EP4685031A1

EP4685031A1 - Method, server, and system

Info

Publication number: EP4685031A1
Application number: EP24190598.3A
Authority: EP
Inventors: Shingo ADACHI; Wisinee WISETJINDAWAT; Miki Yonehara; Andrew Broadbent
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 2024-07-24
Filing date: 2024-07-24
Publication date: 2026-01-28
Also published as: WO2026023326A1

Abstract

A computer-implemented method, a server, and a system are provided. The method comprises: receiving, from a passenger device detector configured to detect passenger devices, passenger device information which includes an estimated number of passenger devices in a first vehicle of the mass transport system; receiving, from a passenger counting device, a number of passengers in the first vehicle; comparing the estimated number of passenger devices from the passenger device detector and the number of passengers from the passenger counting device to generate passenger estimation compensation information. Next the method comprises: receiving, from a second passenger device detector configured to detect passenger devices, further passenger device information which includes an estimated number of passenger devices in a second vehicle of the mass transport system; and using the passenger estimation compensation information to estimate a number of passengers in the second vehicle from the further passenger device information.

Description

Field of the Invention

The present invention relates to a computer-implemented method of estimating a number of passengers in a mass transport system, and to a passenger flow analysis server for estimating a number of passengers in the mass transport system.

Background

The operator of a mass transport system comprising a plurality of vehicles, such as buses, trams, trains, etc, needs data describing the numbers of passengers using the mass transport system in order to predict and meet vehicle requirements and ensure the efficient operation of the mass transport system. For example, bus routes and the timetable of each route may be adjusted to meet a predicted demand. Therefore, there is a need to accurately estimate the number of passengers using a mass transport system and the origin/destination of passengers in a journey.
Some existing solutions for estimating passenger numbers include automatic passenger counting systems such as camera counting systems or ticket counting systems. Such systems can provide an estimation of the number of passengers using each section of a transport network such as a section between adjacent bus stops of a bus route. However, these solutions give little insight into the origin/destination of each passenger. Additionally, mass transport systems often include vehicles and/or routes, such as old buses or trainlines with small train stations, which do not have such automatic passenger counting systems. Therefore, accurate estimations of passenger numbers cannot be determined in these cases. Additionally, such automatic passenger counting systems can be expensive and complex to install and so it is undesirable to provide one in every vehicle of a transport system.
Additionally, device detection technology exists for detecting wireless signals from user devices. Most passengers using a mass transport system are expected to travel with a personal device such as a smart phone, a tablet, or a laptop. Such devices typically perform wireless advertising (e.g., Bluetooth (RTM) or Wi-Fi (RTM)) to request a wireless connection (this can be known as a probe request).
The present invention has been devised in light of the above considerations.

Summary of the Invention

Broadly, the present invention relates to a method of using a combination of passenger counting devices and passenger device detectors to monitor passengers using a mass transport system. The method uses data from passenger counting devices and from passenger device detectors to determine a compensation ratio between the number of passenger devices detected in a vehicle and the actual number of passengers in that vehicle. The compensation ratio can then be used to estimate passenger numbers in other vehicles which do not have passenger counting devices.
Thus, in a first aspect, embodiments of the present invention provide a computer-implemented method of estimating a number of passengers in a mass transport system, the method comprising: receiving, from a passenger device detector configured to detect passenger devices, passenger device information which includes an estimated number of passenger devices in a first vehicle of the mass transport system; receiving, from a passenger counting device, a number of passengers in the first vehicle; comparing the estimated number of passenger devices from the passenger device detector and the number of passengers from the passenger counting device to generate passenger estimation compensation information; receiving, from a second passenger device detector configured to detect passenger devices, further passenger device information which includes an estimated number of passenger devices in a second vehicle of the mass transport system; and using the passenger estimation compensation information to estimate a number of passengers in the second vehicle from the further passenger device information.
Advantageously, by using data from a first vehicle having a passenger counting device to determine the compensation information, more accurate estimates of passengers numbers can be determined for vehicles which do not have passenger counting devices. Therefore, accurate measurements of vehicle loads, demand, and available capacity can be obtained enabling an operator of the mass transport system to make more accurate predictions of further demand and capacity. For example, if low passenger numbers are observed on a particular vehicle route, vehicles of the mass transport system assigned to that route may be reassigned thereby saving fuel and increasing available capacity for other routes. Thus, the operation of the mass transport can be adapted to better meet vehicle demand, decrease passenger journey times, and increase the overall efficiency of the mass transport system.
The mass transport system, which may also be referred to as a transport network or public transport network, may be a plurality of vehicles for transporting passengers, wherein the vehicles follow predefined routes. Each route may comprise passenger embarkation points where passengers may embark or disembark from the vehicles. For example, the vehicles may include buses which follow bus routes with bus stops, trams which follow tracks with tram stops, and trains which follow tracks with train stations. The portions of a route between embarkation points may be referred to as sections or route sections. The mass transport system may be operated by a central operator which generates a timetable for operating the mass transport system and determining which route and departure time each vehicle is assigned to at any given time.
The passenger device detector may be a probe request module which is configured to detect wireless signals from passenger devices known as probe requests. For example, the passenger device detector may be configured to detect PAN (personal area network) signals such as Bluetooth, BLE (Bluetooth Low Energy), or Wi-Fi signals transmitted by the passenger devices. The PAN signals may be advertising signals comprising a device ID such as a MAC address or a Bluetooth device address. The passenger device detector may be configured to count how many advertising signals are detected from different devices, e.g., having distinct device IDs. In some examples, the passenger device detector may be configured to receive one or more advertising signals from one or more passenger devices and form a connection (e.g., pair with) the one or more passenger devices.
The passenger counting device may be an automatic passenger counting system for determining a number of people on or entering a vehicle. For example, the passenger counting device may comprise a surveillance system comprising one or more cameras on a vehicle and a module for automatically counting the people in a field of view of the one or more cameras. In other examples, the passenger counting device may comprise a ticket counter or barrier, or a manual counting device operated by e.g., a vehicle driver. The passenger counting device may be configured to count the number of passengers on the first vehicle. Thus, the passenger counting device may be located on the first vehicle. In other examples, the passenger counting device may be located separately to the first vehicle such as on a train platform or at a bus stop.
The passenger estimation compensation information may comprise a ratio between the estimated number of passenger devices and the number of passengers. In some examples, the passenger estimation compensation information may further comprise further data including one or more of: vehicle type, vehicle location, route data, route section data, time data, and/or date data associated with conditions of the first vehicle when the estimated number of passenger devices and the number of passengers were determined. Thus, compensation information may be generated which corresponds to a particular vehicle type, route, time, day, etc as discussed in detail below. By accounting for days and time of journeys compensation information may be determined which accounts for routes, times, and/or days when certain demographics of people are expected to be using the mass transport network (for example, children travelling to school). Different demographics of people may be expected to own different numbers of passenger devices. Therefore, the number of passengers estimated using the compensation information can be more accurate than if only global compensation information were determined.
In some examples the method may further comprise: determining, from the passenger device information, when and/or where each passenger device embarks and/or disembarks from the first and second vehicles. For example, the passenger device information may include time and/or location data indicated when and/or where each passenger device was detected. Therefore, an origin and/or a destination of each passenger device journey on a vehicle may be determined by monitoring when and/or where the passenger device was first detected and when/where the passenger device is no longer able to be detected. In this way an estimation of passenger trips comprising origin and destination information may be estimated. By monitoring passenger journeys the optimization of predefined vehicle routes used by vehicles of the mass transport system may be performed. For example, if a large number of passengers are observed who have the same origin and destination, a more direct vehicle route between that original and destination may be introduces thus saving fuel and reducing passenger journey times.
In some examples, as mentioned above, the passenger device information includes one or more device identifiers. For example, the device identifiers may be transmitted by respective passenger devices using PAN signals. In these examples, the method may comprise determining when one or more of the device identifiers has changed. For example, if a passenger device transmits a first identifier and then that passenger device transmits a second identifier which is different to the first identifier, the method may comprise detecting that change and determining that the same passenger device is being detected. Determining when one or more of the device identifiers has changed may comprise determining that a first passenger device identifier is no longer being detected, detecting a new device identifier, and determining if the detection of the new device identified corresponds to a time when the vehicle has stopped (which would allow the vehicle's doors to be opened and for passengers to get on and off) or not. When it is determined that the vehicle has not stopped, or that the vehicle has not stopped at a predefined embarkation location, then it may be determined that the passenger device has changed its device identifier from the first to the second identifier. In this way, passenger devices can be tracked through the mass transport system even if they change their device identifier.
The method may further comprise receiving location information associated with the first and/or second vehicles. In some examples, the method may further comprise receiving location information associated with the first and/or second passenger device detectors. For example, the passenger device information may include the location information. The location information may include a current location of the first and/or second vehicles. Thus, the location information may be used to determine passenger journeys using the mass transport system. Further, in examples as discussed below, the location information may be used to generate compensation information which corresponds to a specific route or location. Thus, location information associated with the second vehicle may be used to select compensation information for estimating the number of passengers on the second vehicle, for that location or vehicle route.
For example, the method may further comprise: using the location information to determine when each passenger device embarks and/or disembarks from the first and/or second vehicles. Additionally, the location information may be used to determine when one or more of the device identifiers has changed, for example, by determining if the vehicle has stopped at an embarkation location.
In further examples, the method may further comprise: using the location information to determine a plurality of estimated numbers of passengers (e.g., a plurality of passenger counts) corresponding to each of a plurality of route sections travelled by the first vehicle. Each route section may be a portion of a route travelled by the first vehicle in-between two embarkation points where passengers may embark or disembark from the vehicle. For example, a route section may be a portion of a bus route between two bus stops or a section of a railway line between two stations. In this way, more detailed compensation information may be generated which accounts for passenger trends on different routes and route sections.
In this example, generating the passenger estimation compensation information may comprise generating section compensation information corresponding to each of the route sections travelled by the first vehicle. For example, a first section compensation information may be determined for a first route section and second section compensation information may be determined for a second route section.
Estimating the number of passengers in the second vehicle may comprise using the section compensation information and the further passenger device information, for a (or each) route section, to estimate a number of passengers in the second vehicle for that route section travelled by the second vehicle.
In some examples, the method may comprise receiving date, day, and/or time information associated with the passenger device information and the number of passengers from the passenger counting device. In this way, the day or the week, time of year, and/or time of day may also be accounted for during the estimation of the passenger numbers.
In some examples, the method may comprise receiving location information, and date and/or time information associated with the passenger device information and the number of passengers from the passenger counting device. Therefore, routes, route sections, and time/day/date may each be accounted for in the generated compensation information. Accordingly, compensation information may be determined for different parts of a vehicle timetable and for different vehicle routes thereby accounting for more passenger trends and enabling more accurate estimation of passenger numbers and predictions of future vehicle demand.
The passenger estimation compensation information may comprise a compensation ratio between the number of passengers detected by the passenger counting device and the estimated number of passenger devices detected by the passenger device detector. In some examples, the compensation ratio may be determined from an average of multiple, distinct, estimates of the number of passenger devices from one or more different passenger device detectors. Likewise, the compensation ratio may be determined from an average of multiple, distinct, counts of the number of passengers from one or more passenger counting devices. For example, compensation information may be determined using multiple passenger device and passenger counts observed for e.g., a particular route, and/or a particular time of day and/or a particular day. For example, passenger estimation compensation information may be generated for each route and/or journey time of the mass transport system. By using more data to generate the compensation information, outlier events may be ignored and general trends in passenger behaviour can be observed. Therefore, a more accurate estimation of passenger numbers may be performed.
Passenger estimation compensation information may be generated by comparing passenger device information received from a plurality of passenger device detectors associated with a plurality of vehicles of the mass transport system, and passenger numbers received from a plurality of passenger counting devices associated with the plurality of vehicles. In this way, multiple different instances of compensation information may be generated for each vehicle, or the data from each vehicle may be combined as described above to generate overall passenger compensation information (e.g., for a particular route, section and/or date/time).
The passenger estimation compensation information may be used to estimate the passenger numbers for a plurality of second vehicles of the mass transport system e.g., for a plurality of routes and journey times. For example, the plurality of second vehicles may be vehicle of the mass transport system which do not have passenger counting devices (but do have passenger device detectors).
The method may further comprise: allocating vehicles with passenger counting devices to selected routes of the mass transport system, and allocating vehicles which do not have passenger counting devices to other selected routes of the mass transport system based on the estimated number of passengers. Allocating vehicles to selected routes may also comprise allocating vehicles to selected timetables associated with those routes.
In some examples, the method may further comprise: determining updated timetable information based on the estimated number of passengers. For example, determining updated timetable information may comprise allocating certain vehicles to selected routes, allocating more or fewer vehicles to selected routes, adapting routes, adjusting vehicle journey times, etc. For example, if a large number of passengers are predicted based on the estimated passenger numbers to be on a certain vehicle at a certain time (e.g., a bus), the method may comprise adjusting a departure time of a second vehicle (e.g., a train) to enable those passengers to use the second vehicle.
In some examples, the method may further comprise: using passenger device information from one or more passenger device detectors to determine estimated journeys of one or more passengers across the mass transport network, each passenger journey involving one or more vehicles and one or more vehicle routes of the mass transport network. Accordingly, in this example, the computer-implemented method may comprise determining updated timetable information based on the estimated passenger journeys.
The method may further comprise: transmitting an update signal to the passenger device detector and/or the second passenger device detector based on the passenger device information from that passenger device detector. For example, the update signal may be configured to increase or decrease a monitoring frequency of the respective passenger device detector. The monitoring frequency may determine a frequency at which the respective passenger device detector assesses a number of passenger devices within its detection range. For example, the transmitted signal may be configured to increase the monitoring frequency if the estimated number of passenger devices is above a predetermined threshold number. For example, if a large number of passenger devices are detected in a vehicle, the monitoring frequency may be increased to improve the accuracy of the detection and reduce the likelihood of the devices being miscounted. Alternatively, if a low number of passenger devices are detected on the vehicle, then the signal may be configured to reduce the monitoring frequency in order to reduce power and resource consumption of the passenger device detector.
In another or the same example, the update signal may be configured to change a scanning mode of the passenger device detector and/or the second passenger device detector between an active scanning mode and a passive scanning mode, or vice versa. When the passenger device detector is configured to detect Bluetooth signals, the transmitted update signal may be configured to cause the passenger device detector to switch between passive scanning for Bluetooth signals and active scanning for Bluetooth signals in order to change the level of data collected. For example, if a vehicle is stopped at a passenger embarkation point, then the signal may be configured to switch the passenger device detector to active scanning mode to collect additional information about passenger devices, e.g., the names of the devices, in addition to the device IDs to improve the accuracy of the detection and reduce the likelihood of the devices being miscounted. Alternatively, if a vehicle is not stopped, the signal may be configured to switch the passenger device detector to passive scanning mode to minimise the data collection.
In a second aspect, embodiments of the present invention provide a method of using precomputed compensation information to estimate a number of passengers. That is, the second aspect of the present invention provides a computer-implemented method of estimating passengers in a mass transport system, the computer-implemented method comprising: receiving, from a passenger device detector configured to detect passenger devices, passenger device information including an estimated number of passenger devices in a vehicle of the mass transport system; receiving compensation information, the compensation information indicating a ratio between a number of passengers and the number of passenger devices; and using the compensation information and the estimated number of passenger devices to estimate the number of passengers in the vehicle.
The compensation information may be determined using historical ratios between numbers of passengers and passenger devices in the mass transport system. For example, the compensation information may be determined using any of the methods described above for the first aspect.
In this way, an estimate of passengers can be determined for a vehicle which does not have a passenger counting device by counting the number of passenger devices on the vehicle and adjusting that count using the compensation information.
In a third aspect, embodiments of the present invention provide a passenger flow analysis server configured to perform the computer implemented method of the preceding aspects. For example the passenger flow analysis server may be configured to: receive, from a passenger device detector configured to detect passenger devices, passenger device information which includes an estimated number of passenger devices in a first vehicle of the mass transport system; receive, from a passenger counting device, a number of passengers in the first vehicle; compare the estimated number of passenger devices from the passenger device detector and the number of passengers from the passenger counting device to generate passenger estimation compensation information; receive, from a second passenger device detector configured to detect passenger devices, further passenger device information which includes an estimated number of passenger devices in a second vehicle of the mass transport system; and use the passenger estimation compensation information to estimate a number of passengers in the second vehicle from the further passenger device information.
In a fourth aspect, embodiments of the present invention provide a system for monitoring passengers of a mass transport system, the system comprising: a plurality of passenger device detectors configured to detect passenger devices on respective vehicles of the mass transport system to generate passenger device information which include an estimated number of passenger devices in each respective vehicle; one or more passenger counting devices configured to count a number of passengers in one or more of the respective vehicles of the mass transport system; and the passenger flow analysis server according to the third aspect.
Additional aspects of the invention may relate to systems configured to execute the computer-implemented method of any one of the first, and second aspects of the invention. Specifically, the system may comprise a processor which is configured to execute the respective computer-implemented methods of the first, and second aspects of the invention.
Additional aspects of the invention may provide a computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the steps of the computer-implemented method of the first and second aspects of the present invention. Further aspects of the invention may provide a computer-readable storage medium, having stored thereon, the computer program of the previous aspects of the invention.
The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

Summary of the Figures

Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which:

Fig. 1 shows a diagram of a system for estimating a number of passengers according to aspects of the present invention;
Fig. 2 shows a flow diagram of a method for estimating a number of passengers according to aspects of the present invention;
Fig. 3 shows another diagram of the system for estimating a number of passengers;
Fig. 4 shows a flow diagram of a method for optimising operation of a mass transport system;
Fig. 5 shows a flow diagram of a method for monitoring passenger journeys which use a mass transport system;
Fig. 6 is a table of example passenger numbers received from passenger counting devices;
Fig. 7 is a table of example passenger device information received from passenger device detectors;
Fig. 8 is a table of example compensation rations determined from numbers of passengers and estimated numbers of passenger devices;
Fig. 9 is a table of example of combined counting data
Fig. 10 is a table of example output of the trip estimation unit of the passenger flow analysis
Fig. 11 is a table of example output of the journey estimation unit
Fig. 12 is a table of example vehicle allocation output of the optimisation unit; and
Fig. 13 is a table of example timetable change output of the operations module.

Detailed Description of the Invention

Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art.
Fig. 1 shows a diagram of a system for estimating a number of passengers in a mass transport system. The mass transport system comprises a plurality of vehicles 200 for transporting passengers. The vehicles may be any mode of transport in a mass transport network such as buses, trams, trains, etc. In Fig. 1 a first vehicle 200A and a second vehicle 200B are shown, however the system may comprise many more vehicles 200.
The system for estimating the number of passengers comprises a passenger flow analysis server. The passenger flow analysis server comprises a data collection module 102 for receiving data from sensors located in the mass transport system, an estimate comparison module 104 for comparing the data from the sensors to generate compensation information, and an estimate compensation module 106 for estimating the number of passengers using the generated compensation information. The sensors comprise passenger device detectors 204 which are configured to detect passenger devices travelling on vehicles of the mass transport system and passenger counting devices 202 which are configured to count the passengers travelling in the vehicles.
In this example, the data collection module 102 is configured to receive passenger device information from a passenger device detector 204A in the first vehicle. The passenger device information comprises an estimated number of passenger devices in the first vehicle. The data collection module 102 is also configured to receive, from a passenger counting device 202A, a number of passengers in the first vehicle. The passenger counting device may be located in the first vehicle (e.g., as a ticket counting device or camera systems) or outside of the first vehicle (e.g., as ticket barriers). Additionally, the data collection module also receives further passenger device information a second passenger device detector in second vehicle which includes an estimated number of passenger devices in the second vehicle.
Herein, the term "module" is used to refer to a functional module which is configured or adapted to execute a particular function. The modules may be implemented in hardware (i.e. they may be separate physical components within a computer), in software (i.e. they may represent separate sections of code, which when executed by a processor, cause the processor to perform a particular function), or in a combination of both.
The vehicles 200 are connected to the passenger flow analysis server 100 over a network, for example a wide area network (e.g., the internet). In some examples the vehicles include a cellular internet modem allowing them to establish a remote (wireless) connection with the passenger flow analysis server 100.
Fig. 2 shows a flow diagram of a method for estimating a number of passengers according to aspects of the present invention. For example, the method may be performed by the passenger flow analysis server of Fig. 1.
First, in step S100, the data collection module 102 receives the passenger device information from the passenger device detector in the first vehicle. The passenger device information includes an estimated number of passenger devices in the first vehicle.
Next, in step S102, the data collection module 102 receives a passenger count from the passenger counting device in the first vehicle, the count representing the number of passengers in the first vehicle.
Next, in step S104, the estimate comparison module compares the estimated number of passenger devices from the passenger device detector and the number of passengers from the passenger counting device to generate the passenger estimation compensation information. The passenger estimating compensation information includes a ratio of the number of passengers to the number of passenger devices detected in the first vehicle.
Next, in step S106, the data collection module receives the further passenger device information from the second passenger device detector in the second vehicle. The further passenger device information includes an estimated number of passenger devices detected in the second vehicle.
Finally, in step S108, the estimate compensation module uses the passenger estimation compensation information and the further passenger device information to estimate a number of passengers in the second vehicle. For example, the estimated number of passengers can be generated by multiplying the compensation ratio by the estimated number of devices in the second vehicle.
Whilst shown in series, steps S100, S102, and S106 can be performed in any order (including simultaneously). For example, the passenger count received in step S102 may be received before or at the same time as receiving the passenger device information in S 101. However both the passenger device information and the number of passengers must be received before step S104 can be performed. Similarly, step S106 can occur at any point prior to step S108.
Fig. 3 shows another diagram of the system for estimating the number of passengers in more detail.
The system comprises a passenger flow analysis server 100, in communication with passenger device detectors 204 and passenger counting devices 202 associated with various vehicles 200A, 200B, of a mass transport system such as buses or metro trains. Generally, some vehicles are equipped with both passenger counting devices and passenger device detectors, as shown for the first vehicle 200A in Fig. 3. However, some vehicles may not be equipped with passenger counting devices, such as the second vehicle 200B in Fig. 3, which only has a passenger device detector. The flow analysis server is in communication with a plurality of passenger device detectors and a plurality of passenger counting devices associated with a plurality of vehicles of the mass transport system (only two shown) (with and without passenger counting devices). Finally, the system comprises an operation module 300.
As described above, the passenger flow analysis server 100 is configured to receive numbers of passengers from the passenger counting devices and estimated numbers of passenger devices from the passenger device detectors in order to generate passenger estimation compensation information. Additionally, the passenger flow analysis server is configured to combine the number of passengers counted by the plurality of passenger counting devices and the estimated numbers of passenger devices monitored by passenger device detectors to generate combines passenger estimation compensation information. The passenger flow analysis server can then estimate numbers of passengers for all of the services of the mass transport network.
The passenger flow analysis server comprises a compensation module 106, a trip estimation module 108, a journey estimation module 110, and an optimisation module 112. The passenger flow analysis server can be implemented on a cloud server, and the data connections between the passenger counting devices or probe monitoring devices and the passenger flow analysis server would be by internet connections.
Additionally, the passenger flow analysis server 100 receives operation information of the mass transport network from the operation module 300. The operating information includes, but is not limited to, timetables for vehicle routes (stored in timetable management module 302), the allocation of vehicles to selected routes and times, planned vehicle allocations and historical record of vehicle allocations and journeys (stored in vehicle allocation module 304). The compensation module 106, the trip estimation module 108, the journey estimation module 110, and the optimisation module 112 are configured to used the operation information in their respective computations. For example, the trip estimation module is configured to use the vehicle timetable and passenger device estimations to determine a trip taken by a passenger.
The compensation module 106 compares the number of passengers 1021 and the number of passengers' devices onboard 1022 in each section and then computes the compensation ratio of these numbers for each route and time period, which will be stored in compensation ratio database 1062. The compensation module 106 also estimates the number of passengers in each section for vehicles without passenger counting devices based on the compensation ratio stored in database 1062, and then stores the result as combined counts 1024. In some examples, the data stored in 1024 may be referred to as probe paths. That is, it may store a path (in time and space) along which an identified device travelled during data collection. This both provides: (i) a count of the number of devices (a path per device); and (ii) original/destination information of the respective device.
The trip estimation module 108 computes the estimated passenger trips within each single route, describing the initial and last stops and the used service/vehicle based on the passenger counts, the device counts in 1022 and the respective compensation ratio in database 1062.
The journey estimation module 110 computes the estimated passenger journeys over the whole mass transport network with multiple routes, describing the initial and last stops and the used service/vehicles based on the device counts 1022 and the trip estimation, and then stores this estimated journey data, or the estimated origin/destination matrix in database 1015.
The optimisation module 112 allocates vehicles with and without passenger counting devices to services to reduce the uncertainty of estimation of the number of passengers onboard in services operated by the vehicles without passenger counting devices. The optimisation module 112 also plans the timetable changes to improve operation of the mass transport network based on the estimated origin/destination matrix, or the estimated passenger journeys over the whole mass transport network.
The data stored in the passenger count database 1021, device count database 1022, compensation ratio database 1062, combined count database 1014, and estimated origin/destination matrix database 1015 are illustrated in Figures 6, 7, 8, 9, and 10, respectively. The estimated trip computed by the trip estimation unit (1002) is explained in Figure 11. In addition, the vehicle allocation and the timetable changes computed by the optimisation unit (1004) are explained in Figure 12, and 13, respectively.
The passenger counting devices 202 in the vehicles 200 count the number of passengers onboard in each section between adjacent stops in vehicles and upload the counting data to the passenger flow analysis server. Any existing methods known per se in the art can be used to implement such passenger counting devices. For example, there are methods based on an analysis of onboard CCTV images to count passengers or an analysis of dedicated video cameras or infrared sensors to detect the number of passengers who are passing doors to get on/off, etc.
A passenger device detector 204 monitors the paths of passengers' devices onboard. It uploads the probe path data, including the number of passengers' devices in each section, to the passenger flow analysis server. The onboard passenger device detector 204 includes a probe receiver 2042, a locator 2043 and a control unit 2041. The probe receiver 2042 receives probe requests from passengers' devices in wireless communication protocols like Wi-Fi and Bluetooth and sends the data to the control unit 2041. The probe receiver 2042 may change its configurations, e.g., frequency and types of monitored packets, to monitor probe requests based on the instruction of the control unit 2041. The locator 2043 identifies the vehicles' locations and the current timestamp where probe monitoring devices are equipped and sends the data to the control unit 2041.
The control unit 2041 processes the probe request data and the vehicles' locations to estimate the paths of devices. The probe path data describes the range from the initial stop to the last stop where the devices are onboard by distinguishing the potential change of devices' address while the vehicles are running between two adjacent stops. The control unit 2041 also uploads the probe path data to the passenger flow analysis server 100. In addition, the control unit 2041 may determine the configuration for the probe receiver 2042 based on the vehicle's location, timestamp and the compensation ratio computed in the passenger flow analysis server 100.
The detailed process of the passenger device detector 204 is explained in Figure 5. Also, the probe path data sent from passenger device detector 204 to the passenger flow analysis server is described in Figure 7. The control unit 2041 of the passenger device detector 204 may be implemented in the passenger flow analysis server 100.
The operation module 300 manages the timetables of routes and vehicle allocation in the mass transport network. The operation module 300 allows the timetable and vehicle allocation to be updated by the optimisation module 112 of the passenger flow analysis server. The operation system distributes the timetable and vehicle allocation data to the driver navigation units 206 of the vehicles 200.
The driver navigation units 206 show the corresponding vehicles' timetables and related information to the drivers based on the data distributed by the operation module 300. For example, a driver navigation unit shows the route the vehicle should run, the names and the arrival and departure times of upcoming bus stops as the driving instructions to the driver. The driver then will drive the vehicle by following the instructions, e.g., departure time of each bus stop, etc.
By combining the data from passenger counting devices and probe monitoring devices as described herein, the passenger flow data in the mass transport network, including the number of passengers in the vehicle without a passenger counting device, and the origin-destination of passenger's journey, can be estimated accurately. By optimising the allocation vehicles with and without passenger counting devices, the accuracy is improved. Using the improved accuracy of passenger flow data, the operation of the mass transport, e.g., the timetable of bus operation, can be made to better match the travel demand in the area.
The compensation information and the passenger device information may also comprise information about a vehicle route, route section, time, or day for which the information corresponds to. Therefore, the estimated number of passengers may be generated for a particular route, section, time, and/or day.
Fig. 4 shows a flow diagram of a method for optimising operation of a mass transport system. This method can be run periodically, for example weekly, daily, or bi-hourly. In a first step, S200, a given route is selected (e.g., bus route, train route, etc.) from a set of routes to be optimised. Then, in step S202 the compensation information is updated by comparing the stored passenger counts to the stored device counts for the selected route. This step can be performed by the compensation module 106. For the sections in the route operated by vehicles with passenger counting devices, the compensation module 106 computes the number of passenger devices. Then, the compensation module computes the ratio for the number of passenger devices to the number of passengers for each section. For each section, the compensation module finds the corresponding group in the compensation ratio database 1062. Then, the compensation module updates the compensation ratio (device/passenger ratio) of the group with the average over a predetermined period (e.g., the last 30 days). The compensation module also updates an amount of collected data in the same period.
In step S204, the compensation module applies the updated compensation ratio in order to estimate the number of passengers based on the device count. The compensation module retrieves the passenger counts and device counts from the respective databases for the selected route, and for the sections in the route operated by vehicles without passenger counting devices, the compensation module computes the number of passenger devices. Then, the compensation module computes the estimated number of passengers by dividing the number of devices by the corresponding compensation ratio. The compensation module then stores the computed estimated number of passengers, for example in the combined counts database 1024.
In step S206, which can be performed by the trip estimation module 108, the passenger's trips on the selected route are estimated and sent to the journey estimation module 110. An example of estimated passenger trips are shown in Fig. 10. The trip estimation unit retrieves the passenger device information, in this example probe path information, for the selected route. Then the trip estimation unit attaches the compensation ratio of the appropriate group from the compensation ratio database 1062 to each probe path. The trip estimation unit then adjusts the attached ratios to minimise the discrepancy over sections between the number of passengers counted by the passenger counting devices and the estimated number of passengers computed by dividing the number of ongoing probe paths (i.e., the device count) by the attached ratios. If the adjusted ratio is less than one, this means that a passenger has multiple devices. Therefore the passenger's trip has several probe IDs with the same (or a very similar) path. On the other hand, if the adjusted ratio is more than one, this means that several passengers in a group have a single device (or a number of devices fewer than the number of passengers in the group). In this instance, several passengers' trips would have the same probe ID. The trip estimation module organises these combined and duplicated trips as shown in Fig. 10, and sends the data to the journey estimation module.
Steps S200 - S206 are repeated until all routes of interest are completed. When they have been 'Yes', the method moves to step S208 which may be performed by the journey estimation module. The journey estimation module retrieves the estimated trips from the trip estimation module and connects all plural trips over different routes which have common probe IDs during transit over different routes (trips). The journey estimation module then stores these journeys the estimated origin/destination matrix database 1015.
Finally, the method moves to step S210 which may be performed by the optimization module. The optimisation module allocates vehicles with passenger counting devices to the service or route with a higher percentage (e.g. 20%) of missed data compared with collected data in the compensation ratio table. The optimization module assumes a basic allocation of vehicles, and then explores the choices to swap vehicles with and without passenger counting devices so as to better cover the sections with significant missed data with vehicles with passenger counting devices. An example of the vehicle allocation is shown in Fig. 12. The optimisation module also changes a part of the timetable, e.g., the arrival and/or departure times of specific stops, based on the latest travel demand or the estimated origin/destination matrix or estimated journeys stored in the estimated origin/destination matrix database 1015. For example, if journeys show that 20 passengers are waiting for 10 minutes in transit to catch a bus, but that the previous bus is scheduled to depart just before 30 seconds, then delaying the previous bus for one minute reduces the overall travel time for the passengers. An example of this kind of timetable change is shown in Fig. 13.
Fig. 5 shows a flow diagram of a method for monitoring passenger journeys which use a mass transport system. The passenger device detector 204 starts the process with step S300. The entire process may be periodically repeated, for example every 100 milliseconds. In step S300, the control unit 2041 of the passenger device detector updates the configuration of the probe receiver 2042, and the probe receiver monitors probe requests made by passenger devices. This data is sent to the control unit. To update the configuration of the probe receiver, the control unit receives the corresponding compensation ratio where (along which section, or route) the vehicle is running from the passenger flow analysis server based on the vehicle's location and timestamp. Then, the control unit modifies the configuration of the probe receiver (e.g., frequency and types of monitored packets) based on the compensation ratio and the monitored number of devices. For example, if the corresponding compensation ratio is higher than 2.0 and/or the monitored number of devices is large enough, the control unit may change the monitoring frequency to be doubled so as to prevent packet loss in the situation of many devices and packets. Otherwise, the frequency may be lowered to the default value to minimise data processing.
In step S302, the control unit detects the vehicle status. Specifically, the control unit determines if the vehicle is stopped at a specific stop (e.g., a bus stop) if the location data from the locator 2043 matches the predefined location of the specific stop, and the location data is steady enough (e.g., for more than a few seconds, for example 10 seconds).
In step S304, if the detected vehicle status is that it has not stopped ('no'), the control unit detects address changes of the probe requests due to device-based randomisation and connects the changed addresses (devicelD1 has ceased to broadcast, devicelD2 has begun to broadcast, but no passengers have left the vehicle and therefore devicelD2 is the new ID for deviceID1). The method then moves to step S306 when the vehicle has stopped, whereupon the updated probe path data is sent to the passenger flow analysis server.
Fig. 6 is a table of example passenger numbers received from passenger counting devices located on vehicles of a mass transport system. Each record in the table represents a number of passengers in a vehicle with a passenger counting device, which is travelling in a route section between two adjacent bus stops - in this example, from a departing bus stop to a subsequent, arriving bus stop. For other vehicles, the section may be defined by different nodes such as predefined distances or train stations, etc.
The first column 401 comprises ID numbers assigned to each record, which are represented by each row of the table. The second column 402 comprises timestamps associated with each record. In this example, each timestamp is a time of arrival at the arriving or departing bus stop by the vehicle. In examples, where the vehicle doesn't stop at the arriving or departing bus stop, for example, because no passengers embarked or disembarked from the vehicle, then the time to pass by the bus stop may be used for this column. The next columns 403, 404, 405, 406, and 407 contain a route ID, service ID, vehicle ID, and IDs for the departing and arriving bus stop of the record where the data for each column in the table was collected. A service means a single transport service in a route by a vehicle. On a different date, the same service, e.g., a bus service scheduled to depart bus stops at specific times defined in the timetable, may be operated by different vehicle. The vehicle ID shows which physical vehicle was used to operate the identified service on the day. The next column 408 contains the number of passengers detected on the vehicle in the identified route section by the passenger counting device. The final columns 409 and 410 contain the number of passengers who have boarded the bus at the departing bus stop, and disembarked the bus at the arriving bus stop.
Fig. 7 shows a table of example passenger device information received from passenger device detectors located on vehicles of the mass transport system. Each record (i.e., row) in the table represents a device which was detect on board in a vehicle in service operation in a route section e.g., between two adjacent bus stops - from a departing bus stop to a subsequent arriving bus stop.
The first column 501 contains an ID which is assigned to each record. The second column 502 contains an ID for the probe monitoring device (i.e., the passenger device detector) which is associated with each record. The third column 503 contains a timestamp for each record, that is, the time of arrival of the probe monitoring device at the departing or arriving bus stop. In examples, where a vehicle does not stop at the departing or arriving bus stop, i.e., no passenger got on or off, then the time to pass by the stop can be used for this column.
The next columns 504, 505, 506, 507, and 508 contain the route ID, service ID, vehicle ID, and the departing and arriving bus stop IDs for each record, indicating where the data from each row was collected.
The next column 509 contains a single device / probe ID or a plurality of device / probe IDs, for each row, which were detected by probe monitoring device and identified to be from a passenger device. If a passenger device is determined to use a same network address for the whole route section, then this column will contain just a single ID. On the other hand, if a passenger device changes its network address while the vehicle is in the route section (i.e. between stops), for example, due to randomisation techniques, then this column contains a plurality of device IDs. The device IDs may be an irreversible hash value of the network address used by a detected passenger device. The final column 510 contains a connected ID which is assigned by the probe monitoring device to each passenger device for the duration of a trip in the same vehicle by that passenger device.
In general cases, a plurality of records associated with each route section would be expected, for example where the columns 502, 503, 504, 505, 506, 507, and 508 of Fig. 7 have the same values. The passenger flow analysis server 100 is able to count the number of passenger devices in that route section, for that vehicle, by counting the number of records appearing in the table for that section.
For records where the vehicle ID 506 and the connected ID 510 are the same, the passenger flow analysis server 100 is configured to identify the passenger device's trip path, by recording the origin/destination of that passenger device. For example, the departing bus stop ID of the record for that passenger device with the earliest timestamp is determined to be as the device origin, and the arriving bus stop ID of the record with the latest timestamp for that passenger device is determined to be the destination.
Fig. 8 shows a table of example compensation ratios determined from the numbers of passengers detected by the passenger counting devices and the estimated numbers of passenger devices detected by the passenger device detectors. Each record (i.e., row) in the table represents a group and its compensation information associated with that group. Each group comprises a range of route sections (e.g., defined by bus stops) and a time period which are determined to exhibit similar characteristics such as the compensation ratio between the number of passengers and numbers of passenger devices.
The first column contains a record ID for each record. The next columns 602, 603, and 604 contain the route ID, node (e.g., bus stop) ID range, and time period which together form a definition for each group associated with the respective record in the table. The next column comprises compensation ratios, which are the estimated ratio of the number of passenger devices to the number of passengers for each group.
The next column 606 contains a number of collected data corresponding to the group to compute the compensation ratio in the fixed period, e.g., 30 days. On the other hand, the following column 607 stores a number of missing data corresponding to the group in the same period, in other words, operated by vehicles without passenger counting devices. If the ratio of column 607 to column 606 in a record exceeds a specific value, e.g., 1/10, then the optimisation unit 1004 is configured to allocate vehicles with passenger counting devices to the services covered by that group.
The table in Figure 9 shows the combined counting data stored in the combined counts database 1024 of the passenger flow analysis server 100. The table combines the passenger counting data stored in the passenger counts database 1021 and the estimated number of passengers computed by the compensation unit compensation module 106 based on the device count / probe path data stored in the device count database 1022. A record in the table represents the number of passengers in a vehicle with and without a passenger counting device in service operation in a section between two adjacent bus stops - from a departing bus stop to the next arriving bus stop. The columns of this table include all columns of the table in Figure 6. This table has an additional column on the data source 709. This table contains all records in the table in Figure 6 with the data source 'Counting'. In addition, this table also contains records on the number of passengers in a vehicle without a passenger counting device. These records are computed by the compensation module 106 based on the probe path data stored in the device count database 1022 and have the data source 'Probe estimation'.
The table in Figure 10 shows the output of the trip estimation module 108 of the passenger flow analysis server 100. A record in the table represents a trip by a passenger on a single route within the mass transport network. The record is based on the passenger device's trip but is compensated with the compensation ratio. The column 801 stores the ID of the record. The columns 802, 803 and 804 store the IDs of the route, service, and vehicle the estimated passenger took. The columns 805 and 806 store the IDs of the initial (origin) and last (destination) bus stops of the trip. The columns 807 and 808 store the timestamps at the initial and last bus stops in the trip. The columns 809 and 810 store the probe IDs collected at the initial and last sections of the trip.
The table in Figure 11 shows the output of the journey estimation module 110, which is then stored in the estimated origin/destination matrix database 1015 of the passenger flow analysis server 100. A record in the table represents a network-wide journey by a passenger. The record may be a trip in the table of Figure 10 or may connect plural trips in the table of Figure 10, which have the common probe IDs during the transits over different routes (trips). The column 901 stores the ID of the record. The columns 902 and 903 store the IDs of the initial (origin) and last (destination) bus stops of the journey. The column 904 stores the IDs of the trips, which constitute the journey. The columns 905 and 906 store the timestamps at the initial and last bus stops in the trip.
The table in Figure 12 shows the vehicle allocation output of the optimisation module 112 of the passenger flow analysis server 100, which is then sent to the vehicle allocation module 304 of the operation module 300. A record in the table expresses which vehicle with or without a passenger counting device should serve the service defined in the timetable. The example row in the table shows that service ID 426 should be operated by vehicle ID 2100 with a passenger counting device.
The table in Figure 13 shows the timetable change output of the optimisation module 112 of the passenger flow analysis server 100, which is then sent to the timetable management module 302 of the operation system 300. A record in the table expresses how the arrival and departure times of the timetable should be changed regarding the latest travel demand or the estimated origin-destination matrix in the estimated origin/destination matrix database 1015. The example row in the table shows that the departure time of service ID 425 at bus stop ID 325 should be delayed for a minute.
The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.
While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.
For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.
Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.
Throughout this specification, including the claims which follow, unless the context requires otherwise, the word "comprise" and "include", and variations such as "comprises", "comprising", and "including" will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.
It must be noted that, as used in the specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from "about" one particular value, and/or to "about" another particular value. When such a range is expressed, another embodiment includes from the one particular value and/or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent "about," it will be understood that the particular value forms another embodiment. The term "about" in relation to a numerical value is optional and means for example +/- 10%.

Claims

A computer-implemented method of estimating a number of passengers in a mass transport system, the method comprising:
receiving, from a passenger device detector configured to detect passenger devices, passenger device information which includes an estimated number of passenger devices in a first vehicle of the mass transport system;

receiving, from a passenger counting device, a number of passengers in the first vehicle;

comparing the estimated number of passenger devices from the passenger device detector and the number of passengers from the passenger counting device to generate passenger estimation compensation information;

receiving, from a second passenger device detector configured to detect passenger devices, further passenger device information which includes an estimated number of passenger devices in a second vehicle of the mass transport system; and

using the passenger estimation compensation information to estimate a number of passengers in the second vehicle from the further passenger device information.
The method of any preceding claim further comprising: determining, from the passenger device information, when and/or where each passenger device embarks and/or disembarks from the first and second vehicles.
The method of any preceding claim wherein the passenger device information includes one or more device identifiers, and the method comprises determining when one or more of the device identifiers has changed.
The method of any preceding claim further comprising receiving location information associated with the first and second vehicles and/or the first and second probe monitoring devices.
The method of claim 4 further comprising: using the location information to determine when each passenger device embarks and disembarks from the first and/or second vehicles and/or when one or more of the device identifiers has changed.
The method of claims 4 or 5 further comprising: using the location information to determine a plurality of estimated numbers of passengers corresponding to each of a plurality of route sections travelled by the first vehicle.
The method of claim 6 wherein generating the passenger estimation compensation information comprises generating section compensation information corresponding to each of the route sections travelled by the first vehicle.
The method of claim 7 wherein estimating the number of passengers in the second vehicle comprises using the section compensation information and the further passenger device information, for each route section, to estimate a number of passengers in the second vehicle for each route section travelled by the second vehicle.
The method of any preceding claim wherein the passenger estimation compensation information comprises a ratio between the number of passengers detected by the passenger counting device and the estimated number of passenger devices detected by the passenger device detector.
The method of any preceding claim, wherein the passenger estimation compensation information is generated by comparing passenger device information received from a plurality of passenger device detectors associated with a plurality of vehicles of the mass transport system, and passenger numbers received from a plurality of passenger counting devices associated with the plurality of vehicles;
The method of any preceding claim, further comprising: allocating vehicles with passenger counting devices to selected routes of the mass transport system, and allocating vehicles which do not have passenger counting devices to other selected routes of the mass transport system based on the estimated number of passengers.
The method of any preceding claim, further comprising: determining updated timetable information based on the estimated number of passengers.
The method of any preceding claim, further comprising: transmitting an update signal to the passenger device detector and/or the second passenger device detector based on the passenger device information from that passenger device detector, the update signal being configured to:
increase or decrease a monitoring frequency of the respective probe monitoring device,

wherein the monitoring frequency determines a frequency at which the respective passenger device detector assesses a number of passenger devices that it can detect;
and/or

change a scanning mode of the passenger device detector and/or the second passenger device detector between an active scanning mode and a passive scanning mode, or vice versa.
A passenger flow analysis server for estimating a number of passengers in a mass transport system, wherein the passenger flow analysis server is configured to:
receive, from a passenger device detector configured to detect passenger devices, passenger device information which includes an estimated number of passenger devices in a first vehicle of the mass transport system;

receive, from a passenger counting device, a number of passengers in the first vehicle;

compare the estimated number of passenger devices from the passenger device detector and the number of passengers from the passenger counting device to generate passenger estimation compensation information;

receive, from a second passenger device detector configured to detect passenger devices, further passenger device information which includes an estimated number of passenger devices in a second vehicle of the mass transport system; and

use the passenger estimation compensation information to estimate a number of passengers in the second vehicle from the further passenger device information.
A system for estimating numbers of passengers of a mass transport system, the system comprising:
a plurality of passenger device detectors configured to detect passenger devices on respective vehicles of the mass transport system to generate passenger device information which include an estimated number of passenger devices in each respective vehicle;

one or more passenger counting devices configured to count a number of passengers in one or more of the respective vehicles of the mass transport system; and

the passenger flow analysis server according to claim 14.