected at the step S51 is switched to another unprocessed train sequentially.
Then the delay time calculation unit 24, when it receives a positive result at the step S59 (YES at S59) by finishing the execution of the processing of the steps S52 to S58 to all the trains to which the train numbers are given with regard to the day selected at the step S50 soon, judges whether or not the execution of the processing of the steps S51 to S59 with regard to all the days during the analysis target period is finished (S60).
The delay time calculation unit 24, when it receives a negative result (NO at S60) in the judgment, returns to the step S50 and then repeats the processing of the steps S50 to S60 while the day selected at the step S50 is switched to another unprocessed day sequentially. Then the delay time calculation unit 24, when it receives a positive result at the step S60 (YES at S60) by finishing the execution of the processing of the steps S51 to S59 with regard to all the days during the designated analysis target period soon, calls the delay time refinement unit 25 and then finishes the delay time calculation processing.
Actually, the delay time refinement unit 25, when called by the delay time calculation unit 24, starts the delay time refinement processing shown in
Successively, the delay time refinement unit 25 extracts and thus acquires the actual schedule (an arrival time at each of stations and a departure time from each of stations) of the train selected at the step S71 on the day selected at the step S70 from the actual schedule read from the actual schedule management table 11 (
Successively, the delay time refinement unit 25: selects one of unprocessed station sections from among the station sections on a target railway (S74); calculates an arrival delay time and a train stop delay time excluding secondary delay respectively while the schedule restrictions acquired at the step S73 are taken into consideration in the station section of the train selected at the step S71 (S75); and registers the calculation result in the post-refinement train delay time management table 15 (S76).
For example, a case where an arrival delay time of “III Station” in the station section of “II Station to III Station” of the train having the train number of “1” is “00:04:00” and an arrival delay time of “III Station” in the station section of “II Station to III Station” of the train having the train number of “2” is “00:06:00” as shown in
On this occasion, as shown in
Meanwhile, for example, a case where an arrival delay time at “II Station” in the station section of “I Station to II Station” is “00:04:00” and an arrival delay time at “III Station” in the station section of “II Station to III Station” is “00:10:00” of the train having the train number of “1” as shown in
On this occasion, as shown in
Here, the post-refinement train delay time management table 15 is a table produced by the delay time refinement unit 25 in the state of: copying only the entire information of the train number column 12A, the train type column 12B, and the up-or-down classification column 12C in each of the rows of the train delay time management table 12 stated above in reference to
Successively, the delay time refinement unit 25, with respect to the train selected at the step S71 on the day selected at the step S70, judges whether or not the execution of the processing of the step S72 and later to all the station sections on the target railway is finished (S77). Then the delay time refinement unit 25, when it receives a negative result (NO at S77) in the judgment, returns to the step S74 and then repeats the processing of the steps S74 to S77 while the station selected at the step S74 is switched to another unprocessed station sequentially.
Then the delay time refinement unit 25, when it receives a positive result at the step S77 (YES at S77) by finishing the execution of the processing of the step S75 and later to all the station sections on the target railway with regard to the train selected at the step S71 soon, judges whether or not the execution of the processing of the steps S72 to S77 to all the trains to which the train numbers are given with regard to the day selected at the step S70 is finished (S78).
The delay time refinement unit 25, when it receives a negative result (NO at S78) in the judgment, returns to the step S71 and then repeats the processing of the steps S71 to S78 while the train selected at the step S71 is switched to another unprocessed train sequentially.
Then the delay time refinement unit 25, when it receives a positive result at the step S78 (YES at S78) by finishing the execution of the processing of the steps S72 to S77 to all the trains to which the train numbers are given with regard to the day selected at the step S70 soon, judges whether or not the execution of the processing of the steps S71 to S78 with regard to all the days during the designated analysis target period is finished (S79).
The delay time refinement unit 25, when it receives a negative result (NO at S79) in the judgment, returns to the step S70 and then repeats the processing of the steps S70 to S79 while the day selected at the step S70 is switched to another unprocessed day sequentially. Then the delay time refinement unit 25, when it receives a positive result at the step S79 (YES at S79) by finishing the execution of the processing of the steps S71 to S78 with regard to all the days during the analysis target period soon, calls the delay model generation unit 26 and then finishes the delay time refinement processing.
Actually, the delay model generation unit 26, when called by the delay time refinement unit 25, starts the delay model generation processing shown in
Successively, the delay model generation unit 26, on the basis of the confirmation result at the step S80, generates a delay model by: using information stored in the post-refinement train delay time management table 15 (
Successively, the delay model generation unit 26 stores the generated delay models in the delay model management table 18 (
Then the delay model generation unit 26 successively: calls the past result analysis unit 27 when the analysis mode designated by a user in the analysis mode designation screen 30 is a past analysis mode; calls the future prediction analysis unit 28 when the analysis mode designated by a user in the analysis mode designation screen 30 is a future prediction mode; and then finishes the delay model generation processing.
Actually, the past result analysis unit 27, when called by the delay model generation section 26, starts the past result analysis processing shown in
Successively, the past result analysis unit 27 multiplies the coefficients of the delay factors in the delay model read at the step S91 by the values (magnifications) of the visualization parameters given to the delay factors in the visualization parameter designation screen 50 at that time and the actual values of the corresponding delay factors read from the delay-factor-specific result management table 16 (
Here, since the initial value of a visualization parameter is “1”, “1” is applied as the value (magnification) of a visualization parameter at the stage of the step S7 in the schedule analysis processing stated above in reference to
Further, when the analysis target period is not only one day but lasts for several days, at the step S92, the past result analysis unit 27, with respect to each delay factor, calculates the average value of the actual values of the delay factor during the several days and multiplies the coefficient of the delay factor by the calculated average value and the value (magnification) of the visualization parameter given to the delay factor at that time. As a result, it is possible to calculate an average delay time during the analysis target period corresponding to the values of the corresponding visualization parameters in the station section selected at the step S90 of the train selected at the step S90 by using the corresponding delay model by using the corresponding delay model.
Successively, the past result analysis unit 27 links a delay time calculated at the step S92 to a corresponding delay model (S93). Further, the past result analysis unit 27 successively judges whether or not the execution of the processing of the steps S91 to S93 with respect to the combinations of all the station sections of all the trains in the plan schedule of the analysis target is finished (S94).
Then the past result analysis unit 27, when it receives a negative result (NO at S94) in the judgment, returns to the step S90 and then repeats the processing of the steps S90 to S94 while the combination of the train and the station section selected at the step S90 is switched to another unprocessed combination sequentially.
Then, the past result analysis unit 27, when it receives a positive result (YES at S94) at the step S94 by finishing the execution of the processing of the steps S91 to S93 with regard to the combinations of all the trains and all the station sections in the plan schedule of the analysis target soon, calls the analysis result output unit 29 and then finishes the past result analysis processing.
Actually, the future prediction analysis unit 28, when called by the delay model generation unit 26, starts the future prediction analysis processing shown in
Then, the future prediction analysis unit 28 successively, when a user clicks the execution button 75 (
Successively, the future prediction analysis unit 28 reads the delay model corresponding to the station section selected at the step S103 of the train selected at the step S103 from the delay model management table 18 (
Further, the future prediction analysis unit 28 calculates the predicted values of the delay times of respective delay factors in the station section of the train selected at the step S103 by applying the assumed values of the delay factors acquired at the step S102 to the delay model acquired at the step S104 (S105).
Further, the future prediction analysis unit 28: multiplies the delay times of the delay factors calculated at the step S105 by the values (magnifications) of the corresponding visualization parameters; and then adds up the multiplication results (S106). Here, since the initial value of a visualization parameter is “1” as stated above, the delay times of the delay factors are multiplied by “1” respectively at the stage of the step S9 in the schedule analysis processing stated above in reference to
Successively, the future prediction analysis unit 28 links the predicted value of the delay time calculated at the step S106 to a corresponding delay model (S107). Further, the future prediction analysis unit 28 then judges whether or not the execution of the processing of the steps S104 to S107 to the combination of the train and the station section to be targeted in the plan schedule of the analysis target is finished (S108).
Then the future prediction analysis unit 28, when it receives a negative result (NO at S108) in the judgment, returns to the step S103 and then repeats the processing of the steps S103 to S108 while the combination of the train and the station section selected at the step S103 is switched to another unprocessed combination sequentially.
Then the future prediction analysis unit 28, when it receives a positive result (YES at S108) at the step S108 by finishing the execution of the processing of the steps S104 to S107 to all the combinations of the trains and the station sections to be targeted in the plan schedule of the analysis target soon, calls the analysis result output unit 29 and then finishes the future prediction analysis processing.
Actually, the analysis result output unit 29, when called by the past result analysis unit 27 or the future prediction analysis unit 28, starts the analysis result output processing shown in
Specifically, the analysis result output unit 29 displays the schedule analysis result display screen 60 including the diagram 61 in which a second line segment 62A is displayed by a color tone corresponding to the magnitude of a delay time or a predicted delay time in accordance with the delay time associated with a delay model in each station section of each train by the past result analysis unit 27 at the step S93 in the past result analysis processing (
Successively, the analysis result output unit 29: displays the visualization parameter designation screen 50 (
Then the analysis result output unit 29, when the value (magnification) of the visualization parameter of any one of delay factors is changed in the visualization parameter designation screen 50, gives instructions to the past result analysis unit 27 or the future prediction analysis unit 28 so as to recalculate a delay time of each train in each station section in accordance with the value (magnification) of the visualization parameter of the delay factor reset at that time (S114).
Specifically, the analysis result output unit 29, when the analysis mode at that time is the past analysis mode: notifies the revised values (magnifications) of the visualization parameters of respective delay factors to the past result analysis unit 27; and then gives instructions to the past result analysis unit 27 so as to recalculate a delay time of each train in each station section. In this way, the past result analysis unit 27 that has received the instructions reexecutes the past result analysis processing stated above in reference to
Then the analysis result output unit 29, when the analysis mode at that time is the future prediction mode: notifies the revised values (magnifications) of the visualization parameters of respective delay factors to the future prediction analysis unit 28; and then gives instructions to the future prediction analysis unit 28 so as to recalculate a predicted delay time in a requested station section of a requested train. In this way, the future prediction analysis unit 28 that has received the instructions reexecutes the future prediction analysis processing stated above in reference to
Then the analysis result output unit 29 successively, when the reexecution of the past result analysis processing by the past result analysis unit 27 or the reexecution of the future prediction analysis processing by the future prediction analysis unit 28 is completed, updates the diagram 61 in the schedule analysis result display screen 60 through processing similar to the step S110 on the basis of the processing result of the past result analysis processing or the future prediction analysis processing executed at that time (S115).
Successively, the analysis result output unit 29 returns to the step S112 and then repeats the processing of the step S112 and later in the same manner as described above. Then the analysis result output unit 29, when the finish button 64 in the schedule analysis result display screen 60 is clicked soon, closes the schedule analysis result display screen 60 and the visualization parameter designation screen 50 and then finishes the analysis result output processing.
In a schedule analysis support device 1 according to the present embodiment as stated above, during a past analysis mode, a delay time of each train in each station section is classified into delay times of respective delay factors in a targeted plan schedule and on the other hand, during a future prediction mode, a delay time obtained by predicting delay times of respective delay factors in each station section of each train and multiplying the delay times of the delay factors by the magnifications corresponding to the setting of the visualization parameters of the delay factors designated by using the visualization parameter designation screen 50 and adding up the multiplication results is displayed as a line segment 62A of a tone corresponding to the magnitude of the delay time in the schedule analysis result display screen 60 in an overlapping manner and hence a user can acquire the influences of delay factors in each of the station sections of each of the trains easily and, to that extent, it is possible to reduce work time and facilitate work in the analysis of a plan schedule. The schedule analysis support device 1 according to the present embodiment therefore can lead to a subsequent temporary operational measure and rapid operation schedule revision and resultantly the reduction of delay and congestion can be expected.
Further, in the schedule analysis support device 1, a delay time of each train in each station section after the influence of secondary delay is excluded while schedule restrictions are taken into consideration is calculated and a delay model having delay factors as the variables is generated in each of the station sections of each of the trains on the basis of the calculation result and hence an accurate delay model excluding the influence of the secondary delay can be generated and, to that extent, a user can execute more accurate schedule analysis.
Meanwhile, although the case of applying the present invention to the analysis of a plan schedule of a transport system using a train as the target mobile body is described in the aforementioned embodiment, the present invention is not limited to the case and can be widely applied also to the analysis of a plan schedule in a transport system using a bus, a ship, or an airplane, other than a train, as such a mobile body.
Further, although the case where the schedule analysis support device 1 has the two analysis modes of a past analysis mode and a future prediction mode as the analysis modes is described in the aforementioned embodiment, the present invention is not limited to the case and can also apply to the case where the schedule analysis support device 1 has only either of the past analysis mode and the future prediction mode.
Furthermore, although the case of displaying the processing result of the past result analysis processing and the processing result of the future prediction analysis processing in the representation forms stated above in reference to
The present invention relates to a schedule analysis support device and can widely apply to the analysis of a plan schedule in various transport systems in which a mobile body is operated in accordance with the plan schedule.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2018-078764 | Apr 2018 | JP | national |
