EP3100216A1

EP3100216A1 - Method and system for rebalancing a facility for shared use of vehicles, and facility implementing such a method and/or system

Info

Publication number: EP3100216A1
Application number: EP15703477.8A
Authority: EP
Inventors: Raphaël BARROIS; Olivier BOUVET; Thomas CHAUMENY; Xavier DUTREILH; Sylvain GERON; Clément LAMBRINOS
Original assignee: Bluecarsharing SAS
Current assignee: Bluecarsharing SAS
Priority date: 2014-01-31
Filing date: 2015-01-22
Publication date: 2016-12-07
Also published as: WO2015113882A1; JP2017504923A; US20160343068A1; SG11201605751RA; CN106030628A; CA2937414A1; KR20160113675A; FR3017229A1; US10417585B2; JP6514235B2; HK1226183A1

Abstract

The invention concerns a method (200) for rebalancing vehicle pickup/return stations, said method (200) comprising the following steps: - determining (204), for each station : • a number of vehicles available in said station, • an average waiting time in the station, and - filtering (206) the stations by said parameters, said filtering (206) providing: • a first list of so-called departure stations, and • a second list of so-called arrival stations, - designing (208) at least one operation for moving a vehicle depending on said lists, and - carrying out (214) at least one operation for moving a vehicle.

Description

'Method and system for rebalancing a shared-use vehicle installation, installation implementing

The present invention relates to a method of rebalancing a shared use installation of vehicles comprising a plurality of stations geographically distant from each other. It also relates to a system implementing such a method and an installation implementing such a method and / or such a system.

The field of the invention is the field of shared use of a fleet of vehicles, including a fleet of electric vehicles, on a plurality of stations allowing users to take or deposit such vehicles, for example in the framework of the automated rental of vehicles.

State of the art

Automated vehicle leasing is a growing field. Agglomerations wishing to reduce the number of vehicles on their territory are setting up automated vehicle rental facilities.

The automated rental facilities, and more generally the facilities for shared use, of a fleet of vehicles comprise several stations, each enabling the users to start the use of a vehicle, that is to say of take possession of a vehicle, and deposit a vehicle after use.

These installations require operations to rebalance the stations. Indeed, it is very rare that a user taking possession of a vehicle at a given station, deposits the same vehicle at the same station after using the vehicle. In addition, flows of users occur naturally, for example in the morning from a residential area to an economic zone and vice versa in the evening. Therefore, it is common to have "empty" stations where no vehicle is available while other stations are overloaded with vehicles, the distribution of vehicles in the stations being in fact not in conformity with the needs of the users.

Such a situation brings several drawbacks, including a very long waiting time for users going to "empty" stations, or for those wishing to drop a vehicle in an overloaded station, and degraded access to the shared use service. cars.

In an attempt to overcome these disadvantages, operators have been deployed to rebalance the stations. These operators move vehicles on overcrowded stations to empty stations.

However, such rebalancing is not effective in reducing waiting time, and increasing access to the vehicle usage service. In addition, such a rebalancing, although relatively inefficient, is very expensive in terms of labor and is very time consuming.

An object of the present invention is to overcome the aforementioned drawbacks.

Another object of the invention is to propose a method and a system for rebalancing vehicles in a facility for shared use of vehicles that are more efficient than the existing methods and systems in terms of reducing waiting time and access to the vehicle. service.

Another object of the invention is to provide a method and system for rebalancing vehicles in a facility for shared use of vehicles less expensive than existing methods and systems.

Finally, another object of the invention is to propose a method and system for rebalancing vehicles in a shared vehicle utilization facility that is more flexible and more responsive to changes occurring in the installation.

Presentation of the invention

The invention proposes to achieve at least one of the aforementioned objects by a rebalancing process of stations for taking / removing vehicles proposed for shared use, said method comprising the following steps: detecting, for each station, each departure and each arrival of a vehicle in said station and storing data relating to said station in association with a given hourly data,

- determination for each station:

■ a number of vehicles available in said station,

^■ a waiting time of each vehicle in said station,

determination and storage of an average waiting time for each station as a function of the data stored in association with a given time slot, and

filtering the stations according to the number of available vehicles and the average waiting time, said filtering providing:

^■ a first list, called departure stations, and ■ a second list, called arrival stations;

designing at least one vehicle moving operation between a station of the first list and a station of the second list, and

- Performing at least one operation of moving a vehicle.

The invention therefore makes it possible to carry out a rebalancing taking into account, not only the number of vehicles in a station, but also other criteria representative on the one hand of the real activity of each station of the installation and on the other hand part of the desired activity on each station of the installation. Thus, the rebalancing performed according to the invention is more flexible to adapt to the actual activity of the installation at each station. In addition, since the actual activity may change over time, the rebalancing according to the invention takes these changes into account, individually at the level of each station, and is therefore more reactive.

Furthermore, taking into account the criteria representative of the actual activity of each station and the desired activity at each station, the method according to the invention makes it possible, for example, not to feed a station in which it there are not many activities, even if the number of vehicles on this station is low. Similarly, the method according to the invention makes it possible for example not to unload a station in which there are many requests, even if there are many vehicles on this station compared to a standard situation. Consequently, the method according to the invention makes it possible to avoid making unnecessary displacements of vehicles, which, on the one hand, reduces the costs of rebalancing, and on the other hand, makes it possible to have more operators for other stations where the rebalancing can be conducted faster and more responsively reducing the average waiting time.

The waiting time may in particular correspond to a time between a removal of the vehicle on said station and a vehicle taken on said station following said removal. Thus, for a station, when the average waiting time is long, it means that, for this station, the average demand, which corresponds to the inverse of the waiting time, is low. On the contrary, when the average waiting time is low it means that the average demand is large. It is possible to take into account the variation of the waiting time, and consequently of the demand, according to the different time slots of a day or a week. For example, for a station, demand can be high at the time of arrival and departure of work and very low in the day. The method according to the invention advantageously makes it possible to take these variations into account.

The durations between a vehicle engagement operation and a vehicle removal operation can be performed by detecting for each vehicle:

- the time, and possibly the day and the month, of removal of the vehicle;

- the time, and possibly the day and the month, of taking the vehicle.

This data can be detected, for example, by a presence sensor arranged at a parking space and / or, in the case of electric vehicles, by a sensor arranged in the parking cable. charging the electric vehicle. The detected data can be transmitted to a remote server, connected to each station, and comprising calculation means for determining the average waiting time.

Each stored waiting time can thus be associated with at least one criterion such as the type of day, the waiting times used to calculate the average waiting time being only the times associated with this criterion.

According to an advantageous embodiment, the time slot can be a sliding time slot, of predetermined amplitude, and calculated according to the current time. More specifically, the time slot can be defined by adding a predetermined number of hours (s) to the time. Thus, it is possible to take into account future variations in terms of waiting times or requests and to anticipate the rebalancing necessary to cope with future demand.

At least one of the filtering or design steps is performed taking into account for each station at least one datum relating to at least one station adjacent to said station, the one or one of the data being a number of vehicles in a together, said area, including said station and the stations closest thereto.

The process thus takes into account the stations not in isolation but also the impact of the state of each station on the neighboring stations, for a more precise balancing. According to the invention, a zone associated with a station can be defined as comprising:

all the stations located at a distance and / or at a predetermined travel time as a function of the station with respect to which it is determined: in this case the number of stations in the zones is variable and two defined zones associated with two stations may include a different number of stations; or

a predetermined number of stations closest to the station with respect to which it is determined: in this case two defined areas associated with two stations include the same number of stations.

The filtering may in particular comprise the following steps performed for at least one, preferably each, station:

comparing the total number of vehicles in the station or the zone associated with said station with a first threshold;

when said total number is greater than said first threshold, adding said station to a list, said list of stations to be unloaded;

comparing said total number with a second threshold, lower than said first threshold;

- When said total number is less than said second threshold, adding said station to a list, said list of stations to feed.

Such a filtering makes it possible to determine the stations for which a rebalancing operation can be envisaged, by eliminating the stations which are sufficiently provided with vehicles according to the number of vehicles in the station or the zone. In addition, this operation makes it possible to simply perform an effective first selection, using simple calculations and not being too difficult to implement.

The method according to the invention may advantageously comprise a determination, for at least one station, in particular each station of the list of stations to be discharged, of a parameter, called charge rate, depending on:

- the number of vehicles available in the station, and

average waiting time for a given time period in said station;

the filtering step being performed by taking into account the value of said charge rate.

Thus, the method according to the invention makes it possible to apply an order to the stations to be unloaded as a function of the number of vehicles in the station and the average waiting time. Thus, it is possible to discharge in priority the stations that have the most vehicles available compared to the average waiting time, ie the stations that are the most loaded compared to the demand in these stations.

According to a particular and nonlimiting limiting embodiment, the charge rate, noted TDQ, for the station / ^' can be determined according to the following relation:

TDC _t = TAM _i x NVS _t

with:

- TAM, the average waiting time in the station / ^' , and

- NVIf the number of vehicles available in the station / ^' .

In addition, the filtering step may comprise a step of selecting start stations, in particular from the list of stations to be discharged as a function of the charge rate, said selection step comprising the following steps:

determining a number of operators available to move a vehicle;

for each available operator, and by successive iterations:

^■ determination of the station, especially in said list of stations to be discharged, which is associated with the largest charge rate;

^■ adding said station to the list of departure stations;

^■ updating the load rate associated with said selected station by subtracting a vehicle from the number of vehicles available in said station.

Thus, when there are fewer operators available than stations to be discharged, the method according to the invention makes it possible to discharge the most heavily loaded stations in priority. The station or stations thus selected are placed in the list, or cluster, called "departure stations".

In addition, when a station is selected as a departure station and added to the list of departure stations, the charge rate associated with that station is updated (as if an operator had unloaded a station). vehicle from this station) before performing a new iteration of the selection steps. Thus, when a station is very busy, it may be selected again for a new rebalancing operation as a start station.

The method according to the invention may further comprise a determination of a parameter, said flow, as a function of said average waiting time and a mean time, said deposition, the filtering step comprising a selection of least one start station according to said flow, said selection comprising at least one iteration of the following steps:

comparing the flow associated with a station to a first threshold;

when said stream is less than said first threshold, adding said station to the list of departure stations. The average time of removal is the average time between taking a vehicle and the subsequent removal of a vehicle in the station.

The flow corresponds in particular to the inverse of the waiting time at which we subtract the inverse of the average time of removal, the first threshold being generally very negative, which means that the removal in the station is very strong compared to the taking of vehicle.

If necessary, the step of selecting using the flow is performed before the step of determining the number of operators available to determine the number of elements of the start list, the number of operators taking into account the number of stations whose flow is below this threshold. It corresponds to the actual number of available operators from which we subtract the number of stations having fulfilled the previous flow criterion.

We can also and / or alternatively compare the flow with another range of values, in particular a range of values around 0 for the flow as defined above, and when the flow is in the range of values, exclude the station from the station list to unload, because it is then considered as self-regulating. Such filtering using the flow parameter makes it possible to determine the stations for which a rebalancing operation can be envisaged, taking into account situations that are not necessarily treated by the aforementioned operations as a function of the average waiting time. and an average time of removal of vehicles in the station.

Each of the thresholds used in the filtering step may be determined by a statistical analysis of past time slots, and may be updated based on field observations.

According to an exemplary and non-limiting particular embodiment, the flow, denoted F ,, for a station / ^'can be determined according to the following relationship:

F, = -!

TAM _t TEAM _;

with:

- TAMj the average waiting time in the station / ^' , and

- TEAM, the average time between two consecutive arrivals of vehicles on the station ^/.

Furthermore, the method according to the invention may advantageously comprise a determination, for at least one station, in particular each station of the list of stations to be supplied, of a parameter, called discharge rate, depending on:

- the number of vehicles available in the station, and

average waiting time for a given time period in said station;

the filtering step being performed taking into account the value of said discharge rate.

Thus, the method according to the invention makes it possible to apply an order to the stations to feed according to the number of vehicles in the station and the average waiting time. Thus, it is possible to prioritize the stations that have the least available vehicles compared to the average waiting time, that is to say the stations that are the most unloaded compared to the demand in these stations . According to a particular and nonlimiting limiting embodiment, the discharge rate, denoted TDD ,, for the station / ^' , can be determined according to the following relation:

TDD _; = TAM _; x (NVS _j + 1)

With:

- TAMj the average waiting time in the station / ^' , and

- NVSj the number of vehicles available in the station / ^' .

In addition, the filtering step may advantageously comprise a step of selecting arrival stations, in particular in the list of stations to be powered according to the discharge rate, said selection comprising the following steps:

determining a number of operators available to move a vehicle;

for each available operator, and by successive iterations:

^■ determination of the station, in particular in said list of stations to feed, which is associated with the smallest discharge rate;

■ adding said station to the list of arrival stations;

^■ updating the discharge rate associated with said selected station by adding a vehicle to the number of vehicles available in said station.

Thus, when there are fewer operators available than stations to be loaded, the method according to the invention makes it possible to load the most unloaded stations in priority. The station or stations thus selected are placed in a list, or cluster, called "arrival stations".

In addition, when a station is selected as the arrival station and added to the list of arrival stations, the discharge rate associated with that station is updated (as if an operator had powered the station with a vehicle) before performing a new iteration of the selection steps. Thus, when a station is very discharged, it is possible that it is selected again for a new rebalancing operation as the arrival station. The method according to the invention, and in particular the design step, can also advantageously comprise a step of determining a balancing time, performed before the moving step, comprising the following operations, performed by successive iterations:

for each station in the list of departure stations, determination of a parameter, said start, based on at least the load ratio;

for each station in the list of arrival stations, determining a parameter, said arrival, based on at least the discharge rate;

calculating the balancing time between each of the stations of the list of departure stations and each of the stations of the list of arrival stations, according to said departure and arrival parameters and a travel time between said stations departure and arrival stations; and

- Choice of displacements according to balancing times respectively calculated.

It is thus possible to take into account, for each operation of vehicle movements, on the one hand the possibility that a user goes to a station close to that concerned and secondly the time that would take the movement of the vehicle from the departure station to the arrival station. The departure parameter can also be calculated based on a probability of transfer to at least one closer station, in particular to all stations in the zone associated with said station.

The arrival parameter can also be calculated according to a probability of transfer to at least one closer station, in particular to all the stations of the zone associated with said station. According to a particular embodiment and not limiting, the starting parameter for the station / ^' , denoted PDD can be determined according to the following relation:

PDD, ¹

(TAM _i x NVS _i ) + px (Σ V _k × TAM _k × NVZ _k )

with

p the probability of transfer to at least one nearest station,

- k is the number of stations in the area associated with the station / ^'

- NVZ _k the number of vehicles in the zone associated with station k, and

- V _k = l if the station k is empty and V _k = 0 otherwise.

According to a particular and nonlimiting limiting embodiment, the arrival parameter for the station / ^' , denoted PDA, can be determined according to the following relation:

1

PDA

(TAM _i x (NVSi + V) + px (Σ ^V k ^{x TAM} k ^x ( ^NVZ k +!))

with

p the probability of report to at least one nearest station, k the number of stations in the area associated with the station / ^',

NVZ _k the number of vehicles in the zone associated with station k, and V _k = l if station k is empty and V _k = 0 otherwise.

According to a particular embodiment and in no way limiting, the balancing time between a start station / ^' and an arrival station j, noted TDEj, can be determined according to the following relation:

TDE _{i → j} = PDD _t - (TDP _{i → j} x PDA _j )

with:

- PDD the start parameter of the start station / ^' ,

PDA _j , the arrival parameter of the arrival station j, and

- TDPj, the journey time from the station / ^' to the station j.

Thus, when the balancing time is very low, it means that the travel time and / or the arrival parameter is large compared to the starting parameter. Therefore, this means that the vehicle will in principle be caught in the arrival station before it is in the departure station.

In the opposite case, this means that the travel time and / or the arrival parameter is very small compared to the departure parameter. This means that it is very likely that the vehicle will be taken from the departure station before it is in the arrival station.

The design step may further take into account the balancing time by at least one iteration of the following steps:

determination of the maximum balancing time among the calculated balancing times,

determination of the departure and arrival stations with which said maximum balancing time is associated and selection of these stations in association,

- exclusion of said stations from the list of departure and arrival,

- update the number of vehicles in the departure and arrival stations as if the trip had already been made. Thus, the method according to the invention makes it possible, in priority, to carry out balancing operations for which the balancing time is the minimum.

Advantageously, the design step may comprise a step of selecting an operator to perform a vehicle movement, said selection comprising the following steps:

- determination of the geographical position of each available operator;

according to said geographical position, determination of the operator closest to the departure station concerned by said displacement;

sending a message to said nearest operator to signal the task to be performed;

- update the operator status as unavailable. Thus, the method according to the invention makes it possible to select the operator closest to the departure station concerned by the vehicle moving operation. Consequently, the method according to the invention makes it possible to minimize the travel time of an available operator to the departure station.

According to another aspect of the invention, there is provided a rebalancing system for vehicles for taking / removing vehicles proposed for shared use, said system comprising:

means for detecting each departure and each arrival of a vehicle in the station, for example presence sensors at each of the station's parking spaces, such as cameras or sensors inserted in the vehicle charging cable in the station, the case of electric vehicles;

means for determining, for each station, the following parameters:

^■ a number of available vehicles,

^■ a waiting time of the vehicle in the station, such as a sensor or sensors detecting the time, or the date of arrival of each vehicle and the time, or the date of departure of a vehicle;

means for storing data relating to the waiting times in association with a given hourly data;

means for determining an average waiting time for each station as a function of the data stored in association with a given time slot, comprising for example a calculation means receiving from each station these data and configured to calculate the average waiting time ;

means for storing the determined average waiting time;

means for filtering the stations to be balanced according to said parameters, said filtering providing:

^■ a first list, called departure stations, and

^■ a second list, called arrival stations; and means for designing at least one vehicle moving operation between a station of the first list and a station of the second list. The filtering means and / or the design means may comprise:

one or more configured / programmed calculation means for calculating desired data, such as those described above, by means of previously indicated relations, such as those specified above;

means, such as geolocation means, carried by the operators in the field to determine and inform the position of each operator. Advantageously, some or all of the calculation means can be centralized in a central site connected to each station and to each operator.

According to another aspect of the invention, there is provided a shared use facility for vehicles, in particular an automated vehicle rental facility, said installation comprising:

a plurality of vehicle pick-up stations; and

means for implementing the steps of the rebalancing method according to the invention, or a rebalancing system according to the invention.

Other advantages and characteristics will appear on examining the detailed description of non-limiting embodiments, and the appended drawings in which:

- FIGURE 1 is a schematic representation of a vehicle rental installation implementing the method according to the invention; and FIGURES 2-7 are a diagrammatic representation in the form of a diagram of an exemplary embodiment of a method according to the invention. It is understood that the embodiments that will be described in the following are in no way limiting. In particular, it will be possible to imagine variants of the invention comprising only a selection of characteristics described subsequently isolated from the other characteristics described, if this selection of characteristics is sufficient to confer a technical advantage or to differentiate the invention from the state of the prior art. This selection comprises at least one feature preferably functional without structural details, or with only a part of the structural details if it is this part that is only sufficient to confer a technical advantage or to differentiate the invention from the state of the prior art.

In particular, all the variants and embodiments described are combinable with each other if nothing stands in the way of this combination at the technical level. In the figures and in the following description, the elements common to several figures retain the same reference.

The examples described below relate to an automated rental of electric cars on several rental sites.

FIG. 1 is a schematic representation of a shared-use installation of vehicles, such as, for example, in the context of an automated rental of electric vehicles, implementing the method according to the invention.

The installation 100 shown in FIG. 1 comprises a central site

102, several stations 104i-104 _n allowing users to take or deposit vehicles. The central site 102 is connected to each of the stations through a wireless communication network 106, for example GPRS, or a wired network, for example of the DSL or LAN type. Of Preferably, each station 104 is connected to the central site via the two networks of terminals, which allows a continuous connection even if one of the networks is faulty.

Each station 104 comprises a management terminal 110 for the management of the station, such as for example a rental terminal of a vehicle, and a plurality of charging terminals 112-116, each charging terminal being provided for charging a station. vehicle with an electric battery at a parking location, namely parking spaces 118-122.

Some stations 104 also include a subscription terminal 108 for registering a new subscriber.

Each of the station 118-122 of a station 104 includes a presence detector module 124-128, namely weighing means, a camera and / or an electrical connection detector with a vehicle, connected to the terminal of management 110 of the station 104, to detect the presence or absence of a vehicle at a parking space 118-122, and in fact the departure of a vehicle from the station 104, and the arrival of a vehicle at the station 104. Management terminal 110 is programmed to record the times, and dates, of arrivals and departures of vehicles on station 104.

The central site 102 may be connected directly to each of the management terminals 110 of a rental site 104 through the network 106 and / or to the charging terminals 112-116.

The central site 102 includes a central management server 132, a calculation and analysis model 134, central dice, a communication model 136, central d it, and a database 138 in which it is stored. in association with each station identifier 104, the number of vehicles available in this station, the hours (dates) of arrival and departure of the vehicles on this station.

The instal lation 100 further comprises _140i-140m modules geolocated ization and common ication, for example, smartphones or prom ises G PS, worn by operators 142i-142 _m located in the area covered by the instal lation 100 through a wireless communication network 144. The calculation and analysis module 134 is configured to calculate, for each station and according to the data communicated by the station and predefined relations, the following data:

^■ the number of vehicles available in the station,

^■ the average waiting time of the vehicle in the station, corresponding to the time between a delivery of the vehicle and a consecutive take, and the average time of removal, corresponding to the time between a vehicle and a subsequent discount, for a given time range,

^■ the number of vehicles in a set, said area, including the stations closest to said station, and

^■ the flow, based on the average waiting time and the average time of removal;

and storing these data in the database 138 in association with the station identifier.

The calculation and analysis module 134 is furthermore configured to classify one or more stations 104 in the list of stations to be fed or in the list of stations to be discharged, again according to the data communicated by the stations.

The calculation and analysis module 134 is furthermore configured to calculate a charge rate for each station 104 classified in the list of stations to be supplied, and a discharge rate for each station in the list of stations to be discharged, always in according to the data communicated by the stations.

The calculation and analysis module 134 is further configured to:

- select as starting stations, stations classified in the list of stations to be unloaded,

- select as stations of arrival, stations classified in the list of the stations to be loaded,

- calculate the balancing times between each start station and each arrival station, - determine the closest operators for each departure station, and

- select the vehicle movement operations by assigning each available operator, a vehicle movement, according to the position of the operator and the balancing time.

The installation 100 makes it possible to manage a plurality of electric vehicles proposed for shared use, for example as part of a rental. The users are able to interact with the different terminals as well as with the different elements of the vehicles and the operators are provided to intervene on the vehicles in order to perform rebalancing tasks.

Alternatively, the calculation and analysis module 134 may be replaced by a plurality of calculation and analysis modules each dedicated to calculating a parameter or to performing a calculation or analysis operation, among those described above.

FIGURES 2-7 are a diagrammatic representation in the form of a diagram of the different steps of a method of rebalancing the vehicle take-off stations of a shared-use installation of vehicles according to the invention.

FIGURE 2 is a general representation of an exemplary method according to the invention and FIGURES 3-6 a detail of various steps of the process shown in FIGURE 2.

The method 200 shown in FIGURE 2 includes a step 202 of collecting data from each of the stations of the installation. This data collection step is performed in real time and the data is collected continuously.

Step 202 is followed by a step 204 for determining different parameters for each of the stations. In this step 204, the parameters determined for each station include:

- the average waiting time,

- possibly the average time of removal,

- the number of vehicles in the station concerned, - the area associated with the station and the total number of vehicles in the area, and

- the flow associated with the station.

The value of each of these parameters is stored in association with the station identifier.

This step 204 is followed by a filtering step 206, supplying several lists: on the one hand, a list of stations to be unloaded and a list of departure stations drawn up from the list of stations to be unloaded and, on the other hand, a list of stations to feed and a list of arrival stations elaborated from the list of stations to feed.

This filtering step comprises on the one hand:

a first step of selecting stations 208 to determine the stations that are part of the lists of stations to be discharged and stations to be powered,

a second selection step 210 of the departure stations, in particular among the stations to be discharged, and arrival stations, in particular among the stations to be supplied. The filtering step 206 is followed by a step 212 of designing the vehicle moving operations, which comprises a selection of the vehicle movement operations to be performed as a function of the balancing time and the operators to carry out these displacements with respect to their geographical location.

In the present example, the method 200 ends with a step 214 of vehicle moves.

Steps 204 to 214 are performed at a predetermined frequency or at the request of a supervisor.

FIGURE 3 is a schematic representation in the form of a diagram of step 202 of data collection.

Step 202 begins with a step 302 of arrival or departure of a vehicle on the station.

The arrival or departure of a vehicle is detected during a step 304.

Step 306 determines the time and date of arrival or departure of a vehicle by consulting a clock. The data relating to the time, or the date, are transmitted to a remote site during a step 308, possibly in association with an identification data of the station and the vehicle concerned.

The remote site stores these data in a database, in association with an identifier of the station during a step 310.

This step 202 is performed at each arrival and departure of a vehicle on each station of the installation.

FIGURE 4 is a schematic representation in the form of a diagram of step 204 for calculating parameters associated with each station.

Step 204 is performed either at a predetermined frequency or at the request of an operator, in particular for supervising the park as a whole.

Step 204 comprises a step 402 of reading the current time on a clock.

According to the current time, a step 404 determines a comparative time slot according to a predefined rule, such as for example:

Time Zone = Current time +5 hours.

During a step 406, all the data relating to the departures and the arrivals on each station since a given date, for example the 30 days preceding the day at which the process is carried out, previously stored for this time slot, are loaded from a base of data. These data form a statistical sample from which it is possible to model the behavior of the station.

Furthermore, in one embodiment of the method, it is possible that each waiting time is also associated with one or more criteria such as the type of day (for example, week, weekend or school period or school holidays). , the data chosen for the average waiting time being those associated with the same criteria as the time when the process is performed. Based on the data loaded a step 408 determines the number of vehicles in station i at the present time, denoted NVS ,, also according to the departure and arrival data of the vehicles stored for the station.

A step 410 determines the average waiting time, noted TAM ,, for the station i, calculated by averaging the time between a catch and a removal of consecutive vehicles on the time slot. This waiting time can be determined differently than what is described, especially only when there is a vehicle in the station, or as the time between two rentals.

Steps 408 and 410 may be performed one after the other or at the same time.

Then, a step 412 determines the total number of vehicles, denoted NVZ, in a zone previously associated with the station by summation of the vehicles being on all the stations of the zone. To do this, step 412 determines the zone i associated with the station i, that is to say all the stations constituting the zone i, namely in the example the station i and the four stations closest to the station i, loads the data relating to each of these stations, determines the number of vehicles in each of these stations and carries out a summation of the numbers of vehicles found in each of the stations constituting the zone i.

Next, a step 414 determines the flow, denoted F ,, associated with the station / ^' , according to the following relation:

F, = - TA! M _t TEAM _;

with TEAM, the average time of removal, between a take and a subsequent delivery on the station / ^' .

The values obtained from the various parameters, namely the parameters NVS ,, TAM ,, NVZ, and F ,, are stored in a database in association with an identifier of the station, during a step 416.

Steps 406 to 412 are performed for each of the n stations of the installation. FIGURE 5 is a diagrammatic representation in the form of a diagram of step 208 for selecting the stations of the lists to be fed and unloaded.

The selection step 208 is performed for each of the n stations of the installation.

It comprises a step 502 comparing the total number of vehicles NVZj in the zone i associated with the station i to a first threshold THRESHOLD1.

If NVZi> SEUIL1, a step 504 determines the TDQ load rate associated with this station according to the following relation:

TDC _t = TAM _i x NVS _t

The station i is added to a list of stations to be unloaded in a step 506, as this means that the area includes too many vehicles.

If at step 502, NVZi <SEUIL1, then a step 508 compares NVZi with a second threshold THRESHOLD2, with SEUIL2 <THRESHOLD1.

If NVZi <SEUIL2 then a step 510 determines the value of a discharge rate, noted TDD ,, for the station according to the following relation:

TDD _; = TAM _; x (NVS _j + 1)

The station i is added to a list of stations to feed during a step 512, because it means that the area does not include enough vehicles.

If at step 508, NVZi> SEUIL2, then the station i is excluded from the rest of the process during a step 514. The zone is indeed considered as moderately fueled with vehicles and therefore does not need to be rebalanced.

FIGURE 6 is a diagrammatic representation in the form of a diagram of step 210 for selecting departure stations and arrival stations.

This step 210 comprises a step 602 testing for each of the n stations of the installation, if this station has not been excluded from the rest of the process, for example during step 514.

If the station is excluded, then the next station is tested. If the station is not excluded, a step 604 compares the flux F, calculated for this station to one or predetermined thresholds. For example, the flow can be compared to two predetermined thresholds to know if the flow F, is around the value zero.

If the comparison is not verified, a step 606 determines whether the flow is below a negative threshold value.

If so, then a step 608 adds this station to a list of departure stations. Indeed, this means that many more vehicles arrive at the station than it leaves and it is appropriate to remove vehicles from the station. Otherwise, the next station is tested.

If in step 604, the comparison is verified, it means that the flow is around zero or positive. In this case, the station is excluded from the rest of the process during a step 610. This means that the time of deposition and the waiting time are equivalent, the station is therefore self-regulating and it is therefore no need to balance it. The next station is tested.

When all the stations i are tested, a step 612 is performed to determine the number "m" of operators available.

A step 614 determines the number of stations already included in the list of departure stations, for example during step 608. Depending on this number and the number m of operators available, an "m" number of operators remaining is determined in a step 616.

In a step 618, the list of stations to be discharged, constituted in step 506, is tested to determine the station with the highest TDC charge rate.

During a step 620, the corresponding station is added to the list of departure stations.

The rate of charge associated with the added station is updated as if it had benefited from a vehicle removal, during a step 622.

Then steps 618-622 are iterated until there are as many starting stations as there are free operators, or until there are no more stations in the list of stations to be unloaded. before reaching the number 'm' of remaining operators. Then, in a step 624, the list of stations to be supplied, constituted during step 512, is tested to determine the station with the smallest TDD discharge rate.

During a step 626, the corresponding station is added to a list of arrival stations.

The discharge rate associated with the station added to the list of arrival stations is updated as if it had benefited from a vehicle addition, during a step 628.

Then steps 624-628 are iterated until there are as many arrival stations as free operators, or until there are no more stations in the list of stations at feed if the number of operators available is greater than the number of balancing operations.

FIGURE 7 is a diagrammatic representation in the form of a diagram of the step 212 for selecting the movements of vehicles and operators for carrying out these displacements.

Step 212 comprises a step 702 for calculating a start parameter for each of the stations in the list of departure stations. The starting parameter is calculated according to the following relation:

1

DP ,.

1

(-) + /> * (Σ

TAM, x NVS _: TAM _k x NVZ _k

with

Then, a step 704 determines an arrival parameter for each of the stations in the list of arrival stations. The arrival parameter is calculated according to the following relation:

1

PDA ,.

1 _^ -, V

() + P x (Σ)

TAM _i x (NVS _i + 1) ^ TAM _k x (NVZ _k + iy with:

p the probability of transfer to at least one nearest station,

- k is the number of stations in the area associated with the station / ^'

- NVZ _k the number of vehicles in the zone associated with station k, and - V _k = l if station k is empty and V _k = 0 otherwise.

Note that the stations in the start list could be determined using these parameters. This would require, however, a greater power of computing means.

During a step 706, the balancing times TDE, entre between each start station i of the list of departure stations (constituted during steps 608 and 620) and each arrival station j of the list of stations arrival (constituted in step 628) are calculated according to the following relationship

TDE _{i → j} = PDD _t - (TDP _{i → j} x PDA _j )

with:

- PDD the start parameter of the start station / ^' ,

PDA _j , the arrival parameter of the arrival station j, and

- TDPj, the journey time from the station / ^' to the station j.

In step 708, the maximum balancing time TDE ^, _max is selected from all calculated balancing times. This corresponds to the best balancing.

In a step 710, the start station and the arrival station with which the maximum balancing time TDE ^, _max are associated are respectively excluded from the list of start stations and from the list of arrival stations.

During a step 712, the various parameters, in particular the number of vehicles of the departure and arrival stations, are updated for the remaining stations in the lists of the departure and arrival stations, as if a balancing had performed between the stations selected in step 710.

It should be noted that it is therefore possible for the same station to be several times in the list of departure or arrival stations. Step 714 determines the geographical positions of each of the available m operators.

In a step 716, the operator closest to the selected start station in step 710 is determined.

A step 718 executes the transmission of a message attributing to the nearest operator the operation of moving a vehicle of the respective departure station and the arrival station concerned.

Steps 708 to 718 are iterated until all operators are assigned to a vehicle move operation or there are no more departure stations or arrival stations in the departure station lists. and arrival stations, if the number of available operators is greater than the number of balancing operations.

Naturally, the invention is not limited to the examples which have just been described.

For example, the following modifications may be made: the vehicles used are not necessarily electric vehicles and therefore not necessarily connected to a charging station,

the parameters of departure and arrival or the selection on the list to be unloaded or fed are not necessarily determined according to the stations close to the station,

some criteria as described above may be optional: for example, we could very well not take into account the flow parameter,

the time slots can also be chosen differently (the ranges are not necessarily slippery), the different thresholds used, in particular for the number of vehicles, can be fixed or variable depending on the zones and the maximum capacity provided for the stations. They can also be in absolute or relative value,

waiting and removal times may also be different from what has been chosen, as long as these parameters are consistently selected.

Claims

A method (200) of rebalancing stations (104) for taking / removing vehicles for shared use, said method (200) comprising the steps of:

detecting (202), for each station, each departure and each vehicle arrival in said station and storing (202) data relating to said station in association with a time data item,

- determination for each station (104):

^■ a number of vehicles available in said station (104),

^■ a waiting time of each vehicle in said station (104), and

determining (204) and storing an average waiting time for each station according to the data stored in association with a given time slot,

filtering (206) the stations (104) according to the number of vehicles available in the station and the average waiting time, said filtering (206) providing:

^■ a first list, called departure stations, and

^■ a second list, called arrival stations,

- designing (208) at least one vehicle moving operation between a station of the first list and a station of the second list, and

- Realization (214) of at least one moving operation of a vehicle.

2. Method (200) according to the preceding claim, characterized in that the waiting time corresponds to a time between a removal of the vehicle on said station and a vehicle taken on said station following said removal.

3. Method (200) according to the preceding claim, characterized in that at least one of the filtering or design steps (212) is performed taking into account for each station at least one data item relating to at least one station adjacent to said station, the one or one of the data being a number of vehicles in a set, said area, including said station and the stations closest thereto.

4. Method (200) according to any one of the preceding claims, characterized in that a zone is defined as comprising:

all the stations located at a distance and / or at a predetermined travel time as a function of the station with respect to which it is determined; or

a predetermined number of stations closest to the station with respect to which it is determined.

5. Method (200) according to any one of the preceding claims, characterized in that the filtering (206) comprises the following steps performed for at least one, preferably each, station (104):

comparing (502) the total number of vehicles in the station or the zone associated with said station (104) with a first threshold;

- when said total number is greater than said first threshold, adding (506) said station (104) to a list, said list of stations to be discharged;

- comparing (508) said total number to a second threshold, less than said first threshold;

- When said total number is less than said second threshold, adding (512) said station to a list, said list of stations to feed.

6. Method (200) according to any one of the preceding claims, characterized in that it comprises a determination (504), for at least one station (104), in particular each station of the list of stations to be unloaded, a parameter, called charge rate, depending on:

- the number of vehicles available in the station (104), and the average waiting time for a given time period in said station (104);

7. Method (200) according to the preceding claim, characterized in that the filtering step (206) comprises a step of selecting (210) starting stations, in particular from the list of stations to be discharged according to the rate charge, said selecting step comprising the steps of:

determining (612-616) a number of operators (142) available to move a vehicle;

for each operator (142) available and by successive iterations:

■ determination (618) of the station, in particular in said list of stations to be discharged, which is associated with the largest charge rate;

Adding (620) said station to the list of start stations;

■ updating (622) the charge rate associated with said selected station by subtracting a vehicle from the number of vehicles available in said station.

8. Method (200) according to any one of the preceding claims, characterized in that it further comprises a determination (414) of a parameter, said flow, as a function of said average waiting time and a time average, said deposition, corresponding to the average time between taking a vehicle and the subsequent removal of a vehicle in the station, the flow corresponding in particular to the inverse of the average waiting time to subtract the inverse of the average time of deposition, the filtering step (206) comprising a selection (210) of at least one start station according to said flow, said selection comprising at least one iteration of the following steps: comparing (606) the flux associated with a station (104) at a first threshold, in particular negative;

when said stream is below said first threshold, adding said station to the list of departure stations.

9. Method (200) according to claims 7 and 8, characterized in that the selection step (608) depending on the flow is optionally performed before the step of determining (612-616) the number of operators. available, the number of available operators taking into account the number of stations whose flow is below the threshold.

10. Method (200) according to any one of the preceding claims, characterized in that it further comprises a determination (414) of a parameter, said flow, as a function of said average waiting time and a time average, said deposition, corresponding to the average time between taking a vehicle and the subsequent removal of a vehicle in the station, the flow corresponding in particular to the inverse of the average waiting time to subtract the inverse of the average time of removal, the filtering step comprising at least one iteration of the following steps:

comparing (604) said stream with at least one range of values, in particular a range of values around 0;

when said stream is in the range of values, excluding (610) from said station lists of stations to be discharged or fed.

11. Method (200) according to any one of the preceding claims, characterized in that it comprises a determination (510), for at least one station (104), in particular each station of the list of stations to feed, of a parameter, called discharge rate, depending on:

- the number of vehicles available in the station (104), and

the average waiting time for a given time period in said station (104);

12. Method (200) according to the preceding claim, characterized in that the filtering step (206) comprises a step of selecting (210) arrival stations, particularly in the list of stations to feed, depending on the rate discharge, said selection comprising the following steps:

determining (612) a number of operators (142) available for moving a vehicle;

for each operator (142) available, and by successive iterations:

■ determination (624) of the station, in particular in said list of stations to feed, which is associated with the smallest discharge rate;

■ adding (626) said station to the list of arrival stations;

■ updating (628) the discharge rate associated with said selected station by adding a vehicle to the number of vehicles available in said station.

Method (200) according to claims 6 or 7 and 11 or 12, characterized in that the designing step comprises a step (212) for determining a balancing time, performed before step (214) of displacement, comprising the following operations, performed by successive iterations:

for each station in the list of departure stations, determining (702) a parameter, said start, based on at least the charge rate;

for each station in the list of arrival stations, determining (704) a parameter, said arrival, based on at least the discharge rate;

calculation (706) of the balancing time between each of the stations of the list of departure stations and each of the stations of the list of arrival stations, as a function of said departure and arrival parameters and of a time of journey between said departure and arrival stations; and - Choice of displacements according to balancing times respectively calculated.

14. Method (200) according to the preceding claim, characterized in that:

the starting parameter is also calculated as a function of a transfer probability to at least one closer station, in particular to all stations in the zone associated with said station; and or

- The arrival parameter is also calculated based on a probability of transfer to at least one station closer, especially to all stations in the area associated with said station.

15. Method (200) according to claims 13 and 14, characterized in that the design step (208) takes into account the balancing time by at least one iteration of the following steps:

determination (708) of the maximum balancing time among the calculated balancing times;

determination (708) of the departure and arrival stations with which said maximum balancing time is associated and selection of these stations;

- exclusion of stations from the departure and arrival lists;

- update the number of vehicles in the departure and arrival stations as if the trip had already been made.

The method (200) of any of the preceding claims, characterized in that the designing step (208) comprises a step of selecting an operator to perform vehicle movement, said selecting comprising the steps of:

- determination (714) of the geographical position of each available operator;

according to said geographical position, determination (716) of the operator closest to the departure station concerned by said displacement; transmitting (718) a message to said nearest operator to signal the task to be performed;

- update the operator status as unavailable.

17. Method (200) according to any one of the preceding claims, characterized in that the time slot is a sliding time slot, whose amplitude is predetermined, and calculated according to the current time.

18. Method (200) according to any one of the preceding claims, characterized in that the stored waiting times are associated with at least one criterion such as the type of day, the waiting times used to calculate the waiting time. average wait being only the times associated with this criterion.

19. A system for rebalancing stations (104) for taking / removing vehicles for shared use, said system comprising:

means (124-128) for detecting each departure and each arrival of a vehicle in the station,

means (132, 134) for determining, for each station, the following parameters:

■ a number of available vehicles,

^■ a waiting time for each vehicle in the station,

means (138) for storing data relating to the waiting time in association with a given hourly data,

means (132, 134) for determining an average waiting time as a function of the data stored in association with a given time slot,

means (138) for storing the determined average waiting time,

means (134, 138) for filtering the stations to be balanced according to said parameters, said filtering providing:

^■ a first list, called departure stations, and

^■ a second list, called arrival stations, and means (134, 138) for designing at least one vehicle moving operation between a station of the first list and a station of the second list.

20. A shared-use installation (100) for vehicles, said installation (100) comprising:

a plurality of stations (104) for taking / removing vehicles; and

means for carrying out the steps of the rebalancing method (200) according to any one of claims 1 to 18, or a rebalancing system according to claim 19.