Urban bus station layout multi-objective optimization method based on genetic algorithm
Through a multi-objective optimization method based on genetic algorithm, the layout of bus stations is optimized, and the multi-objective problems of bus station coverage area and running time are solved, and the service quality and operation efficiency of the bus system are improved.
Patent Information
- Application Number
- CN202510353074.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-25
- Publication Date
- 2025-08-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, the research on bus stop optimization lacks optimization under the influence of multiple goals such as the coverage area of bus stops and the operating time of bus system, resulting in insufficient operational efficiency and service level of urban bus systems.
A multi-objective optimization method based on genetic algorithm is adopted to establish a bus stop layout model, and the bus stop layout is optimized, including optimization of site coverage and bus running time by selecting bus evaluation indicators and genetic algorithms.
The coverage rate and operation efficiency of bus stops have been improved, the service quality and operation efficiency of the bus network have been improved, and the model has good reliability and applicability.
Smart Images
Figure CN120409759A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of public transportation, and in particular to a multi-objective optimization method for urban bus station layout based on genetic algorithms. Background Art
[0002] Public transportation is a crucial urban infrastructure, and its development directly impacts citizens' travel convenience and well-being. With the acceleration of urbanization, the mismatch between urban spatial structure and bus network layout has led to a continuous decline in public transportation demand. To mitigate this phenomenon, it is necessary to systematically study the rationality of bus network and station layout, optimize bus station layout, and effectively improve the operational efficiency and service level of the public transportation system.
[0003] However, current research mostly focuses on optimization at the bus network level, with bus stop optimization being only one of the optimization objectives. There is a lack of research specifically targeting the optimization of bus stops, especially optimization problems under the influence of multiple objectives such as bus stop coverage area and bus system operating time. Summary of the Invention
[0004] The present invention provides a multi-objective optimization method for urban bus stop layout based on genetic algorithm, which solves the problems raised by the above background technology.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a multi-objective optimization method for urban bus stop layout based on genetic algorithms, the bus stop optimization method including establishing a multi-objective optimization model, selecting bus evaluation indicators, solving with genetic algorithms, evaluating optimization results, and determining optimization solutions. The multi-objective optimization model establishment includes establishing the relationship between bus supply and user demand in different areas and establishing the relationship between the spacing between each station and the acceptable walking distance. The bus evaluation indicator selection module includes a station coverage unit and a bus operation time unit.
[0006] Optionally, the establishment of a multi-objective optimization model specifically refers to: setting the first station of each line to , the last station is set to , suppose a line except the first and last stations are selected as the initial candidate stations , the candidate sites of the repeated lines are recorded as , called shared candidate sites, suppose there are k shared sites on the bus route, recorded in the set In the middle, all stations except the first and last stations are considered as candidate stations and optimized in the subsequent station selection; Establish a bus network, such as Figure 2 , the bus network includes 4 bus lines, and is set to ,use Indicates the The starting point of the line, Indicates line The terminal station, for any route , the order of intermediate stops is always from arrive , which are candidate sites, routes Terminals and routes The relationship between the +1 starting stations is , ; Assume that there are k shared candidate sites on the bus route, and the k shared candidate sites constitute the set of all shared sites, set as G k , set G k There are 16 shared candidate sites in the set G1, the first shared candidate site k=1, which can be used by routes 1 and 4 at the same time. 72} indicates the bus stop shared by bus routes 1 and 4. is defined as a decision variable, which is equal to 1 if candidate stop j is selected as a stop on route i, and equal to 0 otherwise. is defined as a decision variable, and is equal to 1 if shared stop k is selected as the station, and equal to 0 otherwise. j It is also defined as a decision variable. If a bus stop is added later, it is equal to 1, otherwise it is equal to 0.
[0007] Optionally, the constraint variables for establishing the multi-objective optimization model include: Limit the maximum number of bus routes sharing candidate stations, according to Formula 1: , where f i,j represents the frequency of bus arrivals of route i at station j, is a binary variable, j [1, 2, ..., m], the value is obtained by formula 3, where F j Provide a vehicle cap for site access; The candidate site is selected according to formula 2: [1, 2, …, m-1], where is a binary variable, and its value is calculated by Formula 4; Site selection binary variable, according to formula 3: ; The shared candidate site selection binary variable is based on Formula 4: ; Add a new binary variable for site selection, according to Formula 5: ; The binary variable of area overlap with the distance between two stations less than 600m, according to formula 6: ; The binary variable of area overlap with the distance between two stations less than 1000m, according to formula 7: ; The binary variable for judging the addition of stations with the distance between two stations greater than 1000m, according to formula 8: ; Objective function is the overall operation time of the bus, according to formula 9: , where, t j is the average time of the vehicle at station j, x j is the candidate station, L i is the length of the i-th line, V i is the average running speed of the i-th line; The available objective function represents the coverage area of the bus stop, according to formula 10: , where, R is the radius of the bus stop, which can be 300 or 500m, is a binary variable taking 0 or 1. When R takes 300m, c takes , when R takes 500m, c takes , is a binary variable taking 0 or 1, and d is the distance between two stations; If the binary variable takes 1, then the objective function should add the coverage area of the newly added station , according to formula 11: .
[0008] Optionally, the station coverage rate unit specifically refers to: the percentage of the service coverage area of the bus stop in the urban built-up area. Taking the bus stop as the center and the walking distance as the service radius, calculate the ratio of its coverage area to the urban built-up area, according to formula 12: , where, is the station coverage rate, is the coverage area of the i-th bus stop, n is the number of bus stops, and A is the urban built-up area.
[0009] Optionally, the bus operation time unit specifically refers to: the operation time is the overall operation time of the vehicles included in the selected urban bus lines, that is, the sum of the times of all vehicles from the starting station to the terminal station, according to formula 13: .
[0010] Optionally, the optimization result evaluation includes optimization solving, specifically: the particle swarm optimization algorithm and the genetic algorithm are selected to solve the established model, the solving software is Python. By comparing the solving results, it is found that the solving result of the particle swarm optimization algorithm does not match the actual situation. It obtains the optimal solution after the first iterative optimization and does not change during subsequent iterative solving. In this study, the genetic algorithm is used to solve the model to obtain the optimization scheme.
[0011] Optionally, the determination of the optimization scheme is based on experimental verification. Experimental derivation includes basic data and experimental analysis, specifically: through Python web crawler technology, the bus stops and bus line data of Pingdingshan City are crawled using the Gaode Map API. At the same time, road data is obtained from OpenStreetMap, which provides important information for subsequent analysis. Combining with bus GPS data, noise reduction processing is carried out on the outliers and irrelevant data in the experimental data, and the data labels are unified; Four bus lines are selected for site optimization analysis. The established multi-objective optimization model is solved using the genetic algorithm. According to the proposed model and solving method, the site layout is optimized and solved.
[0012] The present invention has the following beneficial effects: 1. The multi-objective optimization method for urban bus stop layout based on the genetic algorithm verifies the reliability of the model by establishing a random bus network from a theoretical perspective, and then takes an actual case as an example for application analysis. The results show that the proposed model has a good optimization effect, making the constructed site layout optimization model not only applicable to various bus networks, with strong practicability, but also able to assist bus enterprises in site adjustment and optimization.
[0013] 2. The multi-objective optimization method for urban bus stop layout based on the genetic algorithm converts the bus stop adjustment into a multi-objective optimization problem to ensure the overall optimality of the bus stop coverage rate and running time, that is, with the goal of maximizing the service quality and running efficiency of the bus network.
[0014] 3. The multi-objective optimization method for urban bus stop layout based on the genetic algorithm not only analyzes the theoretical solving results of the model, but also verifies it again with an actual case to ensure that the constructed model has good reliability and applicability. Description of the Drawings
[0015] Figure 1 It is the calculation flow chart of the multi-objective optimization method for bus stop layout in the present invention; Figure 2 It is the evaluation result within the research scope before and after optimization of the actual case in the present invention; Figure 3 It is the bus operation schedule within the research scope in the present invention. Detailed implementation manners
[0016] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0017] Please refer to Figure 1 and Figure 3 The present invention provides a technical solution: a multi-objective optimization method for urban bus stop layout based on genetic algorithm. The bus stop optimization method includes establishing a multi-objective optimization model, selecting bus evaluation indicators, solving using genetic algorithm, evaluating the optimization results, and determining the optimization plan. Establishing the multi-objective optimization model includes establishing the relationship between bus supply and user demand in different regions and establishing the relationship between the spacing between stations and the walking acceptance distance. The module for selecting bus evaluation indicators includes a station coverage rate unit and a bus operation time unit.
[0018] Specifically, establishing the multi-objective optimization model means: setting the starting station of each line as and the terminal station as Selecting the initial candidate stations except the starting and terminal stations for a line The candidate stations of the repeated lines are denoted as which are called shared candidate stations. Suppose there are k shared stations in the bus route, which are respectively recorded in the set That is, all stations except the starting and terminal stations are used as candidate stations and are optimized in subsequent station selection; Establishing a bus network, such as Figure 2 The bus network includes 4 bus lines and is set as Using to represent the starting station of the th line, and using to represent the terminal station of line For any route the order of the intermediate stops is always from to which are candidate stations. The relationship between the terminal station of route and the starting station of route +1 is , ; Since there are some stations shared by two or more routes, passengers can transfer between different bus lines. Suppose there are k shared candidate stations in the bus line. The k shared candidate stations form the set of all shared stations, denoted as G k, set G k There are 16 shared candidate stations in it. The first shared candidate station is k = 1, which can be used by both Route 1 and Route 4 at the same time. The set G1 = {S5, S 72} represents the stations shared by Bus Route 1 and Bus Route 4. Let be defined as a decision variable. If the candidate stop j is selected as a stop of Route i, it is equal to 1, otherwise it is equal to 0. Among them, is defined as a decision variable. If the shared stop k is selected as a station, it is equal to 1, otherwise it is equal to 0. A j is also defined as a decision variable. If a bus stop is added subsequently, it is equal to 1, otherwise it is equal to 0.
[0019] The constraint variables for establishing the multi-objective optimization model include: To ensure that there is no congestion and bunching at bus stops, therefore, an upper limit must be set for the distance between stops to ensure that the distance for passengers to walk to the bus stop is within an acceptable range, that is, it is necessary to limit the maximum number of bus routes at shared candidate stations to prevent congestion at bus stops. The maximum number of bus routes at shared candidate stations is limited according to Formula 1: , where f i,j represents the arrival frequency of the bus of Route i at stop j. is a binary variable, j ∈ [1, 2,..., m], and its value is obtained from Formula 3. F j is the upper limit of the number of passing vehicles at the station; To ensure that the shared candidate stop k is selected as a stop by at least one bus route, it is determined that the candidate stop is selected according to Formula 2: ∈ [1, 2,..., m - 1], where is a binary variable, and its value is calculated from Formula 4; The station selection binary variable is a judgment variable for determining whether a candidate station is selected. If there is only one route passing through the station, it should be selected according to Formula 3: ; The shared candidate station selection binary variable is a variable for determining whether a shared candidate station is selected by a route. If the number of passing vehicles at the station is greater than the carrying capacity of the station, some routes should be cancelled according to Formula 4: ; The new station selection binary variable is a variable for determining whether a new station is needed, according to Formula 5: ; The binary variable for overlapping area when the distance between two stations is less than 600m. This variable means that when the distance between two stations is less than 600m, there is an overlapping area that needs to be removed. It is necessary to judge the distance between the two stations. When it is greater than 600m, no removal is required. According to formula 6: ; The binary variable for overlapping area when the distance between two stations is less than 1000m. When the distance between two stations is less than 1000m, there is an overlapping area that needs to be removed. Therefore, it is necessary to judge the distance between the two stations. When it is greater than 1000m, no removal is required. According to formula 7: ; The binary variable for judging the addition of stations when the distance between two stations is greater than 1000m. When there is no station set between two stations with a distance greater than 1000m, corresponding stations should be added. According to formula 8: ; To ensure the operation of the bus system and improve the bus operation speed, the objective function is set is the overall operation time of the bus. According to formula 9: , where, t j is the average time of the vehicle at station j, x j is the candidate station, L i is the length of the i-th line, V i is the average operation speed of the i-th line; To ensure that the bus system can serve more citizens and facilitate citizens' travel, the station coverage rate is selected as the objective function of the optimization model, that is, to maximize the overall station coverage area as much as possible. The objective function represents the coverage area of the bus station. According to formula 10: , where, R is the radius of the bus station, which can be 300 or 500m, is a binary variable taking 0 or 1. When R takes 300m, c takes , when R takes 500m, c takes , is a binary variable taking 0 or 1, and d is the distance between two stations; If the binary variable takes 1, then the objective function should add the coverage area of the newly added station . According to formula 11: . Through the above formulas, the overall optimal results of the bus station coverage area and the overall operation time of the bus can be obtained.
[0020] The station coverage rate unit specifically refers to the percentage of the service coverage area of bus stops in the built-up area of the city. According to the "JT / T 1457-2023 Evaluation Index of Bus and Trolleybus Network", the calculation method is as follows: taking the bus stop as the center and the walking distance as the service radius, calculate the ratio of its coverage area to the built-up area of the city, according to formula 12: , where, is the station coverage rate, is the coverage area of the i-th bus stop, n is the number of bus stops, and A is the built-up area of the city.
[0021] The bus operation time unit specifically refers to: the operation time is the overall operation time of the vehicles included in the selected urban bus lines, that is, the sum of the times of all vehicles from the starting station to the terminal station, according to formula 13: .
[0022] The evaluation of the optimization result includes optimization solving, specifically referring to: selecting the particle swarm optimization algorithm and the genetic algorithm to solve the established model. The solving software is Python. By comparing the solving results, it is found that the solving result of the particle swarm optimization algorithm does not match the actual situation. It obtains the optimal solution after the first iterative optimization and does not change during subsequent iterative solving. In this study, the genetic algorithm is used to solve the model to obtain the optimization scheme.
[0023] Determining the optimization scheme is based on experimental verification. The experimental derivation includes basic data and experimental analysis, specifically referring to: through Python web crawler technology, using the Gaode Map API to crawl the bus stop and bus line data of Pingdingshan City. At the same time, road data is obtained from OpenStreetMap, which provides important information for subsequent analysis. Combining with bus GPS data, noise reduction processing is carried out on the outliers and irrelevant data in the experimental data, and the data labels are unified; Select 4 bus lines for station optimization analysis. According to Figure 3 as shown, select bus lines 4, 10, 16, and 21. Establish a multi-objective optimization model and use the genetic algorithm to solve it. According to the proposed model and solving method, optimize and solve the station layout. When the coverage radius is 300m, the overall station coverage area after optimization is the best; when the coverage radius is 500m, the station coverage area of a single line is the best. This result may be because when calculating the station coverage area according to a single line, there is no need to remove the overlapping area of the station coverage area, so that the coverage area is numerically larger. Whether it is the overall bus operation time or the overall station coverage area, both have been greatly improved. This result also reflects from the side that appropriately optimizing the station layout can effectively improve the operation efficiency and service quality of buses, and at the same time, it also proves again that the proposed station optimization model has good feasibility and practicality.
[0024] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention; the terms "first", "second", "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, unless otherwise clearly specified and defined, the terms "installed", "connected", "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.
[0025] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A multi-objective optimization method for urban bus stop layout based on genetic algorithm, characterized in that: The bus stop optimization method includes establishing a multi-objective optimization model, selecting bus evaluation indicators, solving using a genetic algorithm, evaluating the optimization results, and determining the optimization plan. Establishing the multi-objective optimization model includes establishing the relationship between bus supply and user demand in different regions and establishing the relationship between the distances between stations and the acceptable walking distance. The module for selecting bus evaluation indicators includes a station coverage rate unit and a bus operation time unit.
2. The multi-objective optimization method for urban bus stop layout based on genetic algorithm according to claim 1, characterized in that: The establishment of the multi-objective optimization model specifically means that the first station of each line is set as , and the last station is set as . One initial candidate station is selected from the stations other than the first and last stations of a line . The candidate stations of repeated lines are denoted as , which are called shared candidate stations. Suppose there are k shared stations on a bus route, which are respectively recorded in the set . That is, all stations except the first and last stations are used as candidate stations, and optimization is carried out in the subsequent selection of stations; Build a bus network, which includes 4 bus lines and is set as , using to represent the starting station of the th line, and using to represent the terminal station of line . For any route , the order of intermediate stops is always from to , which are the candidate stations. The relationship between the terminal station of route and the starting station of route + 1 is , ; Suppose there are k shared candidate stops on a bus route, and these k shared candidate stops form the set of all shared stops, denoted as G. k , the set G k contains 16 shared candidate stops. The first shared candidate stop is k = 1, which can be used by both Route 1 and Route 4. The set G1 = {S5, S72} represents the stops shared by Bus Route 1 and Bus Route 4. Define as a decision variable. If candidate stop j is selected as a stop on Route i, it is equal to 1; otherwise, it is equal to 0. Among them, is defined as a decision variable. If the shared stop k is selected as a stop, it is equal to 1; otherwise, it is equal to 0. A j is also defined as a decision variable. If a bus stop is added subsequently, it is equal to 1; otherwise, it is equal to 0.
3. The multi-objective optimization method for urban bus stop layout based on genetic algorithm according to claim 1, wherein: The constraint variables for establishing the multi-objective optimization model include: Limit the maximum number of bus lines for shared candidate stations according to Formula 1: , where f i,j represents the arrival frequency of the buses on Line i at Station j, is a binary variable, j ∈ [1, 2, …, m], and its value is obtained from Formula 3, F j is the upper limit of the passing vehicles at the station; The candidate site determination is selected according to Formula 2: [1, 2, …, m - 1], where is a binary variable, and its value is calculated by Formula 4; Site selection binary variable, according to Formula 3: ; Shared candidate site selection binary variable, according to Formula 4: ; Add a binary variable for site selection, according to Formula 5: ; The binary variable of area overlap with the distance between two stations less than 600m, according to formula 6: ; The binary variable of area overlap with the distance between two stations less than 1000m, according to Formula 7: ; Add a binary variable for site discrimination if the distance between two stations is greater than 1000m, according to Equation 8: ; Objective function is the overall bus running time, according to Formula 9: , where t j is the average time of the vehicle at stop j, x j is the candidate stop, L i is the length of the i-th line, V i is the average running speed of the i-th line; Available objective function Indicates the coverage area of a bus stop, according to Formula 10: , where R is the radius of the bus stop and can take 300 or 500 m, is a binary variable taking 0 or 1. When R takes 300 m, c takes , and when R takes 500 m, c takes , is a binary variable taking 0 or 1, and d is the distance between two stops; If the binary variable takes 1, then the objective function should add the coverage area of the newly added site , according to Formula 11: .
4. The multi-objective optimization method for urban bus stop layout based on genetic algorithm according to claim 1, characterized in that: The specific meaning of the site coverage rate unit is as follows: the percentage of the service coverage area of bus stops in the built-up area of the city. Taking the bus stop as the center and the walking distance as the service radius, calculate the ratio of its coverage area to the built-up area of the city according to formula 12: , where is the site coverage rate, is the coverage area of the i-th bus stop, n is the number of bus stops, and A is the built-up area of the city.
5. A multi-objective optimization method for urban bus stop layout based on genetic algorithm according to claim 1, characterized in that: The specific meaning of the bus operation time unit is as follows: the operation time is the overall operation time of the vehicles included in the bus line of the selected city, that is, the sum of the times of all vehicles from the starting station to the terminal station, according to Formula 13: .
6. The multi-objective optimization method for urban bus stop layout based on genetic algorithm according to claim 1, characterized in that: The evaluation of the optimization results includes optimization solving, specifically: the particle swarm optimization algorithm and the genetic algorithm are used to solve the established model, and the solving software is Python. By comparing the solving results, it is found that the solving result of the particle swarm optimization algorithm does not match the actual situation. It obtains the optimal solution after the first iterative optimization and does not change during subsequent iterative solving. In this study, the genetic algorithm is used to solve the model to obtain the optimization plan.
7. A multi-objective optimization method for urban bus stop layout based on genetic algorithm according to claim 1, characterized in that: Determining the optimization plan is based on experimental verification. The experimental derivation includes basic data and experimental analysis, specifically: through Python web crawler technology, the bus stop and bus line data of Pingdingshan City are crawled using the Gaode Map API. At the same time, road data is obtained from OpenStreetMap, which provides important information for subsequent analysis. Combining with bus GPS data, noise reduction processing is performed on the outliers and irrelevant data in the experimental data, and the data labels are unified; Select 4 bus lines for site optimization analysis, establish the multi-objective optimization model, and use the genetic algorithm to solve it. According to the proposed model and solving method, optimize and solve the site layout.