Method and system for single-line planning of bus route in designated area
Through Internet commuting data and machine learning models, the travel probability of citizens choosing to travel for public transportation is predicted, combined with line constraints and individual traffic attraction, a bus line planning scheme is generated, which solves the problems of complexity and low accuracy of bus line network planning and improves planning accuracy and fit.
Patent Information
- Application Number
- CN202411527233.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-30
- Publication Date
- 2024-11-26
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, the estimation accuracy of the OD volume of bus travel is low, and the urban bus network planning is complex, and there is a lack of effective solutions.
A single-line planning method for bus routes in designated areas, including data preparation, data mapping, travel prediction and line search steps, predict the travel probability of citizens choosing bus travel through Internet commuting data and machine learning models, and search is consistent with line constraints and individual traffic. Attract a number of lines and compare them with the existing bus network to generate a reference line planning scheme.
It improves the accuracy of bus line planning, is more in line with citizens' travel needs, and effectively solves the problems of complexity and low accuracy of bus line planning.
Smart Images

Figure CN119026782A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of public transportation, and in particular to a method and system for planning a single line of a public transportation route in a designated area. Background Art
[0002] Bus network planning is an extremely important topic in the bus industry, and there are many related academic monographs and papers studying it.
[0003] The current practice of line network planning is to first collect basic data such as trip OD (origin and destination), and then lay out the lines according to a certain theory. The theories include: 1) laying out one by one according to the size of OD volume (passenger flow between OD, referred to as OD volume); 2) laying out in layers according to urban structure and OD volume: such as express lines, trunk lines, straight lines, etc.
[0004] Data on the OD volume of public transportation trips is not easy to obtain. Existing methods such as mobile phone signaling and manual surveys are not accurate enough. There is no better way to estimate the potential OD of public transportation trips. Usually, a ratio is given based on experience based on the total OD volume, which is difficult to meet the needs of network planning.
[0005] In addition, the planning of the bus network of the entire city is very complex, involving many factors, such as travel demand, road conditions, station distribution, etc. So far, there is no mature and universal method that can be applied to the planning of the bus network of the entire city. In fact, when planning the network, computer assistance and manual analysis are used, and currently the component of manual experience is higher.
[0006] In recent years, Internet traffic big data has gradually matured, providing a better data source for line network planning. For example, Internet commuting big data uses the location information of mobile phone apps to analyze users' work and residence, commuting methods, commuting time and distance. This data is usually given in the form of a grid, generally including: departure grid, arrival grid, travel time (half-hour granularity), travel mode (rail, bus, driving, cycling, walking, etc.), travel distance, travel time, travel path and travel volume of each travel mode. The grid is a grid that divides the city into rectangular grids, such as 100x100 meter grids, 150x150 meter grids, etc.
[0007] The main problems with current technology are: 1) The estimation accuracy of potential bus travel OD is low. Generally, a part of the overall OD volume is taken as bus travel based on experience, and the estimation of the overall OD volume is not accurate; 2) The city’s bus network planning is too complicated and there is no good solution for now. Summary of the invention
[0008] The technical problem to be solved by the embodiments of the present invention is to provide a method and system for planning a single line of bus routes in a specified area, so as to solve the planning problem of the urban bus network.
[0009] In order to solve the above technical problems, an embodiment of the present invention proposes a method for planning a single bus route in a designated area, comprising: Data preparation steps: Collect bus network data, Internet commuting data, bus operation data, and the areas that the routes must pass through; Data mapping step: aggregate individual traffic volumes in the Internet commuting data to bus stops; Travel prediction steps: Build a machine learning model to predict the probability of citizens choosing public transportation; Route search steps: search for routes that meet linear constraints and individual traffic attraction based on the areas that the routes must pass through; Comparative planning steps: select several routes that attract the largest individual traffic volume from the searched routes, compare each route with the existing bus network, and obtain a reference route planning solution.
[0010] Accordingly, an embodiment of the present invention further provides a single-line planning system for bus routes in a specified area, comprising: Data preparation module: collect bus network data, Internet commuting data, bus operation data, and the areas that the routes must pass through; Data mapping module: aggregates individual traffic volumes in Internet commuting data to bus stops; Travel prediction module: Establish a machine learning model to predict the probability of citizens choosing public transportation; Route search module: searches for routes that meet linear constraints and individual traffic attraction based on the areas that the routes must pass through; Comparative planning module: select several routes that attract the largest individual traffic volume from the searched routes, compare each route with the existing bus network, and obtain a reference route planning solution.
[0011] The beneficial effect of the present invention is that by utilizing the method of the present invention, a better bus route planning within a designated area can be given to better meet the public transportation travel needs of citizens. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 Schematic diagram of a target grid and its surrounding grids according to an embodiment of the present invention.
[0013] Figure 2 It is a structural diagram of a machine learning model according to an embodiment of the present invention.
[0014] Figure 3Schematic diagram of a starting area, a passing area, an end area and a buffer zone according to an embodiment of the present invention.
[0015] Figure 4 This is a schematic diagram of adding a site to a line according to an embodiment of the present invention.
[0016] Figure 5 It is a flowchart of a method for planning a single bus route in a designated area according to an embodiment of the present invention. DETAILED DESCRIPTION
[0017] It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The present invention is further described in detail below in conjunction with the drawings and specific embodiments.
[0018] In the embodiments of the present invention, if there are directional indications (such as up, down, left, right, front, back, etc.), they are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.
[0019] In addition, in the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features.
[0020] The development of a city is usually the development and change of each region. Line planning focusing on local areas has many practical application scenarios, such as the occupancy of new communities, the opening of new subway lines, and the identification and adjustment of inefficient lines.
[0021] The present invention firstly establishes a machine learning model for citizens to choose public transportation through Internet travel big data. On this basis, a single-line planning of bus routes in a specified area is given by searching.
[0022] Please refer to Figure 5 The single-line planning method for a bus route in a specified area in an embodiment of the present invention includes a data preparation step, a data mapping step, a travel prediction step, a route search step, and a comparison planning step.
[0023] Data preparation steps: Collect bus network data, Internet commuting data, bus operation data, and the areas that the route must pass through. The data required for route planning in the embodiment of the present invention are: 1) Bus network data, including static information of routes, stations and depots; 2) Internet commuting data, assuming a granularity of 100x100 meters; 3) Bus operation data, including time periods between routes and stations (time period: refers to a period of time, such as 8:00-8:30 is one time period, 8:30-9:00 is another time period, and half an hour is the granularity of the time period. The granularity is consistent with the Internet commuting data) OD; 4) The areas that must be passed through on a manually specified route (starting point area, end point area, and passing area) are usually presented in the form of polygons.
[0024] Data mapping step: aggregate individual traffic volumes in the Internet commuting data to bus stops. That is, map the non-bus part of the Internet commuting data to bus stops, and add new stops if necessary.
[0025] Before route planning, first aggregate the individual traffic volumes in the commuting data to bus stops: a) Find the bus stops in the target OD grid and its surrounding adjacent grids; Figure 1 , the shadow grid is the target grid, and the rest are the surrounding grids; b) If there is only one bus stop around both the O grid and the D grid, the individual traffic volume of the OD grid is mapped to the bus stop pair; c) If there are multiple bus stops around the O grid or D grid, select the station pair with the shortest driving path between the stations; d) If there is no bus stop around the O grid or D grid, the departure volume and attraction volume of the grid are recorded; if the cumulative departure volume or attraction volume of a certain grid is large enough (exceeding the preset threshold) and there are roads around, a new bus stop is added on the roadside and the corresponding OD volume is mapped to the newly added OD station pair; Note that in the above steps, sites with the same name are treated as the same site when searching for sites around the grid, but are treated separately when searching for the shortest path. Usually, there is a site with the same name on each side of the road.
[0026] Travel prediction step: The data preparation step provides citizens' all-mode commuting travel data. For the same grid travel OD, there may be multiple modes, such as driving, rail, bus, cycling and walking. Based on the historical travel data of each travel mode under the grid, a machine learning model can be established to predict the probability of citizens choosing public transportation.
[0027] A neural network model with a multi-layer perceptron was selected. The input of the model includes peak periods (morning peak, off-peak and evening peak. The peak periods in different cities are not the same. The common ones are 7:00-9:00 for the morning peak, 17:00-19:00 for the evening peak, and the rest of the time is off-peak), travel time, travel distance, walking distance for bus travel, and number of transfers for each travel mode. The output of the model is the probability of passengers choosing bus travel.
[0028] This model is called NNM. Figure 2 The specific model training process is not the focus of the present invention and will not be described here.
[0029] Route search steps: Search for routes that meet linear constraints and individual traffic attraction based on the route's must-pass areas. That is, obtain the shortest path set between sites from the site set of the must-pass area, and then search from the shortest path set. For example, if there are two must-pass areas, the first must-pass area has sites a and b; the second must-pass area has sites c and d. Then ac, ad, bc, bd are route candidates. If there are multiple must-pass areas, similar processing is performed.
[0030] Assume that the starting area of the route contains m sites, represented by , ; The end area of the line contains n sites, represented by , ; The route has a passing area (the situation of having multiple passing areas is handled similarly), and the area has x sites, represented by , The purpose of this step is to find the stations from the starting area Departure, passing through regional stations , arrive at the destination area station , and select other passing stations in the surrounding area to form a route. The route meets the linear constraint and the individual traffic attraction constraint (individual traffic attraction refers to the number of cars or bicycles transferred to buses). The surrounding stations refer to the line segments connecting the must-pass areas as a buffer (such as a 5 km buffer). The OD stations are optional passing stations for stations within the buffer. Please refer to Figure 3 A GIS (Geographic Information System) buffer in this field is an area created around a GIS feature (point, line or surface) where the distance from the region boundary to the GIS feature is the same. This distance value is called the buffer size.
[0031] The route search procedure is to first search for each site on the uplink route, and then reversely connect the sites with the same name on the opposite side of the road as the downlink route through the shortest path.
[0032] The specific implementation steps of the uplink line search are: 1) Sort the OD station pairs in the buffer zone according to their individual traffic volume from large to small, denoted as .in N is the number of OD site pairs; 2) From , , Select one station from each station set as the initial form of the line. There are a total of indivual; 3) Select one of the combinations and record it as , for this combination, the initial route is formed according to the shortest path between sites ; 4) Check the wiring Whether the linear constraints are met, there are three results: a. Not satisfied, and adding more stations will not satisfy it, for example, the line length or the non-linear coefficient between any two stations (the actual length between the two stations divided by the straight-line distance between the two stations) has exceeded the established threshold, and continuing to add stations will only make the line longer or the non-linear coefficient larger, then return 2); b. Not satisfied, but may be satisfied after adding passing stations, for example, if the line length is less than the set length, then proceed to the next step; c. If satisfied, proceed to the next step; 5) Inspection Whether the individual traffic attraction constraint is met, there are three results: a. If it is not satisfied, the following site may be added to satisfy it, then proceed to the next step; b. If the linear constraint is also satisfied, then Put it into the result set as a candidate route L , return to 3); otherwise proceed to the next step; 6) Try to change the site Add to line Middle: According to the spatial location of the site, decide between which two sites the site is added; 7) Check the wiring Whether the linear constraints are met, there are three results: a. The condition is not satisfied, and the condition will not be satisfied again if more passing stations are added. For example, the line length or the non-linear coefficient between any two stations has exceeded the predetermined threshold. Return to step 6 and try the next passing station pair. b. Not satisfied, but may be satisfied after adding passing stations, for example, the line length is less than the set length, proceed to the next step; c. If satisfied, proceed to the next step; 8) Inspection Whether the individual traffic attraction constraint is met, there are three results: a. If it is not satisfied, adding a site later may satisfy it, so proceed to the next step; b. Already satisfied. If the linear constraint is also satisfied, then Put it into the result set as a candidate route L, return to 3); otherwise, next step; 9) Go back to step 6) until every site pair in the buffer is traversed. ; 10) Go back to 3) until each combination.
[0033] The method for checking whether the route meets the individual traffic passenger flow transfer in the above steps 5) and 8) is: for each OD station pair of the route (including up and down), the individual traffic data collected in the data mapping step can be used to obtain the commuting OD volume of each travel mode under the OD; then according to the travel prediction step NNM The model estimates the travel probability of bus trips, so it can predict the individual traffic volume that the OD pair of the line can attract; the summary of these attracted individual traffic volumes can be used to obtain the passenger flow that the line attracts individual traffic.
[0034] In step 6) above, the method of adding an OD site pair to the line is to add the O site and the D site to the line separately in a way that minimizes the line non-linear coefficient: Find the station closest to the target station (station O or station D); If it is closest to the first station of the line, add it after the first station; If it is closest to the end station of the line, add before the end station; If it is closest to a certain station in the middle, it is decided whether to add it before or after this station based on the length of the line after adding it.
[0035] Please refer to Figure 4 For example, the line currently has 4 stations: station 1, station 2, station 3, station 4; to add station A to the line: 1) Find the site closest to site A, which is site 3. 2) Calculate the path length of site 2-site A-site 3-site 4, and the path length of site 2-site 3-site A-site 4. It can be seen that the former length is smaller, so add site A before site 3.
[0036] The present invention searches for routes that meet linear constraints and individual traffic attraction according to the designated route must-pass area. The designated area includes the route starting area, the route end area and one or more optional passing areas. The linear constraints include non-linear coefficient, route length, number of stations, distance between the first and last stations and the station, etc.
[0037] Comparative planning steps: Form a reference route planning scheme.
[0038] Select several routes that attract the largest individual traffic volume from the solutions searched in the previous step.
[0039] Each route is compared with the existing bus network to obtain a route planning scheme: the OD station pairs are compared with each route in the existing bus network one by one, and the routes containing the same OD station pairs are found as similar routes; the planned routes and the corresponding similar routes together become a bus route planning scheme, which serves as the basis for subsequent network planning.
[0040] The present invention compares the searched routes that meet the constraints with the existing bus routes, finds out the routes that have overlapping OD station pairs with the existing buses, and uses this as the basis for subsequent line network adjustment plans.
[0041] The bus route single-line planning system for a designated area in an embodiment of the present invention includes: Data preparation module: collect bus network data, Internet commuting data, bus operation data, and the areas that the routes must pass through; Data mapping module: aggregates individual traffic volumes in Internet commuting data to bus stops; Travel prediction module: Establish a machine learning model to predict the probability of citizens choosing public transportation; Route search module: searches for routes that meet linear constraints and individual traffic attraction based on the areas that the routes must pass through; Comparative planning module: select several routes that attract the largest individual traffic volume from the searched routes, compare each route with the existing bus network, and obtain a reference route planning solution.
[0042] As an implementation method, the data mapping module performs aggregation according to the following steps: a) Find the bus stops in the target OD grid and its surrounding adjacent grids; b) If there is only one bus stop around both the O grid and the D grid, the individual traffic volume of the OD grid is mapped to the bus stop pair; c) If there are multiple bus stops around the O grid or D grid, select the bus stop pair with the shortest driving path between the stops; d) If there is no bus stop around the O grid or D grid, the departure volume and attraction volume of the grid are recorded; if the cumulative departure volume or attraction volume of a certain grid exceeds the preset threshold and there is a road around it, a new bus stop is added next to the road and the corresponding OD volume is mapped to the newly added OD bus stop pair.
[0043] As an implementation method, the machine learning model selects a neural network model of a multi-layer perceptron. The input of the machine learning model includes peak segments, travel time of each travel mode, travel distance, walking distance of bus travel and number of transfers. The output of the machine learning model is the travel probability of passengers choosing bus travel.
[0044] As an implementation method, the necessary areas of the route include a starting area, a passing area, and an end area. Assume that the starting area of the route includes m sites, represented by , ; The end area of the line contains n sites, represented by , ; The route has a passing area, which has x sites, represented by , ; The line search module searches according to the following steps: 1) Sort the OD station pairs in the buffer zone according to their individual traffic volume from large to small, denoted as ,in N is the number of OD site pairs; 2) From , , Select one station from each station set as the initial form of the line. There are a total of indivual; 3) Select one of the combinations and record it as , for this combination, the initial route is formed according to the shortest path between sites ; 4) Check the wiring Whether the linear constraints are met; 5) Check the wiring Whether the individual traffic attraction constraints are met. When both the linear constraints and the individual traffic attraction constraints are met, the route is put into the result set as a candidate route. L , return to step 3); 6) Align the site Add to line middle; 7) Check the wiring Whether the linear constraints are met; 8) Check the wiring Whether the individual traffic attraction constraints are met. When both the linear constraints and the individual traffic attraction constraints are met, the route is put into the result set as a candidate route. L , return to step 3); 9) Return to step 6) until every site pair in the buffer is traversed. ; 10) Return to step 3) until each combination.
[0045] As an implementation method, the line search module searches for lines For each OD station pair in the data mapping step, we first use the data collected in the data mapping step to obtain the commuting OD volume of each travel mode under the OD, then estimate the travel probability of public transportation, and then predict the route. The OD pair can attract individual traffic volume, and finally the individual traffic volume attracted is summarized to obtain the route Individual traffic attraction.
[0046] The present invention selects OD station pairs that can attract individual traffic from the must-pass area and its surroundings, adds the stations to the routes according to their spatial positions, and selects the routes that meet the linear shape and individual traffic attraction constraints as candidate planning routes.
[0047] The present invention uses Internet commuting big data to analyze potential public transportation travel needs, and then focuses on the single-line planning of bus routes in designated areas, solving the planning problem of urban bus networks through a divide-and-conquer approach.
[0048] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A single-line planning method for bus routes in a specified area, It is characterized in that include: Data preparation steps: Collect bus network data, Internet commuting data, bus operation data, and the areas that the routes must pass through; Data mapping step: aggregate individual traffic volumes in the Internet commuting data to bus stops; Travel prediction steps: Build a machine learning model to predict the probability of citizens choosing public transportation; Route search steps: search for routes that meet linear constraints and individual traffic attraction based on the areas that the routes must pass through; Comparative planning steps: select several routes that attract the largest individual traffic volume from the searched routes, compare each route with the existing bus network, and obtain a reference route planning solution.
2. The method for planning a single bus route in a designated area as claimed in claim 1, It is characterized in that The data mapping steps include: a) Find the bus stops in the target OD grid and its surrounding adjacent grids; b) If there is only one bus stop around both the O grid and the D grid, the individual traffic volume of the OD grid is mapped to the bus stop pair; c) If there are multiple bus stops around the O grid or D grid, select the bus stop pair with the shortest driving path between the stops; d) If there is no bus stop around the O grid or D grid, the departure volume and attraction volume of the grid are recorded; if the cumulative departure volume or attraction volume of a certain grid exceeds the preset threshold and there is a road around it, a new bus stop is added next to the road and the corresponding OD volume is mapped to the newly added OD bus stop pair.
3. The method for planning a single bus route in a designated area as claimed in claim 1, It is characterized in that The machine learning model selects a neural network model of a multi-layer perceptron. The input of the machine learning model includes peak segments, travel time of various travel modes, travel distance, walking distance of public transportation and number of transfers. The output of the machine learning model is the travel probability of passengers choosing public transportation.
4. The method for planning a single bus route in a designated area as claimed in claim 1, It is characterized in that The necessary areas of the route include the starting area, the passing area and the end area. Assume that the starting area of the route includes m sites, represented by , ; The end area of the line contains n sites, represented by , ; The route has a passing area, which has x sites, represented by , ; The line search procedure includes the following steps: 1) Sort the OD station pairs in the buffer zone according to their individual traffic volume from large to small, denoted as ,in N is the number of OD site pairs; 2) From , , Select one station from each station set as the initial form of the line. There are a total of indivual; 3) Select one of the combinations and record it as , for this combination, the initial route is formed according to the shortest path between sites ; 4) Check the wiring Whether the linear constraints are met; 5) Check the wiring Whether the individual traffic attraction constraints are met. When both the linear constraints and the individual traffic attraction constraints are met, the route is put into the result set as a candidate route. L , return to step 3); 6) Align the site Add to line middle; 7) Check the wiring Whether the linear constraints are met; 8) Check the wiring Whether the individual traffic attraction constraints are met. When both the linear constraints and the individual traffic attraction constraints are met, the route is put into the result set as a candidate route. L , return to step 3); 9) Return to step 6) until every site pair in the buffer is traversed. ; 10) Return to step 3) until each combination.
5. The method for planning a single bus route in a designated area as claimed in claim 4, It is characterized in that In steps 5) and 8), the line For each OD station pair in the data mapping step, we first use the data collected in the data mapping step to obtain the commuting OD volume of each travel mode under the OD, then estimate the travel probability of public transportation, and then predict the route. The OD pair can attract individual traffic volume, and finally the individual traffic volume attracted is summarized to obtain the route Individual traffic attraction.
6. A single-line bus route planning system for a designated area, It is characterized in that include: Data preparation module: collect bus network data, Internet commuting data, bus operation data, and the areas that the routes must pass through; Data mapping module: aggregates individual traffic volumes in Internet commuting data to bus stops; Travel prediction module: Establish a machine learning model to predict the probability of citizens choosing public transportation; Route search module: searches for routes that meet linear constraints and individual traffic attraction based on the areas that the routes must pass through; Comparative planning module: select several routes that attract the largest individual traffic volume from the searched routes, compare each route with the existing bus network, and obtain a reference route planning solution.
7. The bus route single-line planning system for a designated area as claimed in claim 6, It is characterized in that The data mapping module performs aggregation according to the following steps: a) Find the bus stops in the target OD grid and its surrounding adjacent grids; b) If there is only one bus stop around both the O grid and the D grid, the individual traffic volume of the OD grid is mapped to the bus stop pair; c) If there are multiple bus stops around the O grid or D grid, select the bus stop pair with the shortest driving path between the stops; d) If there is no bus stop around the O grid or D grid, the departure volume and attraction volume of the grid are recorded; if the cumulative departure volume or attraction volume of a certain grid exceeds the preset threshold and there is a road around it, a new bus stop is added next to the road and the corresponding OD volume is mapped to the newly added OD bus stop pair.
8. The bus route single-line planning system for a designated area as claimed in claim 6, It is characterized in that The machine learning model selects a neural network model of a multi-layer perceptron. The input of the machine learning model includes peak segments, travel time of various travel modes, travel distance, walking distance of public transportation and number of transfers. The output of the machine learning model is the travel probability of passengers choosing public transportation.
9. The bus route single-line planning system for a designated area as claimed in claim 6, It is characterized in that The necessary areas of the route include the starting area, the passing area and the end area. Assume that the starting area of the route includes m sites, represented by , ; The end area of the line contains n sites, represented by , ; The route has a passing area, which has x sites, represented by , ; The line search module searches according to the following steps: 1) Sort the OD station pairs in the buffer zone according to their individual traffic volume from large to small, denoted as ,in N is the number of OD site pairs; 2) From , , Select one station from each station set as the initial form of the line. There are a total of indivual; 3) Select one of the combinations and record it as , for this combination, the initial route is formed according to the shortest path between sites ; 4) Check the wiring Whether the linear constraints are met; 5) Check the wiring Whether the individual traffic attraction constraints are met. When both the linear constraints and the individual traffic attraction constraints are met, the route is put into the result set as a candidate route. L , return to step 3); 6) Align the site Add to line middle; 7) Check the wiring Whether the linear constraints are met; 8) Check the wiring Whether the individual traffic attraction constraints are met. When both the linear constraints and the individual traffic attraction constraints are met, the route is put into the result set as a candidate route. L , return to step 3); 9) Return to step 6) until every site pair in the buffer is traversed. ; 10) Return to step 3) until each combination.
10. The bus route single-line planning system for a designated area as claimed in claim 9, It is characterized in that Line search module for line For each OD station pair in the data mapping step, we first use the data collected in the data mapping step to obtain the commuting OD volume of each travel mode under the OD, then estimate the travel probability of public transportation, and then predict the route. The OD pair can attract individual traffic volume, and finally the individual traffic volume attracted is summarized to obtain the route Individual traffic attraction.
Citation Information
Patent Citations
Station-based urban public traffic network optimized configuration method
CN101807222A
Bus rapid commuting route design method based on passenger flow OD big data
CN112288122A
Traffic mode selection prediction method based on computational graph
CN113326919A
Urban public transportation planning method
WO2023109720A1
Cited By
Bus network optimization method and system
CN120782077A
Bus line network optimization method and system
CN120782077B
