A method for predicting passenger flow at ground bus stops

By establishing a simultaneous equation model of passenger flow at bus stops and the time distance from the front of the bus, combined with multi-source data, the endogenous problems in passenger flow prediction at bus stops in the existing technology are solved, and more accurate and reliable passenger flow prediction is achieved, and buses are supported. Effective configuration of facilities and line optimization.

CN114611761BActive Publication Date: 2024-11-08SOUTHEAST UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210140020.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-02-16
Publication Date
2024-11-08
Estimated Expiration
2042-02-16

AI Technical Summary

Technical Problem

The existing bus passenger flow prediction technology fails to effectively consider the influencing factors of passenger flow at the ground bus stop level, especially the endogenous problem between the front distance and passenger flow, resulting in inconsistent and biased prediction results.

Method used

Using multi-source data combination, we establish a simultaneous equation model of bus stop passenger flow and front distance. By obtaining and processing bus card swiping data, vehicle entry and exit data, POI data, road network data, etc., we establish a data set of influencing factors and use it. The three-stage least squares estimation model considers the influencing factors of passenger flow at the site level and the endogeneity of the front distance.

Benefits of technology

A more reliable passenger flow forecast at bus stops is achieved, which can accurately reflect the impact of line adjustment and new sites on passenger flow, providing a basis for bus facilities configuration and location selection, and improving the accuracy and reliability of the model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114611761B_ABST
    Figure CN114611761B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for predicting passenger flow at ground bus stops, comprising the following steps: (1) obtaining bus card swiping data and counting passenger flow at existing bus stops; (2) obtaining vehicle entry and exit data and calculating average headway time of existing bus stops; (3) obtaining point of interest (POI) data, residential area data, road network data, and land use data; (4) establishing a data set of factors affecting passenger flow; (5) establishing a data set of factors affecting headway time; (6) establishing a simultaneous equation model; (7) using a three-stage least squares estimation model; (8) establishing a data set of factors affecting predicted stops; and (9) using the estimated passenger flow model for prediction. The present invention combines multi-source data and considers the influence of endogeneity in the model, and can accurately predict bus passenger flow at the stop level, which is helpful for effectively configuring bus stop facilities and evaluating the rationality of the location selection of new bus stops.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of public transport passenger flow prediction, and in particular to a method for predicting passenger flow at a ground public transport station. Background Art

[0002] As an important green travel mode, ground public transportation is of great significance to alleviating traffic congestion and achieving sustainable development of transportation. Passenger flow forecasting can provide a basis for the construction, adjustment and planning of public transportation, and is of great significance for determining and evaluating public transportation operation management strategies. Most existing studies focus on bus passenger flow forecasting at the line level, and less on the station level. Bus station passenger flow forecasting can predict the impact of line changes, new stations, etc. on bus passenger flow from the station level, which is conducive to the effective configuration of bus station facilities and the rationality of evaluating the location selection of new bus stations.

[0003] The factors affecting passenger flow at bus stops are complex. In order to obtain more reliable prediction results, it is necessary to combine multiple sources of data and incorporate multiple factors into the prediction model. Headway is an important factor affecting passenger flow at a station, but in the regression model, passenger flow and headway will produce endogeneity due to the mutual causality, resulting in biased and inconsistent model estimation results. However, the existing bus passenger flow prediction does not consider the endogeneity problem. Summary of the invention

[0004] Purpose of the invention: To provide a bus stop passenger flow prediction method which combines multi-source data, takes into account the endogeneity of the model, is simple and convenient to apply, and has more reliable results.

[0005] Technical solution: The present invention provides a method for predicting passenger flow at a ground bus stop, comprising the following steps:

[0006] (1) Obtaining bus card swiping data and counting passenger flow at each existing bus stop; the bus card swiping data includes the passenger's card swiping time and boarding stop and location information;

[0007] (2) obtaining vehicle entry and exit data and calculating the average headway time of each existing bus stop; the vehicle entry and exit data includes the station name, station latitude and longitude, and vehicle entry time;

[0008] (3) Obtain POI data, residential area data, road network data, land use data, etc.;

[0009] (4) Process the data and establish a data set of factors affecting passenger flow within the attraction range of the bus station (set to 500 meters, based on the reasonable distance that passengers can walk to the station). The factors affecting passenger flow include land use factors, transportation infrastructure, station attributes, and socioeconomic factors;

[0010] (5) Process the data and establish a dataset of factors affecting headway time at bus stops. The factors affecting headway time include the number of intersections within 500 meters of the stop and the average operating cost of the stop.

[0011] (6) Establish a simultaneous equation model consisting of the bus stop passenger flow regression equation and the headway regression equation;

[0012] (7) The three-stage least squares method is used to estimate the simultaneous equation model and obtain the passenger flow equation;

[0013] (8) Establish a data set of factors affecting passenger flow within a set range around the station to be predicted, substitute it into the passenger flow equation estimated in (7), and obtain the prediction result.

[0014] Furthermore, in step (2), the calculation steps of the average headway time of the station are as follows: first, the data is sorted according to the station name, station longitude and latitude, and vehicle entry time, then the entry time difference of all adjacent vehicles entering the same station is calculated, and finally, the average entry time difference of the vehicles at the same station is taken to obtain the average headway time of the station.

[0015] Furthermore, in step (3), POI data is obtained through the electronic map API interface, including catering service data, shopping service data, medical institution data, government agency data, science, education and culture service data, financial insurance service data, company data, parking lot data, and subway station data; residential area data is obtained from real estate websites through web crawler technology, including residential area name, longitude and latitude, average house price and total number of households in the residential area; road network data comes from the open source map OpenStreet Map; and land use data comes from the public data set EULUC.

[0016] Furthermore, in step (4), the variables of land use factors in the influencing factor data set include residential land density, commercial land density, industrial land density, public service land density, and land use mix; transportation infrastructure includes the number of subway stations, the number of public bicycle stations, the number of other bus stations, the number of parking lots, and the length of low-grade road networks; station attributes include headway, whether it is the first or last station, the number of station lines, and the distance of the station from the city center; socioeconomic factors include housing prices, the number of households, and the number of jobs.

[0017] Furthermore, in step (4), the land use mixing degree is represented by the entropy index, and the calculation formula is:

[0018]

[0019] Where t is the total number of land use types within the site attraction range, P w It is the proportion of the wth type of land area to the site attraction range.

[0020] Furthermore, in step (4), the number of positions is replaced by the number of POI points that can provide positions, and the position POI data is obtained by merging POI data of categories such as catering services, shopping services, medical institutions, government agencies, financial insurance services, and companies and enterprises.

[0021] Furthermore, in steps (4) and (5), the data processing is performed by ArcGIS.

[0022] Furthermore, in step (5), the calculation formula of the average site operating cost is:

[0023]

[0024] Among them, s is the total number of site lines, k is the kth line of the site, l k is the length of the kth line, b k is the number of trains in operation on the kth route, d k is the number of stations on the kth line, and C is the unit mileage cost.

[0025] Furthermore, in step (6), the expression of the simultaneous equation model is as follows:

[0026]

[0027]

[0028] Where: Y is the passenger flow at the bus stop, H is the headway time at the bus stop, n is the total number of factors affecting passenger flow, X i is the i-th passenger flow influencing factor, α i is the coefficient corresponding to the i-th passenger flow influencing factor, m is the total number of factors affecting headway, Z j is the jth headway influencing factor, β j is the coefficient corresponding to the influencing factor of the station headway, and v and u are error terms.

[0029] Furthermore, in step (7), the estimation steps of the three-stage least squares method are as follows: first, the order condition and rank condition are used to test the identifiability of each equation in the simultaneous equation model, and then the two-stage least squares method is used to estimate each equation in the simultaneous equation model to estimate the covariance matrix of the disturbance term of the entire simultaneous equation model, and finally the generalized least squares estimation is performed on the entire simultaneous equation model to obtain the estimation result.

[0030] Beneficial effects:

[0031] (1) The present invention is different from the passenger flow prediction at the line and system levels. The bus passenger flow prediction at the station level can more accurately reflect the impact of factors such as line adjustment and new station construction on passenger flow, provide a basis for the effective configuration of bus station facilities, evaluate the rationality of the location selection of new bus stations, and optimize the adjustment of line stop stations. It can effectively predict the passenger flow at the ground bus station level;

[0032] (2) The factors affecting bus passenger flow are relatively complex. Comprehensively considering various factors affecting bus passenger flow can improve the accuracy of the model. The present invention establishes an influencing factor data set based on four categories: land use factors, transportation infrastructure, station attributes, and socio-economic factors. The impact of station headway on passenger flow and the endogeneity of the two are considered. The factors affecting station passenger flow are considered from multiple aspects, and a simultaneous equation model is used to solve the endogeneity problem, making the prediction results more reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0034] The technical solution of the present invention is described in detail below with reference to the accompanying drawings.

[0035] A method for predicting passenger flow at ground bus stops, the implementation steps are as follows: Figure 1 As shown in the example, the area and stations under the jurisdiction of Nanjing Public Transport Group are selected to predict the hourly passenger flow during the morning peak (7-9 o'clock) of the newly built station. The specific implementation steps are as follows:

[0036] Step 1: Obtain the bus card swiping data in the study area and count the passenger flow of each existing bus stop according to the station name and card swiping time.

[0037] Step 2: Obtain vehicle entry and exit data in the study area, sort the data according to station name, station longitude and latitude, and vehicle entry time, calculate the entry time difference of all adjacent vehicles entering the same station, and take the average entry time difference of vehicles at the same station to obtain the average headway time of each existing bus station.

[0038] Step 3: Obtain POI data, residential area data, road network data, land use data, etc. POI data is obtained from the Amap API, including catering service data, shopping service data, medical institution data, government agency data, science, education and culture service data, financial insurance service data, company data, parking lot data, and subway station data; residential area data is obtained from real estate websites through web crawler technology, including residential area name, longitude and latitude, average house price and total number of households in the residential area; road network data comes from the open source map Open Street Map; land use data comes from the public data set EULUC.

[0039] Step 4: Establish a data set of factors affecting passenger flow, including four categories: land use factors, transportation infrastructure, station attributes, and socioeconomic factors. Import the bus station data and the data obtained in step 3 into ArcGIS, establish a 500-meter buffer for all bus stations, and merge the catering service data, shopping service data, medical institution data, government agency data, financial insurance service data, and company data in the POI data into job POI data.

[0040] The spatial connection and neighbor analysis commands of ArcGIS are used to count the values ​​of various influencing factors in the buffer zone.

[0041] The land use mix degree is calculated using the following formula:

[0042]

[0043] Among them, land use types are divided into five categories: residential, commercial, industrial, public services and others, so t is 5; the site attraction range area is the area of ​​the 500-meter buffer zone of the site.

[0044] The variables of the passenger flow influencing factors dataset are shown in Table 1:

[0045] Table 1

[0046]

[0047] Step 5: Establish a dataset of factors affecting headway time. Use the road network data and the 500-meter buffer zone of the bus station obtained in step 4 to count the number of intersections in each buffer zone, and use the following formula to calculate the average operating cost of each station.

[0048]

[0049] Step 6: Establish a simultaneous equation model consisting of the bus passenger flow regression equation and the headway time regression equation. In order to make the passenger flow conform to the normal distribution as much as possible, the passenger flow at the station is taken as logarithm. Since there are stations with a passenger flow of 0 during the morning peak, the logarithm form of ln(1+Y) is used. The expression of the model is as follows:

[0050]

[0051]

[0052] Step seven, use the three-stage least squares method to estimate the simultaneous equation model: first, use the order condition and rank condition to test the identifiability of the simultaneous equation model, then use the two-stage least squares method to estimate each equation in the simultaneous equation model, then estimate the covariance matrix of the disturbance term of the entire simultaneous equation model, and finally perform generalized least squares estimation on the entire simultaneous equation model.

[0053] The estimated passenger flow equation is:

[0054] ln(1+Y)=-0.0638H+0.7815X 1 -0.4751X 2 +0.0341X 3 +0.1574X 4 +0.2424X 5 -0.0323X 6 +0.0211X 7 -0.0166X 8 +0.0017X 9 -0.0021X 10 +0.2481X 11 +0.0911X 12 -0.0042X 13 +0.05676X 14 +0.00005X 15 +0.0002X 16 +1.8657

[0055] Step 8. Obtain the data set of passenger flow influencing factors of the newly built or adjusted stations according to steps 3 and 4, calculate the headway of the newly built station according to the departure schedule and the estimated time of arrival of the vehicle at the newly built station, and calculate the headway of the adjusted station according to the vehicle entry and exit data, the new line departure schedule and the estimated time of arrival of the vehicle at the station.

[0056] Step nine, substitute the values ​​of the influencing factor data set obtained in step eight into the passenger flow equation estimated in step seven, and restore the result to the passenger flow value, that is, obtain the passenger flow prediction result of the new site or the adjusted site.

[0057] The embodiment of the present invention further provides a device for predicting passenger flow at a ground bus stop, comprising: a processor and a memory, and the number of the processor and the memory can be one or more.

[0058] As a computer-readable storage medium, the memory can be used to store software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the method for predicting passenger flow at ground bus stops in any embodiment of the present application. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory may further include a memory remotely arranged relative to the processor 60, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0059] The processor executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory, that is, realizes the above-mentioned method for predicting passenger flow at ground bus stops.

[0060] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for predicting passenger flow at a ground bus stop provided by an embodiment of the present invention is implemented.

[0061] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (Random Access Memory, RAM), a read-only memory (Read-Only Memory, ROM), an erasable programmable read-only memory (Erasable Programmable ROM, EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0062] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0063] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0064] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or combinations thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0065] The embodiments are only for illustrating the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for predicting passenger flow at a ground bus stop, characterized in that: The following steps are involved: (1) Obtaining bus card swiping data and counting passenger flow at each existing bus stop; the bus card swiping data includes the passenger's card swiping time, boarding stop and location information; (2) Obtaining bus entry and exit data and calculating the average headway time of each existing bus stop; the bus entry and exit data includes the station name, station latitude and longitude, and vehicle entry time; (3) Obtaining point of interest (POI) data, residential area data, road network data, and land use data; (4) Establish a data set of factors affecting passenger flow within the attraction range of each existing bus station, including land use factors, transportation infrastructure, station attributes, and socioeconomic factors; (5) Establishing a data set of factors affecting headway time for each existing bus stop, wherein the factors affecting headway time include the number of intersections within a set range around the bus stop and the average operating cost of the bus stop; (6) Establish a simultaneous equation model consisting of the bus stop passenger flow regression equation and the bus stop headway regression equation; (7) The three-stage least squares method is used to estimate the simultaneous equation model and obtain the passenger flow equation; (8) Establishing a data set of factors affecting passenger flow within the attraction range of the site to be predicted, substituting it into the passenger flow equation estimated in step (7) to obtain the prediction result; The calculation formula for the average site operating cost in step (5) is: Among them, s is the total number of site lines, k is the kth line of the site, l k is the length of the kth line, b k is the number of trains in operation on the kth route, d k is the number of stations on the kth line, and C is the unit mileage cost; The expression of the simultaneous equation model is as follows: Where: Y is the passenger flow at the bus stop, H is the headway time at the bus stop, n is the total number of factors affecting passenger flow, X i is the i-th passenger flow influencing factor, α i is the coefficient corresponding to the i-th passenger flow influencing factor, m is the total number of factors affecting headway, Z j is the jth headway influencing factor, β j is the coefficient corresponding to the influencing factor of the station headway, v and u are error terms; The estimation steps of the three-stage least squares method are: (A) Use order conditions and rank conditions to test the identifiability of each equation in the simultaneous equation model; (B) Each equation in the simultaneous equation model is estimated using the two-stage least squares method; (C) Based on (B), estimate the covariance matrix of the disturbance term of the entire simultaneous equation model; (D) Perform generalized least squares estimation on the entire simultaneous equation model.

2. A method for predicting passenger flow at a ground bus stop according to claim 1, characterized in that: In step (2), the calculation steps of the average headway time of the station are as follows: first, the bus vehicle entry and exit data are sorted according to the station name, station latitude and longitude and vehicle entry time; then, the entry time difference of all adjacent vehicles entering the same station is calculated; finally, the vehicle entry time difference at the same station is averaged to obtain the average headway time of the station.

3. The method for predicting passenger flow at a ground bus stop according to claim 1, characterized in that: In the step (3), POI data is obtained through an electronic map API interface, including catering service data, shopping service data, medical institution data, government agency data, science, education and culture service data, financial insurance service data, company data, parking lot data, and subway station data; residential community data is obtained from real estate websites through web crawler technology, including residential community name, longitude and latitude, average house price and total number of households in the residential community; road network data comes from the open source map Open Street Map; and land use data comes from the public data set EULUC.

4. The method for predicting passenger flow at a ground bus stop according to claim 1, characterized in that: In step (4), the land use factors include residential land density, commercial land density, industrial land density, public service land density, and land use mix; the transportation infrastructure includes the number of subway stations, the number of public bicycle stations, the number of other bus stations, the number of parking lots, and the length of secondary roads and branch roads; the station attributes include headway, whether it is the first or last station, the number of station lines, and the distance of the station from the city center; the socioeconomic factors include housing prices, the number of households, and the number of jobs.

5. The method for predicting passenger flow at a ground bus stop according to claim 2, characterized in that: The land use mixture degree is expressed by the entropy index, and the calculation formula is: Where t is the total number of land use types within the site attraction range, P w It is the proportion of the wth type of land area to the site attraction range.

6. A device for predicting passenger flow at a ground bus stop, characterized in that: The device includes: a memory and one or more processors; the memory is used to store one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the method for predicting passenger flow at a ground bus stop as described in any one of claims 1-5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for predicting passenger flow at a ground bus stop as described in any one of claims 1-5 is implemented.