JP2016148912A - Operation plan support apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To extract a candidate of a user to receive incentive, for facilitating periodic use of the same section, in a car sharing system.SOLUTION: An operation plan support apparatus includes: use history acquisition means of acquiring a trip group, which is a set of trips including information on a use section and use time where and when a user used a vehicle in the past; classification means of classifying the trip group into a first trip group formed of trips of the user with non-uniform use time or use section, and a second trip group formed of trips of the other users; time-zone determination means which determines a periodic-use time zone, which is a time zone in which the vehicle is used by a periodic user, on the basis of the first trip group; and vehicle arrangement determination means which determines the optimal number of vehicles to be arranged for each of stations, at the start or end time of the periodic-use time zone; and candidate extraction means which extracts a candidate of a periodic user, on the basis of the determined optimal number of vehicles to be arranged and the first trip group.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technology that supports creation of a vehicle operation plan in car sharing.

  In recent years, car sharing in which an automobile is shared by a plurality of users has become widespread. In recent years, a one-way type car sharing system that can be used one-way instead of a round trip type that returns to the station where it departs has appeared. As a result, it becomes possible to stay at the destination for a long time, and for example, it is expected to capture the demand for commuting.

In the one-way type car sharing system, there is a problem that the vehicle is biased between stations. Further, in order to maximize profit, it is required to arrange an optimal number of vehicles at each station, and when deviation occurs, it is required to correct this by forwarding the vehicles. As a technology related to this, for example, in Patent Document 1,
It describes how to make an operation plan for car sharing vehicles.

JP 2014-32531 A JP 2014-41475 A

  In order to adjust the vehicle layout between stations, it is necessary to set up personnel to route the vehicle. However, when dedicated staff is used, costs such as labor costs become a problem.

  On the other hand, there is also an idea of letting the user forward the vehicle. For example, a method of giving a preferential reservation right or giving an incentive to a user who travels on a route to be forwarded can be considered. However, since the deviation of the vehicle has a pattern similar to each day, when the user performs vehicle forwarding, it is necessary to secure a user who always travels the same route.

  In order to solve this problem, it is necessary to find a user who regularly travels in the section where the vehicle should be routed from among the members of the car sharing in order to optimize the arrangement of the vehicle between the stations. is there.

  The present invention has been made in consideration of the above problems, and an object of the present invention is to extract candidates for regular users who are users who regularly use the same section in a car sharing system.

  The operation plan support apparatus according to the present invention is an operation for extracting candidates for a regular user who is a user who gives an incentive to regularly use the same section in order to optimize the arrangement of vehicles in a car sharing system. It is a planning support device.

Specifically, when a user has used a vehicle in the past, a use history acquisition unit that acquires a trip group that is a set of trips including information on a use section and a use time, and the trip group is used as a use time or a use. Based on the first trip group consisting of trips made by users with bias in the section and the second trip group consisting of trips made by other users, and the first trip group , Time zone determination means for determining a regular use time zone that is a target time zone for regular use, and vehicle placement decision for determining the optimum number of vehicles to be arranged for each target station at the start time or end time of the regular use time zone Means for extracting a candidate for a regular user based on the determined optimum arrangement number and the first trip group. To.

The classifying means obtains a set (trip group) of histories (trips) when a user has used a vehicle in the past, and a set of trips (first trip group) performed by a user with a bias in use time or use section ) And a set of trips performed by other users (second trip group).
Trips made by users who tend to use the vehicle in a specific time zone or section are classified as the first trip group, and trips performed by other users are classified as the second trip group. The Classification may be performed by clustering or the like, for example, or other methods may be used.

  The time zone determining means is a means for determining a time zone (periodic usage time zone) in which the regular user uses the vehicle. When there are many users who use the vehicle regularly in a specific time zone, the specific time zone is set as a regular use time zone. The regular use time zone can be determined using the first trip group.

  The vehicle arrangement determining means is means for determining the optimum number of vehicles arranged for each target station at the start time or end time of the regular use time zone. Depending on the number of arrangements, the profit in a time zone other than the regular use time zone (hereinafter, non-regular use time zone) is affected. Therefore, the vehicle arrangement determination means obtains the number of vehicles that can realize the most ideal profit state for each station in the non-periodic use time zone.

  The candidate extracting means is means for extracting a user who moves the vehicle in order to realize the optimum arrangement of the vehicle determined by the vehicle arrangement determining means. Specifically, the candidate for the regular user is extracted so that the number of vehicles in the station is close to the optimal arrangement number at the time when the regular usage time zone and the non-regular usage time zone are switched.

  As described above, the operation plan support device according to the present invention determines the regular use time zone, and when the regular user performs the regular use, the regular plan is arranged so that the optimal vehicle arrangement for each station can be realized. Extract user candidates. As a result, the profit of car sharing can be improved without significantly increasing the cost.

  In addition, the operation plan support apparatus according to the present invention further includes a simulation unit that performs a simulation when the candidate extracted by the candidate extraction unit becomes a regular user and calculates a value for evaluating the business property. This may be a feature.

  When the candidate extracted by the candidate extraction means becomes a regular user, it is impossible to know exactly how the arrangement and profitability of the vehicle actually change until it is operated. . Therefore, by performing simulation, it is possible to determine whether or not the candidate has been properly extracted.

  The candidate extracting unit may extract a plurality of patterns of candidates for regular users, and the simulation unit may perform simulation for each pattern.

  In order to know how much change is caused by the setting method of the regular user, a plurality of patterns of candidates for the regular user may be generated and a simulation may be performed for each pattern. By doing in this way, a candidate for a regular user can be optimized.

  Further, the time zone determining means may determine a plurality of regular usage time zones, and the simulation means may perform simulation for each pattern.

  Similarly, in order to know how much change is caused by changing the regular use time zone, a plurality of time zone patterns may be generated and a simulation may be performed for each pattern. In this way, the regular use time zone can be optimized.

  In addition, the value for evaluating the business property may be a profit rate or a service rate. According to such a configuration, it is possible to objectively evaluate the effects caused by regular use from the viewpoint of business.

  In addition, the candidate extraction unit may include a target station at a start time or an end time of the regular use time zone when a target trip is made by a regular user among trips included in the first trip group. The target trip is extracted from the first trip group so that the number of vehicles arranged in the vehicle is close to the optimum arrangement number, and the user corresponding to the trip may be a candidate for a regular user. .

  The vehicle arrangement determining means determines the optimal number of vehicles arranged at the start time or end time of the regular use time zone. Therefore, the candidate extracting means extracts trips that can realize the number of arrangements from the first trip group at the start time or end time of the regular use time zone. Since the trip extracted in this manner is a trip that should be actively used regularly, a user who has performed the trip can be a candidate for a regular user.

  Further, the candidate extracting means may perform prioritization based on usage frequency or usage accuracy when there are a plurality of candidates in the target station.

  If the extracted candidate does not regularly use as planned, there is a risk that optimal vehicle placement will not be possible and profitability will deteriorate. That is, it is desirable that the candidate is a user who uses the vehicle more reliably. Therefore, when there are a plurality of candidates in the target station, prioritization may be performed based on the usage frequency and usage accuracy of the user. By doing in this way, the user who uses a vehicle more reliably can be extracted.

  Further, the time zone determining means classifies trips that are included in the first trip group, and trips to / from the target station by time zone, and based on the number of trips included in each time zone. One or more regular use time zones may be determined, and the vehicle arrangement determining means and the candidate extracting means may perform processing for each of the one or more regular use time zones.

The regular use time zone is preferably a time zone in which more trips included in the first trip group have arrived and departed. Therefore, trips to / from the target station can be classified according to the time zone, and the regular use time zone can be determined based on the number of trips included.
For example, while changing the width of the time zone, a predetermined number of time zones may be determined in descending order of the number of trips included, or the number of trips per unit time may be compared with a threshold value, and the threshold value continuously A plurality of unit times exceeding the above may be collected and used as a regular use time zone. Of course, you may determine a regular use time slot | zone by methods other than this.

  Further, the vehicle arrangement determining means may determine the optimal arrangement number of vehicles by solving an optimization problem formulated by a constraint condition related to movement of the vehicle and a constraint condition related to predicted demand. Good.

  For example, the optimal arrangement number of vehicles can be determined by obtaining a solution to the optimization problem by a mathematical programming solver or the like.

  The present invention can be specified as an operation plan support apparatus including at least a part of the above means. The present invention can also be specified as an operation planning method including at least a part of the above processing. The above processes and means can be freely combined and implemented as long as no technical contradiction occurs.

  ADVANTAGE OF THE INVENTION According to this invention, in the car sharing system, the candidate of the regular user who is a user who uses the same area regularly can be extracted.

It is a system configuration figure of the operation planning device concerning a first embodiment. It is a figure explaining a regular use time zone. It is an example of a usage history table. It is a process flowchart figure of the operation planning apparatus which concerns on 1st embodiment. It is a figure explaining the classification of a trip group. It is a figure explaining the method of determining the candidate of a regular utilization time slot | zone. It is a figure which simplifies and demonstrates the relationship between a station and a link. It is a figure explaining a spatiotemporal network. It is an example of the optimal arrangement pattern. It is a system block diagram of the operation planning apparatus which concerns on 2nd embodiment. It is a process flowchart figure of the operation planning apparatus which concerns on 2nd embodiment.

(First embodiment)
The operation planning apparatus according to the first embodiment will be described with reference to FIG. 1 which is a system configuration diagram. The operation planning apparatus according to the first embodiment is an apparatus that stores a car sharing usage history and extracts user candidates to be used regularly based on the usage history.

First, the vehicle bias and the method for eliminating the bias, which are the premise of the present invention, will be described.
The car sharing system in the present embodiment is a one-way type car sharing system. The one-way type is a one-way type service in which a station that receives a vehicle rental is different from a station that returns the vehicle. In such a service, there may be a case where the vehicle is biased due to demand between stations. For example, if the demand from station B to station A is greater than the demand from station A to station B, the vehicles will gather at station A over time, and service cannot be provided at station B. .

In order to solve such a problem, it is necessary to periodically forward the vehicle and restore the number of vehicles arranged at the station B. However, personnel are required for forwarding, leading to an increase in cost.
Therefore, there is a concept of having a car sharing user forward the vehicle. For example, when there is a user who regularly moves from station A to station B, the user is given some incentive (for example, discount of usage fee or preferential reservation right) and used more actively. It is a way of thinking.

  The operation planning apparatus according to the present embodiment is accumulated based on the idea that a user who moves regularly performs movement of the vehicle, thereby adjusting the arrangement of the vehicle between the stations to an optimum state. It is an apparatus for extracting a candidate for a user who is moving regularly using a usage history.

<System configuration>
The operation planning apparatus 10 according to the first embodiment includes a usage history storage unit 11, a trip classification unit 12, a target time zone determination unit 13, a vehicle placement determination unit 14, a candidate extraction unit 15, and an input / output unit 16.

  The usage history storage unit 11 is a means for storing a history (hereinafter referred to as a usage history) when a user has rented a vehicle in a table format. In the usage history table, a user ID, a departure place, an arrival place, and the like are recorded and updated when the vehicle is returned. Details of the usage history table will be described later. In this specification, each usage history recorded in the usage history table is referred to as a trip.

  The trip classification unit 12 is a means for classifying trips stored in the usage history storage unit 11 into trip groups that can be regularly used and trip groups that are not. The regular user is preferably a user who uses the same section in the same time zone every day. Therefore, the trips performed by the users who are biased in the use time or use section, that is, the users who tend to use the vehicle in a specific time zone or a specific section, and other users are stored Classify the trips that have been performed and extract candidates using the classification results. The former is referred to as a first trip group, and the latter is referred to as a second trip group.

The target time zone determination unit 13 is a means for determining a target time zone for regular use (periodic use time zone) using the trip group classified by the trip classification unit 12.
Further, the vehicle arrangement determination unit 14 is a means for determining the optimum number of vehicles to be arranged for each station (hereinafter, the optimum arrangement number) before and after the regular use time period determined by the target time period determination unit 13.
The candidate extraction unit 15 is a means for extracting candidates for regular use based on the trip group classified by the trip classification unit 12 and the optimum number of arrangements determined by the vehicle arrangement determination unit 14.
A specific method executed by each unit will be described later.

  The input / output unit 16 is a means for accepting an input operation performed by an operator of the apparatus and presenting information. Specifically, it comprises an input device such as a keyboard and a mouse and its control means, a display device such as a liquid crystal display and its control means.

  The operation planning device 10 is a computer having a CPU (Central Processing Unit), a RAM (Random Access Memory), an auxiliary storage device (for example, a flash memory, a hard disk), an input device (for example, a mouse, a touch panel), a display device, and the like. Can be configured. Each function described above may be realized by loading a program stored in the auxiliary storage device into the RAM and executing it by the CPU. In addition, some or all of the functions described above may be realized by an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), or may be processed by another computer (for example, a cloud server). . Moreover, you may implement | achieve using exclusive hardware etc.

  In the case of causing the user to perform vehicle forwarding, it is preferable to extract users who travel on the same or similar route on a daily basis. An example of such a user is a user who uses car sharing for commuting. The present invention is based on the premise that vehicle placement is optimized by giving incentives to such users and using them periodically.

FIG. 2 is a diagram for explaining the regular use time zone. The regular use time zone is a time zone in which a regular user mainly uses the vehicle, and the non-regular use time zone is a time zone in which a vehicle other than the regular user mainly uses the vehicle. The operation planning apparatus according to the present invention mainly has the following functions.
(1) A candidate for a regular use time zone is determined based on the use history. (2) A placement of a vehicle that can be optimally operated in a non-regular use time zone is determined. The users who will be able to drive the vehicle at the boundary time of the regular use time zone are extracted as candidates for the regular users. The user thus extracted is designated as the regular user. Thus, it is possible to improve the efficiency of vehicle operation in the non-periodic use time zone.

Before describing each of the functions described above, the usage history table will be described first.
FIG. 3 is an example of a usage history table stored in the usage history storage unit 11. The usage history table records user and vehicle IDs, rental and return locations (stations), rental and return times. The record to be recorded is called a trip. The usage history table is updated each time the vehicle is returned to the station.
In the present embodiment, all the stations recorded in the usage history are targeted for vehicle arrangement optimization.

  Next, a method for realizing the above-described function will be described with reference to FIG. 4 which is a flowchart of processing performed by the operation planning apparatus according to the present embodiment. The processing shown in FIG. 4 is started when the operation planning apparatus 10 acquires a processing request from an operator.

<Determination of regular use hours>
Steps S11 and S12 are steps for determining a regular use time zone. First, in step S <b> 11, the trip classification unit 12 classifies trips included in the usage history table stored in the usage history storage unit 11.

Explain trip classification.
The operation planning apparatus according to the present embodiment classifies trips recorded in the usage history table into trip groups performed by users who are candidates for regular users and trip groups performed by other users.

FIG. 5A is a diagram showing the number of trips performed by a certain user A for each use time zone. Specifically, a day is divided by unit time, and the number of times the vehicle has been used in the unit time is represented by a histogram. As can be seen from the figure, it can be seen that the user A concentrates on the morning time zone and the evening time zone and uses the vehicle.
On the other hand, FIG. 5B is a histogram similarly showing the number of trips performed by a user B different from the user A. As can be seen from the figure, it can be seen that the user B does not tend to use the vehicle in a specific time zone throughout the day.

In step S11, the trip classification unit 12 generates a histogram of the number of trips for each user, and classifies the trip group performed by a user whose usage time is biased and the trip group performed by a user whose usage time is biased. . Trip classification can be performed, for example, by clustering or the like, but any method may be used as long as trips can be classified based on a deviation in use time in a day.

Here, although an example has been given in which trip classification is performed based on the deviation of use time, similarly, trips can be further classified based on deviation of use sections. The biased use section means that the combination of the departure station and the arrival station in a plurality of trips is more similar.
In the present embodiment, a set of trips performed by a user who has a combination of both and usage time and usage section are defined as the first trip group, and a set of trips performed by other users is the second trip. A group.
For this reason, for example, the result of the classification based on the use time and the result of the classification based on the use section may be logically calculated.

  Next, in step S12, the target time zone determination unit 13 generates a candidate for a regular use time zone using the first trip group. In this step, referring to trips included in the first trip group, a time zone in which the use is most concentrated is extracted. By performing such processing, a time zone suitable for regular use can be determined.

A method for generating a candidate for a regular use time zone will be described.
FIG. 6 is a diagram showing the usage time corresponding to trips included in the first trip group in a tabular format. Here, it is assumed that the time zone is divided in units of 5 minutes, and the rectangle represents the usage time (T1 to T6 are trips).
The target time zone determination unit 13 sums the number of trips per unit time and determines a time zone that exceeds a predetermined threshold. For example, in the example of FIG. 6, if the threshold is 3, a time period from 8:10 to 8:45 can be obtained. The time zone acquired in this way becomes a candidate for the regular use time zone.
The threshold value may be set by the operator of the apparatus, or may be set automatically by the apparatus based on past results.

In addition, when there are a plurality of candidates for the regular use time period per day, a set of a plurality of time periods may be used. For example, when two of the morning time zone and the evening time zone are extracted, the two time zones may be used as the regular use time zone. In the present embodiment, description will be made assuming that there is a regular use time zone twice a day.
Further, when the candidates for the regular use time zone are close to each other (for example, when two or more time zones are extracted within 30 minutes), they may be combined.
In addition to this, while changing the width and position of the time zone, the number of trips included in the time zone is acquired, and a time zone in which the number of trips is a predetermined number or more can be used as a candidate for a regular use time zone. Good.

Steps S13 and S14 are executed for each candidate for the regular use time period determined in step S12.
In step S13, the vehicle arrangement determination unit 14 determines the optimal arrangement of the vehicle in the target time zone. The process performed in this step will be described in detail.
The vehicle placement determination unit 14
(1) The number of parking frames per station (2) Information about the demand for movement between stations is held in advance, and processing is performed based on this information.
The number of parking frames for each station is information representing the maximum number of sharing vehicles that can be parked for each station. The movement demand between stations is information obtained based on the past use history, and is information indicating from which station to which station there is a movement demand in which time zone. The movement demand may be, for example, information obtained by classifying the number of movements between stations for each time zone.

Next, a method for determining the optimal arrangement of the vehicle will be described in detail.
Here, in order to simplify the description, an example in which there are two routes connecting the stations A, B, and C as shown in FIG. 7 will be described. Hereinafter, a path connecting the stations is referred to as a link.

In the present embodiment, a model representing a spatio-temporal network is constructed, and the optimal number of arrangement is calculated using the model. FIG. 8 is a diagram showing a model constructed for the network of FIG. The horizontal axis is a time axis, and in this example, the state of the station and the link at regular intervals is expressed. For example, when time 0 is 0: 0, time 1 can be 0: 5 and time 2 can be 0:10. Each engraved time is referred to as a time step. The step size of the time step is 5 minutes in this example, but may be set in any manner.
In the present embodiment, the first time step is the time when the regular time zone shifts to the non-regular time zone (that is, the time when the morning regular use time zone ends), and the last time step is regular from the non-periodic time zone. The time to shift to the time zone (that is, the time when the regular use time zone in the evening starts).

Stations and links for each time step, that is, circles in the figure are referred to as nodes in the following description. The number of vehicles existing in the node can be expressed by a variable.
Moreover, the relationship between variables can be expressed by a mathematical expression. For example, the total number of vehicles present at station A at time 0 is the number of vehicles that remain at station A until the next time step, that is, the number of vehicles that transition to time 1 via reference numeral 103; This is the total of the number of vehicles that have flowed out to the link AB, that is, the number of vehicles that transition to the time 1 via the reference numeral 104. Similarly, the number of vehicles present at station A (node 105) at time 1 is the number of vehicles present at station A at time 0 and staying at station A until the next time step. This is the total number of vehicles that exist on the link AB and have entered the station A until time 1.

Thus, by representing the number of vehicles existing at each node by the number of vehicles that have flowed out or entered from each node, all the models shown in FIG. 8 can be represented by mathematical expressions. The necessary variables are the following three types.
(1) Number of vehicles existing at a certain station or link at a certain time (2) From one time to the next time, the station or link was leaked and headed to another destination station Number of vehicles (3) Number of vehicles that flowed out of a station from one time to the next time and flowed into other stations or links. Since they flow into the link, the number of vehicles existing at each station and link at each time can be expressed by a mathematical formula by combining them.

Next, in order to express when the vehicle starts to move, information about the usage demand for car sharing is added.
Here, a vehicle that departs from each station is defined based on the use demand (for example, information that “there is a demand from station A to B at eight o'clock)”. The usage demand may be specifically defined for each time step, but it may be assumed that the user arrives based on a predetermined rule. For example, the user may arrive according to the Poisson distribution, or may arrive at random times. Here, one specific user arrival time pattern is generated from the use demand and used.

The vehicle arrangement determination unit 14 further adds a constraint condition (such as the maximum number of parked vehicles) to the mathematical model defined as described above, and determines a solution that satisfies the optimal solution condition by mathematical programming.
The optimum solution condition is for maximizing or minimizing a specific value, and may be anything as long as the specific value can be expressed by a mathematical expression. For example, “maximize the number of vehicle rentals (service rate) in non-scheduled time zone”, “maximize sales”, “maximize profit rate (profit amount)”, etc.
By solving the optimization problem based on multiple mathematical formulas described above using mathematical programming, each station is informed at the time of transition from the regular time zone to the non-regular time zone and at the time of transition from the non-regular time zone to the regular time zone. The number of vehicles to be arranged (hereinafter, optimum arrangement pattern) can be determined. FIG. 9 is an example of the optimum arrangement pattern obtained in this way.
In this example, the optimum arrangement number is obtained at the timing when the morning regular use time zone ends and at the timing when the evening regular use time zone starts, but the timing for obtaining the optimum arrangement number may be other than this. For example, the timing at which the morning regular usage time period starts and the timing at which the evening regular usage time period ends may be added to the target.

  In this example, one arrival time pattern of the user is generated from the use demand, but a plurality of patterns may be generated. In this case, optimization can be performed for each of a plurality of patterns, and the number of vehicles to be arranged at each station can be determined based on the obtained results. For example, with respect to the number of vehicles corresponding to each station, a mode value or an average value obtained for each pattern may be selected to obtain the final number of vehicles arranged. Such a method is called SamplePath optimization.

Next, in step S14, the candidate extraction unit 15 extracts candidates for regular users for realizing the determined optimal arrangement pattern.
First, from the first trip group, users who have used at least once in the regular use time zone are extracted for each station. For example, when adjusting the number of stations A when the regular use time period ends, a user who has a history of arrival at the station A in the regular use time period is extracted. Further, when adjusting the number of stations A when the regular use time period starts, users having a history of departure from the station A are extracted in the regular use time period.

  Note that the station serving as a key when performing extraction is not necessarily the same, and may be a nearby station, for example. For example, when a user who regularly uses station B in the vicinity of station A is expected to change to station A by obtaining an incentive, extraction may be performed for station B.

Similarly, the extraction target does not necessarily have to be a user who is used in the regular use time zone. For example, consider a case where the regular use time period ends at 9 am and there is a user who arrives at the target station at 9:30 am. In this case, when the movement of the user is expected to be 30 minutes earlier by obtaining an incentive, the user may be the extraction target.
As described above, when it is expected that the station used by the user and the use time are changed by giving the incentive, the user may be extracted in consideration of this.

Next, a priority is given to the extracted user, and a user who is a candidate for a regular user is specified. For example, the priority may be higher as the number of uses is higher, or may be higher as the use accuracy is higher. Then, a predetermined number of users are extracted from the top.
For example, when the number of vehicles required for station A is five at time A when the regular use time period ends, a maximum of five users are extracted. In addition, it is preferable to change the number of extraction of a user according to a track record. For example, when there is no regular use and when it is known from past data that there are about two vehicles at station A at time A, it is possible to extract up to three users. On the other hand, if it is known from past data that there are about five vehicles at station A at time A, the user's extraction for station A may be omitted.

  Such processing is performed both at the time when the morning regular use time zone ends and at the time when the evening regular use time zone starts. Thereby, it will be in the state by which the user and the time slot | zone which the said user uses regularly are specified.

  In addition, it is preferable to give a big value with respect to the user who will use a vehicle more reliably as a priority. In addition, when there are a plurality of times during the day when the optimum arrangement number is obtained, it is preferable to give a large value to a user who can use both.

  When the processes of steps S11 to S14 are completed, the result is notified to the operator of the apparatus through the input / output unit 16, and the process is terminated. In addition, when there are a plurality of candidates for the regular use time zone, the processes of steps S13 to S14 are executed a plurality of times. In this case, a plurality of results are presented to the operator for each time period.

  As described above, the operation planning apparatus according to the present embodiment classifies the trip group included in the usage history, and based on the classification result, determines the regular use time zone and the candidate for regular use in the time zone. Extract. Thereby, vehicle operation can be optimized without incurring costs.

(Second embodiment)
In the first embodiment, a user who is a candidate for a regular user can be extracted, but how the vehicle arrangement changes when the extracted user actually performs regular use, or a business I couldn't get information about how it affects sex.
On the other hand, the second embodiment is an embodiment in which a simulation of business property is performed on the assumption that a user who is a candidate for the extracted regular user has performed regular use.

  As shown in FIG. 10, the operation planning apparatus 10 according to the second embodiment is further provided with a simulation unit 17 that performs simulation based on the result obtained in the first embodiment. This is different from the first embodiment. Since other means are the same as those in the first embodiment, detailed description thereof is omitted.

  FIG. 11 is a process flowchart of the operation planning apparatus 10 according to the second embodiment. In the second embodiment, after determining a candidate for a regular user in step S14, in step S15, the simulation unit 17 performs a simulation of profits when the candidate performs regular use. Thereby, a service rate and a profit rate can be calculated | required.

In the second embodiment, in step S14, a plurality of regular user candidates are generated, and a simulation is performed for each of the plurality of patterns. For example, when the number of vehicles required for station A is five at time A when the regular use time period ends, five patterns are generated by changing the number of users to be extracted from 1 to 5 people, Each may be simulated.
Moreover, when there are a plurality of candidates for the regular use time period, the simulation is executed for each candidate. That is, the simulation is executed as many times as the number of patterns of the candidate for the regular user and the number of patterns of the candidate for the regular use time period.

  In addition, after the simulation of all patterns is completed, a comprehensive determination on business performance is performed in step S16. For example, when there is a pattern that satisfies a predetermined condition, the pattern and information about the result are presented to the operator of the apparatus. The predetermined condition is, for example, a condition that defines a service rate or profit rate to be maintained. As a result, it is possible to quantify the pattern with the best conditions and the effects when regular use is introduced in the pattern.

  The simulation may be performed based on the usage history, or when there is data for predicting future usage (for example, reservation data or questionnaire results), the simulation may be performed using the data. .

(Modification)
The above embodiment is merely an example, and the present invention can be implemented with appropriate modifications within a range not departing from the gist thereof.

  For example, in the description of the embodiment, the regular use time zone is set to twice a day, but the regular use time zone may be any number of times per day. Further, the processing target date may be classified according to the day of the week, and the processing may be performed separately on the day suitable for regular use and the day unsuitable. For example, for weekends and holidays, since there is little demand for commuting, processing may be performed excluding the day.

The regular use time zone may be other than the time zone other than morning and evening as illustrated. For example, a pattern of regular use only during the daytime can be considered. Further, the regular use time zone does not necessarily have to be a divided time zone in one day. For example, a plurality of days may be included, such as from 0:00 on Monday to 24:00 on Friday, and from 0:00 on Saturday to 24:00 on Sunday.
For this reason, in step S11, trip classification may be performed in large units such as one week.

Further, in the description of the embodiment, the optimum arrangement is obtained for each of the time when the period shifts from the regular use time period to the non-periodic use period and the time when the period shifts from the non-regular use period to the regular use period. Only one of them may be obtained.
In the description of the embodiment, the processing is performed for all the stations recorded in the use history, but the target station may be selectable.

  In the description of the embodiment, when classifying the first trip group and the second trip group, the processing is performed based on the bias of both the use time and the use section, but based on other information. Trip classification may be performed. For example, when information about the purpose (for example, commuting, attending school, leisure, etc.) for the user to use the vehicle can be acquired, classification may be performed based on the information.

  In the description of the embodiment, the regular use time zone is assumed to be the same for all stations. However, the regular use time zone may be different for each pair of stations (a combination of a departure station and an arrival station). Good.

DESCRIPTION OF SYMBOLS 10 Operation planning apparatus 11 Usage history memory | storage part 12 Trip classification | category part 13 Target time zone determination part 14 Vehicle arrangement | positioning determination part 15 Candidate extraction part 16 Input / output part

Claims (10)

In a car sharing system, an operation plan support device that extracts candidates for a regular user who is a user who gives an incentive to regularly use the same section,
Usage history acquisition means for acquiring a trip group that is a set of trips including information on usage sections and usage times when a user has used a vehicle in the past;
Classifying means for classifying the trip group into a first trip group consisting of trips performed by a user whose use time or usage section is biased and a second trip group consisting of trips performed by other users;
Based on the first trip group, a time zone determining means for determining a regular use time zone that is a target time zone for regular use;
Vehicle arrangement determining means for determining the optimal number of vehicles arranged for each target station at the start time or end time of the regular use time period;
Based on the determined optimum arrangement number and the first trip group, candidate extraction means for extracting candidates for regular users,
An operation plan support apparatus having The candidate extracted by the candidate extraction means further performs simulation when it becomes a regular user, and further includes simulation means for calculating a value for evaluating business feasibility,
The operation plan support apparatus according to claim 1. The candidate extraction means extracts a plurality of patterns of periodic user candidates,
The simulation means performs a simulation for each pattern.
The operation plan support apparatus according to claim 2. The time zone determining means determines a plurality of regular usage time zones,
The simulation means performs a simulation for each pattern.
The operation plan support apparatus according to claim 2 or 3. The value for evaluating the business property is a profit rate or a service rate.
The operation plan support apparatus according to any one of claims 2 to 4. The candidate extracting means includes a vehicle at a target station at a start time or an end time of the regular use time zone when a target trip is made by a regular user among trips included in the first trip group. The target trip is extracted from the first trip group so that the number of arranged is close to the optimum number of arranged, and the user corresponding to the trip is a candidate for a regular user.
The operation plan support apparatus according to claim 1. The candidate extracting means performs prioritization based on use frequency or use accuracy when there are a plurality of candidates in the target station.
The operation plan support apparatus according to claim 1. The time zone determining means classifies trips included in the first trip group, and trips to / from the target station according to time zones, and one trip based on the number of trips included in each time zone. The above-described regular use time zone is determined, and the vehicle arrangement determination means and the candidate extraction means perform processing for each of the one or more regular use time zones,
The operation plan support apparatus according to claim 1. The vehicle arrangement determining means determines an optimal arrangement number of vehicles by solving a constraint condition regarding the movement of the vehicle and an optimization problem formulated by a constraint condition regarding the predicted demand.
The operation plan support apparatus according to claim 1. In a car sharing system, a method performed by an operation plan support apparatus that extracts candidates for periodic users who are users who give incentives to use the same section periodically,
A usage history acquisition step of acquiring a trip group that is a set of trips including information on usage sections and usage times when the user has used the vehicle in the past,
A classification step for classifying the trip group into a first trip group consisting of trips performed by a user whose use time or usage section is biased and a second trip group consisting of trips performed by other users;
Based on the first trip group, a time zone determination step for determining a regular use time zone that is a target time zone for regular use;
Vehicle placement determination step for determining the optimal number of vehicles to be arranged for each target station at the start time or end time of the regular use time zone;
Based on the determined optimum arrangement number and the first trip group, a candidate extraction step for extracting candidates for regular users,
Including methods.

Images