Operation proposal system and operation proposal creation method

Abstract

To provide an operation proposal system capable of creating an operation proposal that fits the demand of people to stay that changes according to the operation status and surrounding conditions.SOLUTION: An operation proposal system 10 for creating an operation proposal of transportation means includes: a computing unit (processor) that performs a predetermined operation; and a storage device to which the computing unit is accessible. The computing unit includes a stay demand prediction unit 21 for predicting the stay demand of passengers at the time of transfer using a stay demand prediction model; and an operation proposal unit 22 that the computing unit generates an operation proposal with a satisfactory evaluation value on the operation proposal based on the predicted stay demand.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an operation proposal system for creating operation proposals.

Background Art

[0002] Efforts are underway to stimulate demand during movement and lead to revenue, such as an increase in commercial facilities within the station premises. Users are also changing their behavior to sequentially determine the places to satisfy their needs according to the operation status of public transportation, rather than satisfying their needs at predetermined places. To understand human behavior, technologies for estimating the time required for transfers and behavior patterns from a person's behavior history are known.

[0003] As background art in this technical field, there is the following prior art. Patent Document 1 (International Publication No. 2018 / 087811) discloses a train schedule proposal system having a data server that stores information on passengers collected at a plurality of stations, and a calculation server that stores a vehicle waiting passenger number prediction program and a vehicle number increase / decrease determination program. The calculation server executes the vehicle waiting passenger number prediction program and the vehicle number increase / decrease determination program at predetermined time intervals respectively. The vehicle waiting passenger number prediction program predicts the number of vehicle waiting passengers at a plurality of stations in a predetermined time period based on the information on passengers, and the vehicle number increase / decrease determination program determines the necessity of increasing or decreasing the number of vehicles in a predetermined time period based on the number of vehicle waiting passengers at a plurality of stations in the predetermined time period (see the abstract).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] By the method described above, it is difficult to grasp small demands that occur during movement and change rapidly from a person's behavior history. However, prediction of staying demands such as shopping is required in accordance with the margin of transfer time at transfer stations.

[0006] An object of the present invention is to create an operation proposal that conforms to a person's staying demand that changes according to the operation situation and the surrounding situation.

Means for Solving the Problem

[0007] A typical example of the invention disclosed in the present application is as follows. That is, an operation proposal system for creating an operation proposal for a transportation means, comprising an arithmetic device that executes predetermined processing and a storage device accessible to the arithmetic device, wherein the arithmetic device uses peripheral information, passenger information, and operation information, generates a movement demand prediction model with the entry station and the entry time zone, which are part of the passenger information, as explanatory variables and the final destination and the probability of moving to the final destination, which are part of the passenger information, as objective variables, and uses the entry station, the entry time zone, the final destination, the operation situation, the attributes of the passengers, and the peripheral information, which are part of the passenger information, as explanatory variables, and has a prediction model estimation unit for predicting the probability of occurrence of staying, the purpose of staying, the distribution of occurrence locations, and Stay demand as objective variables Generate a stay demand prediction model The arithmetic device receives, as inputs to the staying demand prediction model, the operation situation, the attributes of the passengers, the peripheral information, and the probability of moving to each final destination for each entry station, each entry time zone, and each final destination output by the movement demand prediction model, predicts and outputs a combination of the purpose of staying and the staying time period of the passengers at the time of transfer for each purpose of staying, and the arithmetic device has a movement proposal unit that generates a movement proposal with a good movement proposal evaluation value, which is at least one of the bias of the number of passengers boarding and the waiting time of the passengers for transfer, based on the predicted staying demand. The movement proposal unit applies changes in the departure time of the operation flight or the addition of a special flight to the plurality of pattern operation information to create a plurality of operation change plans, and the plurality of created operation change plans and the predicted Stay demand Stay demand ​Based on this, the number of passengers on each operating train is calculated, the created train operation change plan and the number of passengers on each operating train calculated are evaluated, the train operation proposal evaluation value of each train operation change plan is calculated with a predetermined weighting, and a train operation change plan with a high calculated train operation proposal evaluation value is selected.

Effect of the Invention

[0008] According to one aspect of the present invention, the demand during the movement of people can be estimated with high accuracy, and a train operation change plan corresponding to the demand can be proposed. Problems, configurations, and effects other than those described above will be clarified by the description of the following embodiments.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Embodiments for Carrying Out the Invention

[0010] First, an overview of the operation proposal system 10 of the embodiment of the present invention will be described. The operation proposal system 10 of this embodiment collects operation information, passenger information, and surrounding information of transportation means such as trains and buses, estimates stay information from the operation information and passenger information, and estimates a travel demand prediction model, a stay demand prediction model, and Transfer time prediction model from the stay information and surrounding information. Then, the stay demand (stay distribution) is predicted from the daily passenger information using the stay demand prediction model and the transfer time prediction model. Furthermore, according to the operation proposal KPI weighting and the operation proposal constraint conditions, an operation proposal for optimizing the operation proposal evaluation value (KPI) of the transportation means is created from the stay distribution.

[0011] FIG. 1 is a block diagram showing the logical configuration of the operation proposal system 10 of the embodiment of the present invention.

[0012] The operation proposal system 10 includes a peripheral information acquisition unit 11, a passenger information acquisition unit 12, an operation information acquisition unit 13, a peripheral information storage unit 14, a passenger information storage unit 15, an operation information storage unit 16, a prediction model estimation unit 17, a travel demand prediction model storage unit 18, a stay demand prediction model storage unit 19, a transfer time prediction model storage unit 20, a stay demand prediction unit 21, and an operation proposal unit 22.

[0013] The peripheral information acquisition unit 11 acquires peripheral information (for example, input information for predicting travel demand, such as weather forecasts and information on events along the line) obtained from the open data system 50 and stores it in the peripheral information storage unit 14. The peripheral information stored in the peripheral information storage unit 14 will be described later with reference to FIG. 7. The open data system 50 is a computer system that provides data affecting human behavior, such as weather forecasts and regional information. The peripheral information acquisition unit 11 may acquire not only publicly available open data but also data provided individually (for example, for a fee).

[0014] The passenger information acquisition unit 12 acquires passenger information from the ticket gate system 60 and the settlement system 70 and stores it in the passenger information storage unit 15. The passenger information stored in the passenger information storage unit 15 will be described later with reference to FIGS. 8, 9, and 10. The ticket gate system 60 is a computer system that collects entry and exit data of railway ticket gates and boarding and alighting data of fare boxes of shared buses. The settlement system 70 is a computer system that manages purchase records, and for example, based on cashless settlement and credit card usage history, it can collect Data related to the purchaser, purchase date and time, and purchase amount.

[0015] The operation information acquisition unit 13 acquires operation information from the operation management system 80 and stores it in the operation information storage unit 16. The operation information stored in the operation information storage unit 16 will be described later with reference to FIG. 11. The operation management system 80 is a computer system used by railway operators and shared bus operators to manage the operation of railways and buses.

[0016] The prediction model estimation unit 17 generates a travel demand prediction model, a stay demand prediction model, and a transfer time prediction model using the data stored in the surrounding information accumulation unit 14, the passenger information accumulation unit 15, and the operation information accumulation unit 16, and stores each of them in the travel demand prediction model accumulation unit 18, the stay demand prediction model accumulation unit 19, and the transfer time prediction model accumulation unit 20. The travel demand prediction model stored in the travel demand prediction model accumulation unit 18 will be described later with reference to FIG. 12. The stay demand prediction model stored in the stay demand prediction model accumulation unit 19 will be described later with reference to FIG. 13. The transfer time prediction model stored in the transfer time prediction model accumulation unit 20 will be described later with reference to FIG. 14.

[0017] The stay demand prediction unit 21 inputs the daily data of the surrounding information, the daily data of the passenger information, and the daily data of the operation information into the travel demand prediction model, the stay demand prediction model, and the transfer time prediction model to predict the stay demand of the passengers.

[0018] The operation proposal unit 22 creates an operation proposal for trains and buses using the stay demand predicted by the stay demand prediction unit 21 and transmits it to the operation management system 80.

[0019] FIG. 2 is a block diagram showing the physical configuration of the operation proposal system 10.

[0020] The operation proposal system 10 is composed of a computer having a processor (CPU) 1, a memory 2, an auxiliary storage device 3, and a communication interface 4. The operation proposal system 10 may have an input interface 5 and an output interface 6.

[0021] The processor 1 is an arithmetic unit that executes the programs stored in the memory 2. By the processor 1 executing various programs, the functions of each functional unit of the operation proposal system 10 (for example, the surrounding information acquisition unit 11, the passenger information acquisition unit 12, the operation information acquisition unit 13, the prediction model estimation unit 17, the stay demand prediction unit 21, the operation proposal unit 22, etc.) are realized. Note that a part of the processing performed by the processor 1 executing the program may be executed by another arithmetic unit (for example, hardware such as an ASIC or an FPGA).

[0022] The memory 2 includes a ROM which is a non-volatile memory element and a RAM which is a volatile memory element. The ROM stores unchangeable programs (such as BIOS). The RAM is a high-speed and volatile memory element such as a DRAM (Dynamic Random Access Memory), and temporarily stores programs executed by the processor 1 and data used during program execution.

[0023] The auxiliary storage device 3 is a large-capacity and non-volatile storage device such as a magnetic storage device (HDD) or a flash memory (SSD). Also, the auxiliary storage device 3 stores data (such as the peripheral information storage unit 14, the passenger information storage unit 15, the operation information storage unit 16, etc. Data stored in ) used by the processor 1 during program execution, and programs executed by the processor 1. That is, the programs are read from the auxiliary storage device 3, loaded into the memory 2, and executed by the processor 1 to realize each function of the operation proposal system 10.

[0024] The communication interface 4 is a network interface device that controls communication with other devices (such as the operation management system 80) according to a predetermined protocol.

[0025] The input interface 5 is an interface to which input devices such as the keyboard 7 and the mouse 8 are connected and which receives input from the operator. The output interface 6 is an interface to which output devices such as the display device 9 and a printer (not shown) are connected and which outputs the execution result of the program in a form visible to the operator. Note that a terminal connected via a network may provide the input interface 5 and the output interface 6.

[0026] The program executed by the processor 1 is provided to the operation proposal system 10 via a removable medium (such as a CD-ROM or a flash memory) or a network and stored in the non-volatile auxiliary storage device 3 which is a non-temporary storage medium. Therefore, the operation proposal system 10 may have an interface for reading data from the removable medium.

[0027] The operation proposal system 10 is a computer system configured physically on one computer or on a plurality of computers configured logically or physically, and may operate on a virtual computer built on a plurality of physical computer resources. For example, the peripheral information acquisition unit 11, the passenger information acquisition unit 12, the operation information acquisition unit 13, the prediction model estimation unit 17, the stay demand prediction unit 21, and the operation proposal unit 22 may each operate on a separate physical or logical computer, or a plurality of them may be combined and operate on one physical or logical computer.

[0028] Next, the processing executed by the operation proposal system 10 of this embodiment will be described. In the processing described below, a transfer from a train to a bus is assumed, but other transportation means may also be used.

[0029] FIG. 3 is a flowchart of the main processing executed by the operation proposal system 10.

[0030] First, the peripheral information acquisition unit 11 acquires peripheral information from the open data system 50 and stores the acquired peripheral information in the peripheral information accumulation unit 14. Also, the passenger information acquisition unit 12 acquires passenger information from the ticket gate system 60 and the settlement system 70 and stores the acquired passenger information in the passenger information accumulation unit 15. Also, the operation information acquisition unit 13 acquires operation information from the operation management system 80 and stores the acquired operation information in the operation information accumulation unit 16 (100).

[0031] Next, the prediction model estimation unit 17 estimates the stay information based on the passenger information stored in the passenger information accumulation unit 15 and the operation information stored in the operation information accumulation unit 16 (101). For example, by matching the passenger information (movement) and the passenger information (consumption) by passenger ID, if there is a purchase behavior during the movement, it is estimated that the passenger is staying at the transfer station for shopping. Also, if the transfer exceeds the standard transfer time by a large margin, it is estimated that the passenger is staying at the transfer station with an unclear purpose.

[0032] Next, the prediction model estimation unit 17, based on the surrounding information stored in the surrounding information accumulation unit 14, the passenger information stored in the passenger information accumulation unit 15, and the operation information stored in the operation information accumulation unit 16, Transfer time prediction model estimates [the transfer time prediction model] and stores it in the transfer time prediction model accumulation unit 20 (102). For example, a model is created with the transfer attributes (location, transportation, time zone) as explanatory variables and the distribution of the stay time as the target variable. The transfer time prediction model may be composed of a neural network that learns the passenger information and the operation information as teacher data, or a regression prediction model may be generated by performing multivariate analysis on the passenger information and the operation information.

[0033] Next, the prediction model estimation unit 17 estimates the travel demand prediction model based on the surrounding information stored in the surrounding information accumulation unit 14, the passenger information stored in the passenger information accumulation unit 15, the operation information stored in the operation information accumulation unit 16, and the stay information estimated in step 101, and stores it in the travel demand prediction model accumulation unit 18 (103). For example, with the entry station and the entry time as explanatory variables and the probability of the destination as the target variable. A model for obtaining the number of people going to a specific destination is created. The travel demand prediction model may be composed of a neural network that learns the surrounding information, the passenger information, the operation information, and the stay information as teacher data, or a regression prediction model may be generated by performing multivariate analysis on the surrounding information, the passenger information, the operation information, and the stay information.

[0034] Next, the stay demand prediction model estimation unit 17 estimates a stay demand prediction model based on the surrounding information stored in the surrounding information storage unit 14, the passenger information stored in the passenger information storage unit 15, the operation information stored in the operation information storage unit 16, and the stay information estimated in step 101, and stores it in the stay demand prediction model storage unit 19 (104). The method for estimating the stay demand prediction model will be described later with reference to FIG. 4. The stay demand prediction model may be composed of a neural network that learns the surrounding information, passenger information, operation information, and stay information as teacher data, or a regression prediction model may be generated by performing multivariate analysis on the surrounding information, passenger information, operation information, and stay information.

[0035] Next, the surrounding information acquisition unit 11 acquires the surrounding information of the current day from the open data system 50 and stores the acquired surrounding information in the surrounding information storage unit 14. Also, the passenger information acquisition unit 12 acquires the passenger information of the current day from the ticket gate system 60 and the settlement system 70 and stores the acquired passenger information in the passenger information storage unit 15. Further, the operation information acquisition unit 13 acquires the operation information of the current day from the operation management system 80 and stores the acquired operation information in the operation information storage unit 16 (105).

[0036] Next, the stay demand prediction unit 21 inputs the current day's data of the operation information into the travel demand prediction model to predict the travel demand of passengers (106).

[0037] Next, the stay demand prediction unit 21 Travel demand and inputs it into the stay demand prediction model to predict the stay demand of passengers (107). The stay demand prediction process will be described later with reference to FIG. 5, and the predicted stay demand will be described later with reference to FIG. 15.

[0038] Next, based on the distribution of the predicted stay demand, the operation proposal unit 22 creates an operation proposal with good KPIs (108). The operation proposal creation process will be described later with reference to FIG. 6. With an operation proposal with good KPIs, the congestion level of transportation facilities can be leveled, and the waiting time during passenger transfers can be reduced.

[0039] FIG. 4 is a flowchart of the stay demand prediction model estimation process.

[0040] First, the stay demand prediction unit 17 clusters the entry station of the passenger information (movement), the last exit station of a series of movements, the passenger information (attributes), the passenger information (consumption), and the surrounding information to generate N clusters (110), and repeatedly executes the processes of steps 111 to 113 for each cluster.

[0041] After that, the stay demand prediction unit 17 calculates the occurrence probability by dividing the number of cases with stay information for each train delay by the total number of cases with stay demand of the passengers belonging to the cluster, and creates a prediction model for the stay demand due to the train delay of the calculated occurrence probability (111).

[0042] Next, the stay demand prediction unit 17 calculates the number of stations where stays occur for each train delay among the passengers belonging to the cluster, calculates the occurrence location distribution by dividing the number of cases for each station where stays occur by the total number of data where stays occur, and creates a regression prediction model for the train delay of the occurrence probability (112).

[0043] Next, the stay demand prediction unit 17 divides the stay time into predetermined intervals to calculate the number of cases, calculates the probability of each interval by dividing the calculated number of stay cases by the total number of stay cases, and creates a stay demand (113).

[0044] The processes of steps 111 to 113 above are executed for each cluster, and the process is repeated until it is completed for all clusters.

[0045] FIG. 5 is a flowchart of the stay demand prediction process.

[0046] First, the stay demand prediction unit 21 clusters the movement demand predicted in step 106 for each combination of the entry station, the occupation, and the final exit station to generate N clusters (120), and repeatedly executes the processes of steps 121 to 122 for each cluster.

[0047] After that, the stay demand prediction unit 21 identifies the model that matches the cluster among the stay demand prediction models (121).

[0048] Next, the stay demand prediction unit 21 predicts the number of stays and the locations where stays occur based on the number of models and clusters, and creates a stay demand distribution for each cluster (122).

[0049] The processes in steps 121 to 122 above are executed for each cluster, and the processing is repeated until it is completed for all clusters.

[0050] Thereafter, the stay demand prediction unit 21 overlaps the stay demands for each cluster by location to create a stay demand for each location (123).

[0051] FIG. 6 is a flowchart of the operation proposal creation process.

[0052] First, the operation proposal unit 22 repeatedly executes the processes in steps 130 to 133 a predetermined number of times for each predicted stay demand.

[0053] Thereafter, the operation proposal unit 22 applies a change in the bus departure time randomly selected or the addition of an extra bus to the operation information to create a new operation change plan (130).

[0054] Next, the operation proposal unit 22 calculates the number of passengers on each bus based on the created operation change plan and the predicted stay demand (131).

[0055] Next, the operation proposal unit 22 evaluates the created operation change plan and the KPI of the number of passengers on each bus, and calculates the evaluation value of each operation change plan from the operation proposal KPI weighting and the operation proposal constraint conditions (132). The evaluation value can be obtained, for example, by using the linear weighted sum method to multiply each KPI weighting and the evaluation value of each KPI respectively and then summing them up to get one value. If the operation change plan Operation proposal constraint conditions does not meet the requirements, a predetermined penalty is subtracted from the evaluation value. The penalty is set to a very large value (or infinite) for Operation proposal constraint conditions what must be observed, and a value that is desired to be observed as much as possible Operation proposal constraint conditionsIt may be adjusted by setting a finite value for it.

[0056] Next, if the evaluation value of the current operation change plan is higher than that of the previous operation change plan, the operation proposal unit 22 adopts the current operation change plan. If the evaluation value of the current operation change plan is lower than that of the previous operation change plan, the current operation change plan is discarded and the process proceeds to the next iteration (133).

[0057] The processes of steps 130 to 133 above are repeated a predetermined number of times, and an operation change plan with a high evaluation value is selected for each stay demand. This process is executed for each stay demand, and the process is repeated until it is completed for all stay demands.

[0058] FIG. 7 is a diagram showing a configuration example of the peripheral information stored in the peripheral information storage unit 14. The peripheral information is data of the line obtained from the open data system 50 (for example, weather forecast and information on events along the line), is used to predict the travel demand, and includes a date, a range, and content. The date is the date on which the event described in the content column occurs. The range is the range to which the influence of the event extends. The content is the content of the event.

[0059] Next, a configuration example of the passenger information stored in the passenger information storage unit 15 will be described. The passenger information is divided into passenger information (attributes) representing the attributes of passengers, passenger information (consumption) representing the purchase behavior of passengers, and passenger information (movement) representing the movement of passengers, but it may be configured in one table in association with the passenger ID.

[0060] FIG. 8 is a diagram showing a configuration example of the passenger information (attributes) stored in the passenger information storage unit 15. The passenger information (attributes) is information representing the attributes of passengers, and may be, for example, information registered at the time of purchase of a transportation IC card. The passenger information (attributes) includes a passenger ID and an occupation. The passenger information (attributes) Of behavior If it is information effective for the classification of passengers, age, gender, address, etc. may also be recorded.

[0061] FIG. 9 is a diagram showing a configuration example of passenger information (consumption) stored in the passenger information storage unit 15. The passenger information (consumption) is information representing the purchasing behavior of a passenger obtained from the settlement system 70, and may be, for example, information on cashless settlement or the usage history of a credit card. The passenger information (consumption) includes a passenger ID, a date and time, and a store. The date and time is the date and time when the passenger's purchasing behavior was performed. The store is the name of the store where the passenger's purchasing behavior was performed. The passenger information (consumption) may record the purchase amount or the like if it is information effective for classifying the passenger's purchases. Of behavior If it is information effective for classifying the passenger's purchases, the purchase amount or the like may be recorded.

[0062] FIG. 10 is a diagram showing a configuration example of passenger information (movement) stored in the passenger information storage unit 15. The passenger information (movement) is information representing the movement of a passenger obtained from the ticket gate system 60, and may be, for example, OD data such as entrance / exit data of a railway ticket gate or boarding / alighting data of a fare box of a shared bus. The passenger information (movement) includes a passenger ID, an entrance station, an entrance date and time, an exit station, and an exit date and time. The entrance station and the entrance date and time are the place and time when the passenger boarded the transportation means. The exit station and the exit date and time are the place and time when the passenger alighted from the transportation means.

[0063] FIG. 11 is a diagram showing a configuration example of operation information stored in the operation information storage unit 16. The operation information is information on the scheduled operation times of trains and buses created by railway operators and shared bus operators, obtained from the operation management system 80, and is daily data including a transportation means ID, a station, an arrival time, a departure time, and a delay. The transportation means ID is unique identification information for trains and buses. The station is the place (station or bus stop) where the train or bus departs and arrives. The arrival time is the arrival time of the train or bus, and the departure time is the departure time of the train or bus. The delay is the difference between the actual departure time and the scheduled departure time of the train or bus.

[0064] FIG. 12 is a diagram showing an example of a travel demand prediction model stored in the travel demand prediction model storage unit 18. The travel demand prediction model uses the entrance station and the entrance time zone as explanatory variables, and the final destination and the probability of moving to the final destination as objective variables.

[0065] FIG. 13 is a diagram showing an example of the stay demand prediction model stored in the stay demand prediction model storage unit 19. The stay demand prediction model uses the entry station, entry time zone, final destination, operation status, passenger attributes (occupation), and surrounding information as explanatory variables, and the occurrence probability of stay, purpose of stay, occurrence location distribution, and Stay distribution are used as target variables. The occurrence probability is the probability that the stay occurs. The occurrence location distribution is a set of the location where the stay occurs and the probability. Stay distribution is represented by a link to other information.

[0066] FIG. 14 is a diagram showing an example of the transfer time prediction model stored in the transfer time prediction model storage unit 20. The transfer time prediction model uses the target transfer as an explanatory variable, and the stay time interval and the probability of the interval as target variables.

[0067] FIG. 15 is a diagram showing an example of the stay demand predicted by the stay demand prediction unit 21. The stay demand is represented by the distribution of the number of staying people for each occurrence location, and includes the number of staying people / the total number of transfer people, the purpose of stay, the occurrence location of stay, and a link to the stay distribution. The stay distribution represents the number of people for each combination of the purpose of stay and the stay time interval.

[0068] Next, the daily data used for prediction will be described with reference to FIGS. 16 and 17, and the setting parameters will be described with reference to FIGS. 18 and 19.

[0069] FIG. 16 is a diagram showing a configuration example of the daily data of the passenger information (movement) stored in the passenger information storage unit 15. The daily data of the passenger information (movement) has the same format as the passenger information (movement) shown in FIG. 10, but since the entry / exit data is for the current day, the entry time instead of the entry date and the exit time instead of the exit date are used. The daily data of the passenger information (movement) is used for predicting the stay demand, so it is the data of the currently moving passengers, and the exit station and exit time are "not yet". The rest is the same as the passenger information (movement) shown in FIG. 10.

[0070] FIG. 17 is a diagram showing a configuration example of the daily data of the operation information stored in the operation information storage unit 16. The daily data of the operation information is in the same format as the operation information shown in FIG. 11, but since it is the current operation information of the day, the current time After has a departure / arrival status of "not yet". The rest is the same as the operation information shown in FIG. 11.

[0071] FIG. 18 is a diagram for explaining an example of the operation proposal KPI weighting coefficient. The weighting coefficient indicates the KPI that the operation proposal emphasizes, reduces the bias in the number of passengers, aims for the dispersion of congestion, and reduces the waiting time of passengers by changing the train schedule. For example, in the train schedule change, since the numerical value of the weighting setting 2 is larger, an operation proposal with a smaller train schedule change amount is more likely to be output.

[0072] FIG. 19 is a diagram for explaining an example of the operation proposal constraint conditions. The operation proposal constraint conditions include the number of extra trains and the overtime hours of the crew. In condition 1, no penalty is given when the number of extra trains is up to 0, but when the number of extra trains exceeds 0, a penalty is given and an operation proposal with an extra train set is not output. On the other hand, in condition 2, no penalty is given when the number of extra trains is up to 2, but when the number of extra trains exceeds 2, a penalty is given and an operation proposal with 0 to 2 extra trains set is output. Also, from the perspective of the overtime hours of the crew, overtime is not recognized in condition 1, no penalty is given for overtime of 60 minutes or less in condition 2, and a predetermined penalty is given for overtime exceeding 60 minutes.

[0073] FIG. 20 and FIG. 21 are diagrams showing output examples of operation proposals. FIG. 20 shows Output Example 1 when Weighting Setting 1 and Condition 1 are selected, and FIG. 21 shows Output Example 2 when Weighting Setting 2 and Condition 2 are selected. The illustrated output examples may be displayed on the display device 9 of the operation proposal system 10 or output from the communication interface 4 in a predetermined data exchange format. In Output Example 1 shown in FIG. 20, since the timetable change weight is 0 according to Weighting Setting 1, the timetable change time is increased and the waiting time is decreased. In Output Example 2 shown in FIG. 21, since the timetable change weight is 1 according to Weighting Setting 2, the timetable change time is decreased, and a special train is set according to Condition 2.

[0074] As described above, according to the embodiments of the present invention, it is possible to accurately estimate the demand during a person's movement and propose an operation change plan according to the demand. Then, traffic control can be performed to optimize KPIs such as congestion dispersion so as to match the staying demand of people that changes according to the situation.

[0075] Note that the present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the scope of the appended claims. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and the present invention is not necessarily limited to those having all the configurations described. Also, a part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Further, the configuration of another embodiment may be added to the configuration of one embodiment. Also, for a part of the configuration of each embodiment, addition, deletion, or replacement with other configurations may be made.

[0076] In addition, each of the above-described configurations, functions, processing units, processing means, etc. may be realized in hardware, for example, by designing a part or all of them with an integrated circuit, or may be realized in software by a processor interpreting and executing a program for realizing each function.

[0077] Information such as programs, tables, and files that implement each function can be stored in a storage device such as a memory, hard disk, SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

[0078] In addition, the control lines and information lines show those considered necessary for explanation, and do not necessarily show all the control lines and information lines required for implementation. In practice, it is reasonable to consider that almost all components are interconnected.

Claims

1. An operation proposal system for creating operation proposals for transportation means, comprising an arithmetic unit for executing predetermined processing and a storage device accessible to the arithmetic unit, wherein the arithmetic unit uses peripheral information, passenger information, and operation information, and uses the entry station and entry time zone, which are part of the passenger information, as explanatory variables, and generates a travel demand prediction model with the final destination, which is part of the passenger information, and the probability of moving to the final destination as the objective variable, and uses the entry station, entry time zone, final destination, operation status, passenger attributes, and peripheral information, which are part of the passenger information, as explanatory variables, and has a prediction model estimation unit for generating a stay demand prediction model with the occurrence probability of stay, purpose of stay, place distribution of occurrence, and stay demand as the objective variable, The arithmetic unit receives, as inputs to the stay demand prediction model, the operation status, passenger attributes, and peripheral information, and the probability of moving to each final destination for each entry station, each entry time zone, and each final destination output by the travel demand prediction model, predicts the stay demand, which is a combination of the purpose of stay and the stay time period of passengers at transfer times for each purpose of stay, and outputs it, The arithmetic unit has an operation proposal unit that generates an operation proposal with a good operation proposal evaluation value, which is at least one of the bias in the number of passengers boarding and the waiting time for transfer of passengers, based on the predicted stay demand, The operation proposal unit applies changes to the departure times of operation flights or the addition of extra flights to the multiple-pattern operation information to create multiple operation change plans, calculates the number of passengers boarding each operation flight based on the multiple created operation change plans and the predicted stay demand, evaluates the created operation change plans and the calculated number of passengers boarding each operation flight, calculates the operation proposal evaluation value of each operation change plan with a predetermined weighting, and selects an operation change plan with a high calculated operation proposal evaluation value. An operation proposal system characterized by this.

2. The operation proposal system according to claim 1, wherein the peripheral information includes information on the route of the transportation means for which the operation proposal is created, The passenger information includes the entry station of the passenger, the final exit station, attributes, and consumption information. The operation proposal system is characterized in that the operation information includes a predetermined operation time and a delay.

3. The operation proposal system according to claim 1, wherein the stay demand prediction model uses the arrival and departure times, locations, and purposes of passengers, and the delay of the operation as explanatory variables, and the probability of the location and time at which a stay occurs to meet the purpose as the objective variable.

4. The operation proposal system according to claim 1, wherein the operation proposal unit uses the stay demand prediction model to generate an operation proposal with a good operation proposal evaluation value under the operation proposal constraint conditions, and the operation proposal constraint conditions are the maximum number of possible special trains and the overtime hours of the crew members.

5. An operation proposal creation method executed by an operation proposal system of a transportation means, the operation proposal system having an arithmetic unit that executes predetermined processing and a storage device accessible to the arithmetic unit, the operation proposal creation method comprising: the arithmetic unit generating a movement demand prediction model using peripheral information, passenger information, and operation information, with the entry station and entry time zone, which are part of the passenger information, as explanatory variables, and the final destination and the probability of moving to the final destination, which are part of the passenger information, as the objective variable; generating a stay demand prediction model using the entry station, entry time zone, final destination, operation status, attributes of the passengers, and peripheral information, which are part of the passenger information, as explanatory variables, and the occurrence probability of a stay, the purpose of the stay, the distribution of occurrence locations, and the stay demand as the objective variables; storing the generated movement demand prediction model and stay demand prediction model in the storage device; The arithmetic unit receives, as inputs to the stay demand prediction model, the operating status, passenger attributes, and surrounding information, as well as the probabilities of moving to each destination station, each arrival time zone, and each final destination output by the travel demand prediction model, predicts the stay demand, which is a combination of the purpose of stay and the stay time period of passengers at transfer points for each purpose of stay, and outputs the stay demand prediction procedure; The arithmetic unit has an operation proposal procedure for generating an operation proposal with a favorable operation proposal evaluation value, which is at least one of the bias in the number of passengers boarding and the waiting time of passengers at transfers, based on the predicted stay demand; In the operation proposal procedure: The arithmetic unit applies changes to the departure times of train operations or the addition of extra trains to the multiple train operation information to create multiple operation change plans; The arithmetic unit calculates the number of passengers boarding each train operation based on the multiple created operation change plans and the predicted stay demand; The arithmetic unit evaluates the created operation change plans and the calculated number of passengers boarding each train operation, and calculates the operation proposal evaluation value of each operation change plan with a predetermined weighting; The operation proposal creation method is characterized in that the arithmetic unit selects an operation change plan with a high calculated operation proposal evaluation value.

Images