Method and system for optimizing urban area public transit network

Abstract

The invention discloses an optimization method and system of an urban area public transport network, and relates to the field of public transport planning, and the method comprises the steps: constructing a three-index system based on a road segment dynamic interaction index, a passenger flow space-time demand index and a subway collaborative transfer index; according to the public transit network data, using a three-index system to obtain a backbone line; constructing a target function by taking the bus route operation cost, the daily average load factor of the network, the transfer subway passenger flow and the energy efficiency index of the bus network as variables; and randomly selecting a preset number of bus lines and the backbone lines to form a bus network, taking a plurality of bus networks as a population, taking the target function value of each bus network as a fitness value, and obtaining an individual with the optimal fitness value in the population by using a genetic algorithm to serve as an optimized bus network. The circuit operation efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the field of public transportation planning, and in particular to an optimization method and system for an urban area public transportation network. Background Art

[0002] Within urban transportation systems, some areas exhibit complex traffic patterns, such as pocket-shaped road networks. For example, in the Haidian Mountain area, due to factors such as geography and urban planning, traffic is highly concentrated on the subway and some bus routes during peak hours, resulting in significant commuting pressure for residents. Traditional bus network planning methods are clearly inadequate for addressing this situation. They lack effective methods for accurately identifying backbone routes, making it difficult to identify routes that support the bus network, resulting in a lack of focus in route layout. Therefore, improving route efficiency is a critical issue that needs to be addressed. Summary of the Invention

[0003] The embodiments of the present invention provide a method and system for optimizing an urban area bus network, which can solve the problem of how to improve line operation efficiency in the prior art.

[0004] An embodiment of the present invention provides a method for optimizing an urban area bus network, comprising the following steps: Collect bus network data within the urban area; The importance score of bus network data is evaluated using a three-indicator system based on the dynamic interaction index of road sections, the spatiotemporal demand index of passenger flow, and the coordinated transfer index of subways. The weight of each indicator is set and the evaluation results are obtained. The backbone routes are selected based on the evaluation results. The operating cost of bus lines, the average daily load rate of the line network, the passenger flow of subway transfers, and the energy efficiency of the bus network are used as variables. The hierarchical analysis method and multi-objective optimization algorithm are used to determine the weight of each variable, and the objective function is obtained based on the variables and the weights of each variable. A preset number of bus lines and backbone lines are randomly selected to form a bus network. Multiple bus networks are used as a population, and the objective function value of each bus network is used as the fitness value. A genetic algorithm is used to obtain the individual with the best fitness value in the population as the optimized bus network.

[0005] Furthermore, the weight of each indicator is set and the evaluation results are obtained, and the backbone line is selected according to the evaluation results. The specific steps include: obtaining the dynamic interaction index based on the road section , the formula is: ; in, Bus routes for sections Upper section length; Passing section The total number of buses; For bus routes On the road Average length of stay on Obtaining spatiotemporal demand indicators based on passenger flow , the formula is: ; Among them, the time distribution imbalance coefficient of passenger flow is ; The spatial aggregation coefficient is ; For overlapping OD requirements; To meet all passenger flow needs; 、 Represents the site and sites ; Get subway-based collaborative transfer indicators , the formula is: ; Among them, the passenger flow saturation of subway transfer stations is ; The convenience coefficient of bus and subway transfer is ; For passengers transferring to the subway; Get the importance score of each bus line : ; in, is the weight; The bus routes whose importance scores exceed the set threshold are selected as backbone routes.

[0006] Furthermore, the step of obtaining the objective function specifically includes: the objective function is formulated as follows: ; in, To optimize the construction costs of public transportation operations; To optimize the construction cost of public transportation operations; is the average daily full load rate of the network; To optimize the passenger flow from bus transfer to subway; To optimize the passenger flow from bus transfer to subway; represents the energy efficiency index of the bus network; All are weight coefficients; Bus operation and construction costs before and after optimization and , the formula is: ; ; in, For bus routes collection, bus routes ; The number of buses required for bus route l; The purchase cost of each bus; fuel and maintenance costs for each bus; The manpower cost required for each bus; is the length of bus route l; is the required departure interval for bus route l; is the average operating speed of bus line l; Average daily load rate of the network , the formula is: ; in, For the Lines at the station and sites The actual passenger flow between For the first Line stations and sites Rated passenger capacity of public transport vehicles; For the Lines at the station and sites the distance between them; is the total number of stops on a bus route; is the total number of lines; Passenger flow from bus to subway and , the formula is: ; in, For the Standing at the site Passenger flow transferring to the subway; is the number of bus stops where you can transfer to the subway; The energy efficiency index of the bus network is as follows: ; in, For the bus network The length of the line; For the bus network On the line Traffic flow of different bus types; For the bus network On the line The speed of the bus types; For the bus network On the line kind Energy consumption factor when driving at speed; The number of bus models operating in the entire city.

[0007] Furthermore, after obtaining the objective function, the method further includes: constructing constraint conditions; The constraints include: line elastic length constraint, the formula is: ; ; in, are the upper and lower limits of the bus route length; is the average speed of buses running on the route; The maximum travel time for residents; is the time elasticity coefficient of line length; is the spatial elastic coefficient; is the line length; The efficiency constraint of non-straight line operation is as follows: ; in, is the nonlinear coefficient; For the The actual distance between the first and last stops of each bus route; For the The spatial distance between the first and last stops of a bus line; is the running speed weight; is the traffic reliability coefficient; Passenger flow equilibrium constraint, the formula is: ; in, is the maximum cross-sectional passenger flow in the line; is the average cross-sectional passenger flow of the entire line; The passenger capacity constraint of the route is: ; in, is the passenger capacity of line l; is the maximum cross-sectional passenger flow of line l; is the passenger capacity impact coefficient; The frequency of departure is adapted to the constraint, and the formula is: ; in, For the line Frequency of departure; For the line The minimum frequency among all departure frequencies; For the line The maximum frequency among all departure frequencies; Fleet size and management constraints, the formula is: ; in, The maximum fleet size that each bus route’s fleet size cannot exceed; For bus routes Length of line; is the average operating speed of vehicles on bus route l; Start taking values from 1 and take them in sequence , Represents the total number of bus routes; Line demand satisfaction and adjustment constraints, the formula is: ; in, For bus routes the maximum passenger capacity of the vehicle; for right No. Passenger flow demand of bus routes, The meaning is the starting point and end point of the trip; An index variable representing the requirement set Each element in the set Covers all OD pairs consisting of travel origins and destinations in the city; each Represents a set of travel origin and destination combinations, corresponding to the travel demand from one starting point to one destination; is the maximum number of bus routes; Line coverage guarantee constraints , the formula is: ; in, is the total passenger demand of the bus network; is the demand coverage coefficient; The grid density optimization constraint is: ; in, is the area of the wire mesh 𝑅; are the upper and lower limits of wire mesh density respectively; For the network The length of the bus route.

[0008] Furthermore, the bus network data specifically includes: Card data, road geographic information data, subway operation real-time data, traffic flow monitoring data and resident travel survey data.

[0009] An embodiment of the present invention provides an optimization system for an urban area bus network, comprising: The data collection module is used to collect bus network data in the urban area; the backbone line construction module is used to evaluate the importance score of bus network data using a three-indicator system based on the dynamic interaction index of the road section, the spatiotemporal demand index of the passenger flow, and the subway coordinated transfer index; the weight of each indicator is set and the evaluation results are obtained, and the backbone line is selected according to the evaluation results; the model construction module is used to use the bus line operating cost, the average daily full load rate of the line network, the subway transfer passenger flow and the energy efficiency of the bus line network as variables, and use the hierarchical analysis method and multi-objective optimization algorithm to determine the weight of each variable, and obtain the objective function based on the variables and the weights of each variable; the model solution module is used to randomly select a preset number of bus lines and backbone lines to form a bus network, and use multiple bus networks as a population, the objective function value of each bus network as the fitness value, and use the genetic algorithm to obtain the individual with the best fitness value in the population as the optimized bus network.

[0010] The embodiments of the present invention provide a method and system for optimizing an urban area bus network. Compared with the prior art, the methods and systems have the following beneficial effects: The three-indicator system, namely, the dynamic interaction index of road sections, the spatiotemporal demand index of passenger flow, and the coordinated transfer index of subways, takes into account the integration needs of buses and subways. The operating costs of bus lines, the average daily load rate of the line network, the passenger flow of subway transfers, and the energy efficiency of the bus network are used as variables to meet the actual traffic conditions. Ultimately, the bus network is optimized under actual traffic conditions, thereby improving the line operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A general flow chart of a method for optimizing an urban area bus network provided by an embodiment of the present invention; Figure 2 A technical roadmap for an optimization method for an urban area bus network provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0012] To make the above-mentioned objects, features, and advantages of the present invention more readily apparent, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings. The following description sets forth numerous specific details to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art may make similar modifications without departing from the scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0013] See also Figure 1 The embodiment of the present invention provides a method for optimizing a city area bus network, comprising the following steps: Step 1: Collect bus network data within the city, including: Card data, road geographic information data, subway operation real-time data, traffic flow monitoring data and resident travel survey data.

[0014] Data cleaning and fusion are required. This data is processed using tools like Python, traversing all subdirectories and files within the bus data root folder. During the data collection and preprocessing phase, data cleaning algorithms are used to remove obvious errors and outliers. Specific data fusion techniques are employed to integrate data from different sources according to unified standards, ensuring data accuracy and completeness and providing a reliable data foundation for subsequent analysis.

[0015] Step 2: Construct a three-index system based on the dynamic interaction index of road sections, the spatiotemporal demand index of passenger flow, and the coordinated transfer index of subways; use the three-index system to obtain the importance scores of bus routes based on bus network data and dynamically rank them; select the required routes as backbone routes based on the ranking results.

[0016] The indicator system for identifying backbone routes simultaneously assesses three indicators: the dynamic interaction index for road sections, the spatiotemporal demand index for passenger flow, and the coordinated transfer index for subways. The dynamic interaction index for road sections requires accurate data collection on bus speeds and number of stops along the route for proper calculation. The spatiotemporal demand index for passenger flow determines both temporal and spatial distribution. The coordinated transfer index for subways comprehensively considers factors such as transfer convenience at subway stations and the matching of bus and subway passenger flows. Finally, an intelligent algorithm accurately calculates the scores for each route and rationally selects backbone routes.

[0017] Step 3: Taking bus line operating costs, average daily full load rate of the line network, subway transfer passenger flow and energy efficiency of the bus line network as variables, the hierarchical analysis method and multi-objective optimization algorithm are used to determine the weights of each variable, and the objective function is obtained based on the variables and their weights; with line elastic length constraints, line non-linear operation efficiency constraints, passenger flow balance constraints, line passenger capacity constraints, departure frequency adaptation constraints, fleet size and management constraints, line demand satisfaction and adjustment constraints, line coverage guarantee constraints and line network density optimization constraints as constraints, a bus line network optimization model is constructed.

[0018] For the construction of bus network optimization model, the weight coefficients in the objective function change in real time based on multiple factors such as the bus company's operating strategy and the traffic planning policy of the city in the area. The hierarchical analysis method and multi-objective optimization algorithm are used to dynamically determine the objective function to ensure that the objective function can effectively guide the optimization process. The setting of constraint conditions should be combined with actual traffic conditions and operational requirements, and more refined constraint conditions should be added considering the actual operating conditions to achieve reasonable optimization of the bus network.

[0019] Step 4: Randomly select a preset number of bus routes and backbone routes to form a bus network, and use multiple bus networks as the initial population. The objective function value of each bus network is used as the fitness value. The bus network optimization model is solved using a genetic algorithm to obtain the individual with the best fitness value in the population as the optimized bus network.

[0020] 1. Data acquisition and preprocessing module: Multi-source data collection: Integrated bus card Data, road geographic information data, subway operation real-time data, traffic flow monitoring data and residents' travel survey data from multiple sources. Card data can obtain passenger travel information, road geographic information data can provide road network topology, subway operation data can analyze the relationship between bus and subway transfers, traffic flow monitoring data can reflect the real-time road congestion situation, and residents' travel survey data can understand travel demand preferences and determine the main region, providing a basis for subsequent analysis.

[0021] Data cleaning and fusion: Use data cleaning algorithms to remove outlier data, unify data formats and time-space benchmarks. Use data fusion technology to integrate data from different sources and integrate public transportation data. The card data is associated with the road geographic information data to determine the actual operation path and station location of the bus line, ensure the integrity and accuracy of the data, and provide a reliable data source basis for subsequent processing.

[0022] 2. Backbone line identification module: Construction of innovative three-indicator system: 1) Indicators based on dynamic interaction of road segments :Based on the traditional indicators based on road sections, the average stay time weight of buses on road sections is introduced The formula is as follows:

[0023] .

[0024] in Bus routes for sections Upper section length; Passing section The total number of buses; For bus routes On the road The average dwell time on.

[0025] This improvement can more accurately reflect the actual operational importance of the route on the road section, taking into account the operational differences of buses in different sections due to factors such as boarding and alighting at stations, making the identification of backbone routes more in line with actual operating conditions.

[0026] 2) Indicators based on passenger flow temporal and spatial demand :In addition to considering the traditional overlap Demand ratio, combined with actual demand to introduce the time distribution imbalance coefficient of passenger flow and spatial aggregation coefficient .

[0027] The formula is as follows: .

[0028] in For overlapping OD requirements; To meet all passenger flow needs.

[0029] Through this innovative indicator, the changing characteristics of passenger flow in different time periods and spatial areas can be captured, ensuring that backbone lines can effectively cover passenger flow in high-demand periods and high-demand areas, and improving the line's responsiveness to passenger flow demand.

[0030] 3) Indicators based on subway coordinated transfers :In addition to focusing on the proportion of subway transfer passengers, this indicator also introduces the passenger flow saturation of subway transfer stations The convenience coefficient of bus and subway transfer .

[0031] The formula is as follows: .

[0032] in For passengers transferring to the subway.

[0033] This indicator can more comprehensively evaluate the role of bus routes in promoting efficient and coordinated transfers between buses and subways, and guide the bus network and subway network to form a closer and more convenient connection.

[0034] Data identification and screening mechanism: Using big data analysis and intelligent algorithms to collect massive bus routes This system deeply mines and integrates bus card data, relevant road data, and real-time subway operation data. By building a decision-making model, it calculates the importance score of each bus route based on the three innovative indicators mentioned above and dynamically ranks them. Based on the ranking results, combined with actual experience and operational feedback, a certain proportion of routes with the best overall scores are selected as backbone routes, providing the infrastructure for subsequent optimization and model construction.

[0035] Comprehensive calculation of importance score: Comprehensively calculate the scores of the above three indicators and assign corresponding weights to each indicator according to actual conditions , through the formula Calculate the importance score of each bus line The weights can be determined by using methods such as the analytic hierarchy process, combining the bus company's operating strategy, urban traffic planning guidance, and actual operating data to scientifically assign values to ensure that the comprehensive score can accurately reflect the importance of the route.

[0036] Dynamic ranking method: (1) Data update driven ranking: Due to the dynamic characteristics of bus network data, passenger flow will fluctuate in different time periods and seasons, and the passenger flow saturation of subway transfer stations will also change in real time. Therefore, the latest bus network data is collected regularly, including bus IC card data, subway operation real-time data, and traffic flow monitoring data. After each data update, the three indicator scores and comprehensive importance scores of each bus line are recalculated, and then the bus lines are re-ranked based on the new scores. (2) Real-time abnormal data processing: During the data collection process, some abnormal data may appear, such as a sudden accident on a certain line causing a sudden change in passenger flow, or a failure of equipment at a subway transfer station affecting the convenience of transfer. For these abnormal data, real-time monitoring and identification algorithms are used to make judgments. Once abnormal data is found, it is corrected or eliminated in a timely manner, and the indicator scores and comprehensive rankings of the affected lines are recalculated. (3) Feedback adjustment ranking: Feedback information from bus operation departments is collected, such as passenger complaints, changes in operating costs, and evaluation of the effects of line adjustments. Based on this feedback information, the weights in the indicator system are dynamically adjusted.

[0037] 3.Bus network optimization model construction module: Comprehensive optimization objective function innovative design, construction of objective function: .

[0038] in To optimize the construction costs of public transportation operations; To optimize the construction cost of public transportation operations; is the average daily full load rate of the network; To optimize the passenger flow from bus transfer to subway; To optimize the passenger flow from bus transfer to subway; Represents the energy efficiency index of the bus network; weight coefficient ; It is calculated through comprehensive calculation of bus energy consumption data, mileage, passenger volume and other factors, aiming to promote the optimization of bus operations towards green and energy-saving direction; The size of is adjusted in real time based on multiple factors such as the bus company's operating strategy and the traffic planning policy of the city in the area. It is dynamically determined reasonably using the analytic hierarchy process (AHP) and multi-objective optimization algorithm to ensure that the objective function can comprehensively and balancedly guide the optimization of the bus network.

[0039] 1) Public transportation operation and construction costs before and after optimization.

[0040] The operating costs of public transportation mainly depend on the number of bus routes of various levels, route lengths, and departure frequencies, resulting in bus vehicle configuration costs, vehicle operation and maintenance costs, fuel costs, and employee salary costs.

[0041] The formula is as follows: .

[0042] .

[0043] in For bus routes collection, bus routes ; The number of buses required for bus route 1 (units); The purchase cost of each bus (yuan / vehicle / day); The fuel and maintenance cost of each bus (yuan / vehicle / day); The manpower cost required for each bus (yuan / bus / day); is the length of bus route l (meters); is the required departure interval (minutes) for bus route l; is the average operating speed of bus line l (m / min).

[0044] 2) Average daily load rate of the network.

[0045] The average daily load factor of a bus network depends on the distribution of passenger flow along the routes and the planning and design of bus routes. Fluctuations in passenger flow over time, differences in passenger flow in different areas due to function and activity intensity, and the route layout, coverage, and station configuration of bus routes all directly impact the average daily load factor.

[0046] The formula is as follows: .

[0047] in For the Lines at the station and sites The actual passenger flow between For the first Line stations and sites Rated passenger capacity of public transport vehicles; For the Lines at the station and sites the distance between them; is the total number of stops on a bus route; is the total number of lines.

[0048] 3) Passenger flow transferred from buses to subways.

[0049] Bus-to-subway transfer passenger volume refers to the number of passengers transferring from buses to the subway within the public transportation system. This metric is crucial for urban public transportation planning and operations management. It helps transportation planners and operators understand passenger travel patterns, optimize the integration of bus and subway networks, and improve overall transportation system efficiency and passenger satisfaction. The degree of integration between the bus and subway networks can be reflected in the number of passengers transferring from buses to the subway.

[0050] The formula is as follows: .

[0051] in For the Standing at the site Passenger flow transferring to the subway; is the number of bus stops where you can transfer to the subway.

[0052] 4) Energy efficiency of bus network.

[0053] The formula is as follows: .

[0054] in, For the bus network The length of the line; For the bus network On the line Traffic flow of different bus types; For the bus network On the line The speed of the bus types; For the bus network On the line kind Energy consumption factor when driving at speed; The number of bus models operating in the entire city.

[0055] Refined constraint setting: (1) Line elastic length constraint: The traditional line length constraints are as follows: .

[0056] .

[0057] in are the upper and lower limits of the bus route length; is the average speed of buses running on the route; The maximum travel time for residents.

[0058] Taking into account the significant differences in traffic flow in different periods of the regional road network, the time elasticity coefficient of the line length is introduced. and spatial elastic coefficient .

[0059] The constraints are as follows: .

[0060] This enables the line length to be adaptively adjusted according to tidal changes in traffic flow and regional spatial characteristics, avoiding severe congestion due to excessively long lines during peak hours, or affecting service coverage due to excessively short lines during off-peak hours.

[0061] (2) Efficiency constraints of non-linear operation of lines: The traditional non-linear coefficient constraints are as follows: .

[0062] in is the nonlinear coefficient; For the The actual distance between the first and last stops of the bus lines ( ); For the The spatial distance between the first and last stops of bus lines ( ).

[0063] In addition to the traditional non-linear coefficient constraints, the operating speed weight of the line under different traffic modes is increased. and traffic reliability coefficient , and obtain the comprehensive non-linear coefficient constraint.

[0064] The constraints are as follows: .

[0065] This constraint ensures that the line can reasonably select the operating path in a complex road network, improve overall operating efficiency and reliability, and reduce operation delays caused by complex road conditions.

[0066] (3) Passenger flow balance constraints: The passenger flow coefficient constraints are as follows: .

[0067] in is the maximum cross-sectional passenger flow in the line; is the average cross-sectional passenger flow of the entire line.

[0068] Based on the constraints of passenger flow equilibrium coefficient, real-time monitoring data and forecast information of passenger flow are introduced to construct a dynamic equilibrium model.

[0069] The construction of the dynamic equilibrium model includes: Data Collection and Integration: Leveraging public transportation IC card data and intelligent monitoring equipment, we collect real-time passenger flow data at different stops and time periods on each route. This data includes, for example, the number of passengers boarding and alighting at each stop during the morning rush hour. We also collect historical passenger flow data and special event schedules (such as holidays and sporting events). We use data cleaning algorithms to remove outliers and employ data fusion technology to integrate real-time monitoring data with historical data, unifying the data format and spatiotemporal benchmarks to provide accurate data support for subsequent analysis.

[0070] Indicator Calculation and Dynamic Analysis: Based on the integrated data, the passenger flow balance coefficient for each route is calculated in real time. In addition to the traditional ratio of maximum to average cross-sectional passenger flow, real-time data is combined to calculate indicators such as the rate of change of passenger flow over different time periods and sections to more comprehensively reflect the dynamic changes in passenger flow. Historical and real-time data are used to predict passenger flow over the next period of time, taking into account the impact of different weekdays, weekends, and special events on passenger flow. The system then predicts passenger flow for different routes and stations over the next few hours or days.

[0071] Model construction and strategy adjustment: A dynamic equilibrium model is established based on real-time monitoring and predicted passenger flow data. When the predicted passenger flow of a certain route during a certain period exceeds the set threshold, the route adjustment mechanism is automatically triggered. Differentiated adjustment strategies are formulated based on the functional positioning and passenger flow characteristics of different routes. For trunk lines, the main focus is on meeting the passenger flow needs of long-distance and large-volume transportation. When the predicted passenger flow is large, priority is given to increasing the number of vehicles or adjusting the departure interval. For branch lines and micro-circulation lines, emphasis is placed on the connection with the trunk line and the coverage of local areas. The operating range or station settings can be flexibly adjusted according to the real-time passenger flow situation.

[0072] Model Application and Feedback Optimization: Apply the dynamic equilibrium model to actual bus operations. Based on the adjustment strategies generated by the model, continuously collect actual passenger flow data and passenger feedback to evaluate the effectiveness of the dynamic equilibrium model. Analyze whether the model's route adjustments improve the passenger flow balance coefficient and whether indicators such as average waiting time and bus load factor are optimized. Based on the evaluation results, optimize and improve the dynamic equilibrium model's parameters, prediction algorithms, and adjustment strategies to achieve balanced passenger flow distribution and efficient utilization of bus resources.

[0073] (4) Line passenger capacity constraints: The constraints are as follows: .

[0074] in is the passenger capacity of line l; is the maximum cross-sectional passenger flow of line l; is the passenger capacity impact coefficient.

[0075] This constraint utilizes vehicle positioning and passenger flow monitoring technology to understand the route's passenger capacity requirements. Based on passenger flow density and vehicle load conditions on different sections of the route, vehicle operating intervals are dynamically adjusted to achieve passenger capacity allocation, ensuring that vehicle utilization and operational efficiency are maximized while meeting passenger flow demands.

[0076] (5) Frequency adaptation constraints: The constraints are as follows: .

[0077] in For the line Departure frequency.

[0078] This constraint is combined with the real-time location of buses and road congestion conditions to establish a real-time scheduling model and achieve adaptive adjustment of departure frequency.

[0079] (6) Fleet size and management constraints: The constraints are as follows: .

[0080] in The maximum fleet size that each bus route’s fleet size cannot exceed; For bus routes Length of line; is the average operating speed of buses on bus route l.

[0081] Determining the maximum fleet size requires comprehensive consideration of factors such as a city's bus passenger volume, total route length, average vehicle speed, and departure frequency. This can be determined by calculating the minimum number of vehicles required to meet peak-hour bus demand and incorporating a reserve factor of 10%-20%.

[0082] (7) Line demand satisfaction and adjustment constraints: The constraints are as follows: .

[0083] in For bus routes the maximum passenger capacity of the vehicle; for right No. Passenger flow demand of bus routes.

[0084] This constraint leverages big data analysis and intelligent prediction algorithms to dynamically adjust bus routes based on travel demand trends in different areas and time periods, ensuring that bus routes can meet residents' travel needs and improving bus services.

[0085] (8) Line coverage guarantee constraints: The constraints are as follows: .

[0086] in is the total passenger demand of the bus network; is the demand coverage coefficient.

[0087] Based on the actual needs of the existing bus network, a bus coverage optimization and guarantee mechanism will be established. While ensuring overall bus network coverage, a focus will be placed on areas with excessive demand and improving bus network accessibility. By optimizing route layout and station settings, the gap in bus service between different regions will be narrowed, achieving balanced bus service.

[0088] (9) Line network density optimization constraints: The constraints are as follows: .

[0089] in is the area of the wire mesh 𝑅; The upper and lower limits of the wire mesh density.

[0090] By combining traffic flow distribution, population density, and land use, cluster analysis and simulated annealing algorithms are used to dynamically optimize network density. In areas with high passenger flow, such as transportation hubs and commercial centers, network density is appropriately increased. In areas with sparse populations or low traffic flow, network density is reasonably reduced to improve the rationality of network layout.

[0091] The dynamic optimization of line network density includes: Data Collection and Preprocessing: Detailed road network traffic flow data for the study area will be collected, including information on traffic flow by road section and time of day. Accurate population density data will be obtained to clarify the distribution of population in each area. Land use data will be collected to understand the distribution of different functional areas, such as commercial, residential, and industrial areas. The collected data will be cleaned and preprocessed to remove outliers, fill in missing values, and integrate and standardize data from different sources to ensure consistency and comparability.

[0092] Cluster analysis: Traffic flow, population density, and land use data are integrated, using traffic flow, population density, and land use type as characteristic variables. A cluster analysis algorithm is used to cluster the study area. Through multiple experiments and analyses, the appropriate number of clusters is determined, ensuring that areas within the same cluster have similarities in traffic flow, population density, and land use. Based on the clustering results, the study area is divided into different categories, each representing areas with similar characteristics, providing a basis for subsequent optimization of the network density.

[0093] Initial bus network density setting: Based on historical data and existing plans, an initial bus network density value is set for each cluster area. This initial value serves as the starting point for the simulated annealing algorithm.

[0094] Simulated Annealing Algorithm Optimization: Define the objective function. This objective function is constructed based on bus service coverage, average passenger travel distance, and bus operating costs. The objective function is set to minimize average passenger travel distance or minimize bus operating costs while meeting certain service coverage requirements. Determine the initial temperature, cooling rate, and termination criteria for the simulated annealing algorithm. The initial temperature must be high enough to ensure the algorithm can search within a large solution space. The cooling rate determines the speed of temperature drop and influences the algorithm's convergence rate. The termination criteria determine whether the algorithm should terminate. Within each cluster, the initial network density is randomly perturbed to generate a new network density solution. The objective function value of the new solution is calculated and compared with the objective function value of the current solution. If the new solution has a better objective function value, the new solution is accepted; otherwise, the new solution is accepted with a certain probability, which is related to the difference between the current temperature and the objective function value. As the algorithm progresses, the temperature gradually decreases, the probability of accepting a less favorable solution decreases, and the algorithm gradually converges to a more favorable solution. Repeat these steps until the termination criteria are met. The resulting network density solution is considered the relatively better solution within that cluster.

[0095] Scheme Evaluation and Adjustment: Evaluate the optimized network density scheme to check whether it meets actual traffic needs, operational requirements, and planning objectives. Based on the evaluation results, make appropriate adjustments and improvements to the optimized network density scheme to ensure its feasibility and effectiveness.

[0096] Implementation and Monitoring: Apply the optimized network density plan to actual bus network planning and conduct relevant monitoring, continuously collecting data on traffic flow, population distribution, and land use changes. Regularly re-perform cluster analysis and simulated annealing algorithm optimization based on new data to adapt to changes in urban development and transportation demand, and achieve dynamic optimization of network density.

[0097] Finally, based on the backbone routes, the bus network optimization model is used to obtain the optimized bus network, including: Initializing parameters and setting the initial solution: Determine the various parameters required to solve the bus network optimization model, including but not limited to the genetic algorithm's population size, number of iterations, crossover probability, and mutation probability. Construct an initial solution based on backbone routes, which serve as the core component of the initial solution. A certain number of routes are randomly selected from the candidate set to form a complete initial bus network solution. These initial solutions constitute the genetic algorithm's initial population. Each solution can be considered a chromosome, with the routes acting as genes on the chromosome.

[0098] Calculate the objective function value and fitness evaluation: For each bus network plan (chromosome) in the initial population, according to the objective function formula Calculate the objective function value. They represent the bus operation and construction cost before optimization, the bus operation and construction cost after optimization, the average daily load rate of the network, the passenger flow of bus transfer to subway before optimization, the passenger flow of bus transfer to subway after optimization, and the energy efficiency index of the bus network. is the weight coefficient. The objective function value is converted into a fitness value, which is used to measure the pros and cons of each plan in the current population. The larger the objective function value, the higher the fitness, indicating that the bus network plan is better.

[0099] Genetic Operation - Selection: This selection strategy selects a certain number of excellent solutions from the current population based on their fitness values and enters the next generation population. Solutions with high fitness are more likely to be selected, thus preserving and transmitting excellent bus network characteristics within the population.

[0100] Genetic Operation - Crossover: Two bus network solutions (chromosomes) are randomly selected from the selected population and crossover is performed based on a set crossover probability. Using methods such as position-based crossover and sequential crossover, some genes (trails) of the two chromosomes are swapped to generate a new offspring chromosome (a new bus network solution). During the crossover process, the newly generated solution must ensure that it meets all constraints, including route elastic length constraints and non-linear route operation efficiency constraints. If not, adjustments or recrossover are performed.

[0101] Genetic Operation - Mutation: Mutate the offspring chromosomes obtained after crossover according to the set mutation probability. Mutation methods can include randomly changing the direction of a particular route, adding or removing route stations, etc. Similarly, the mutated solution must meet all model constraints; otherwise, the mutation is repeated. Mutation aims to introduce new genes (route combinations) into the population, preventing the algorithm from falling into a local optimum.

[0102] Check termination conditions: Check whether the algorithm's termination conditions are met. Common termination conditions include reaching the maximum number of iterations and the objective function value changing minimally over several consecutive iterations. If the termination conditions are not met, return to step 2 and continue with the objective function value calculation, genetic operations, and other steps. If the termination conditions are met, proceed to the next step.

[0103] Determine the optimized bus network: Select the bus network solution with the highest fitness from the final population. The main routes in this solution are the optimized bus network obtained by the bus network optimization model based on the backbone routes. Evaluate the optimized bus network by analyzing various factors such as route operating costs, network coverage, average passenger travel time, and number of transfers to verify whether the optimization results meet the expected goals.

[0104] The present invention is based on the fusion of multi-source data under the regional road network to construct a bus network optimization model. By collecting data such as bus IC cards, real-time subway operation, traffic flow monitoring and resident travel surveys, and cleaning and integrating them, an innovative three-index system including road section operation efficiency, passenger flow temporal and spatial distribution, and bus-subway coordination is used to accurately identify backbone lines. The model is built by comprehensively considering factors such as bus operating costs, line network full load rate, bus and subway transfer passenger flow, and a series of refined constraints. This model can optimize the layout of the bus network based on the traffic conditions and passenger flow information of the complex road network, improve the bus operation efficiency and the subway coordination effect, effectively alleviate urban traffic congestion, and provide residents with high-quality travel services. It is innovative and practical and has promotion value. The technical roadmap is as follows: Figure 2 shown.

[0105] An embodiment of the present invention provides an optimization system for an urban area bus network, comprising: The data collection module collects bus network data within the urban area. The backbone route construction module evaluates the importance of bus network data using a three-indicator system: a dynamic interaction indicator based on road segments, a spatiotemporal passenger flow demand indicator, and a subway-based coordinated transfer indicator. Weights are assigned to each indicator, evaluation results are obtained, and backbone routes are selected based on the evaluation results. The model construction module uses bus route operating costs, the average daily load factor of the network, subway transfer passenger flow, and the energy efficiency of the bus network as variables, and uses the analytic hierarchy process and a multi-objective optimization algorithm to determine the weights of each variable. The objective function is then derived based on the variables and their weights. The model solution module randomly selects a preset number of bus routes and backbone routes to form a bus network. The multiple bus networks are then grouped as a population, and the objective function value of each bus network is used as the fitness value. A genetic algorithm is then used to identify the individual with the best fitness value within the population as the optimized bus network.

[0106] A specific embodiment is as follows: This embodiment discloses a method for optimizing a bus network in an urban area, and the specific steps are as follows: S1. Data preparation step: Collecting public transportation Card data, road geographic information data, subway operation real-time data, traffic flow monitoring data and resident travel survey data.

[0107] S2. Data cleaning and fusion steps: relying on rule-based screening and data statistical analysis methods, by setting thresholds, Principle, remove illogical data such as negative passenger flow, and complete data cleaning and integration. In addition, the data collection and processing process is to cooperate with multiple departments and platforms to collect public transportation data. Cards, road geography, subway operation, traffic flow, residents' travel and other multi-source heterogeneous data, The principle algorithm cleans data and the spatial interpolation and time series matching technology fuse the data to ensure its reliability and completeness, providing a basis for the subsequent analysis of specific routes.

[0108] S3. Key Route Identification Step: Based on an innovative three-index system consisting of route segment operating efficiency, spatiotemporal passenger flow distribution, and bus-subway collaboration, key routes are selected using big data analysis and intelligent algorithms to mine and process data. Furthermore, this key route identification phase establishes a three-index system: one based on route segment dynamic interaction, one based on spatiotemporal passenger flow demand, and one based on subway coordinated transfers. Innovative calculations are performed based on existing formulas and real-world needs. Data is mined and processed using big data and intelligent algorithms. After comprehensive evaluation, combined with experience and feedback, selected routes are selected as key routes, paving the way for subsequent route optimization.

[0109] S4. Comprehensive model building steps: Comprehensively consider the factors and constraints that affect the objective function, and rationally design multiple factors and refined constraints to build the bus network model. In addition, when building the bus network optimization model, determine the objective function containing multiple factors. 、 、 Relevant parameters such as the line length and the line elasticity are determined according to the actual situation to ensure that the constraints are practical and effective.

[0110] The core purpose of the present invention is to provide a method for optimizing urban area road network bus lines based on multi-source data fusion, to achieve the optimization of complex bus lines, to effectively resolve the problem of reasonable operation of relevant lines in the existing bus network in such areas, to improve line operation efficiency, and to strengthen the organic integration of buses and subways, thereby optimizing residents' travel experience.

[0111] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be determined by the appended claims.

Claims

1. A method for optimizing a bus network in an urban area, characterized in that: The following steps are involved: Collect bus network data within the urban area; The importance score of bus network data is evaluated using a three-index system based on the dynamic interaction index of road sections, the spatiotemporal demand index of passenger flow, and the coordinated transfer index of subways. Set the weight of each indicator and obtain the evaluation results, and select the backbone line based on the evaluation results; Taking bus line operating costs, average daily load factor of the bus network, passenger flow transferring to subways, and energy efficiency of the bus network as variables, the analytic hierarchy process and multi-objective optimization algorithm were used to determine the weights of each variable, and the objective function was obtained based on the variables and their weights. A preset number of bus routes and backbone routes are randomly selected to form a bus network. Multiple bus networks are used as a population, and the objective function value of each bus network is used as the fitness value. A genetic algorithm is used to obtain the individual with the best fitness value in the population as the optimized bus network.

2. The method for optimizing an urban area bus network according to claim 1, wherein: The steps of setting the weights of the indicators and obtaining the evaluation results, and selecting the backbone lines according to the evaluation results, include: Get dynamic interaction indicators based on road segments , the formula is: ； in, Bus routes for sections Upper section length; Passing road section The total number of buses; For bus routes On the road Average length of stay on Obtaining spatiotemporal demand indicators based on passenger flow , the formula is: ； Among them, the time distribution imbalance coefficient of passenger flow is ; The spatial aggregation coefficient is ; For overlapping OD requirements; To meet all passenger flow needs; 、 Represents the site and sites ; Get subway-based collaborative transfer indicators , the formula is: ； Among them, the passenger flow saturation of subway transfer stations is ; The convenience coefficient of bus and subway transfer is ; For passengers transferring to the subway; Get the importance score of each bus line : ； in, is the weight; Bus routes whose importance scores exceed the set threshold are selected as backbone routes.

3. The method for optimizing an urban area bus network according to claim 1, wherein: The specific steps of obtaining the objective function include: The objective function is formulated as follows: ； in, To optimize the construction costs of public transportation operations; To optimize the construction cost of public transportation operations; is the average daily full load rate of the network; To optimize the passenger flow from bus transfer to subway; To optimize the passenger flow from bus transfer to subway; represents the energy efficiency index of the bus network; All are weight coefficients; Bus operation and construction costs before and after optimization and , the formula is: ； ； in, For bus routes collection, bus routes ; The number of buses required for bus route l; The purchase cost of each bus; fuel and maintenance costs for each bus; The manpower cost required for each bus; is the length of bus route l; is the required departure interval for bus route l; is the average operating speed of bus line l; Average daily load rate of the network , the formula is: ； in, For the Lines at the station and sites The actual passenger flow between For the first Line stations and sites Rated passenger capacity of public transport vehicles; For the Lines at the station and sites the distance between them; is the total number of stops on a bus route; is the total number of lines; Passenger flow from bus to subway and , the formula is: ； in, For the Standing at the site Passenger flow transferring to the subway; is the number of bus stops where you can transfer to the subway; The energy efficiency index of the bus network is as follows: ； in, For the bus network The length of the line; For the bus network On the line Traffic flow of different bus types; For the bus network On the line The speed of the bus types; For the bus network On the line kind Energy consumption factor when driving at speed; The number of bus models operating in the entire city.

4. The method for optimizing an urban area bus network according to claim 1, wherein: After obtaining the objective function, the method further includes: constructing constraint conditions; The constraints include: Line elastic length constraint, the formula is: ； ； in, are the upper and lower limits of the bus route length; is the average speed of buses running on the route; The maximum travel time for residents; is the time elasticity coefficient of line length; is the spatial elastic coefficient; is the line length; The efficiency constraint of non-straight line operation is as follows: ； in, is the nonlinear coefficient; For the The actual distance between the first and last stops of each bus route; For the The spatial distance between the first and last stops of a bus line; is the running speed weight; is the traffic reliability coefficient; Passenger flow equilibrium constraint, the formula is: ； in, is the maximum cross-sectional passenger flow in the line; is the average cross-sectional passenger flow of the entire line; The passenger capacity constraint of the route is: ； in, is the passenger capacity of line l; is the maximum cross-sectional passenger flow of line l; is the passenger capacity impact coefficient; The frequency of departure is adapted to the constraint, and the formula is: ； in, For the line Frequency of departure; For the line The minimum frequency among all departure frequencies; For the line The maximum frequency among all departure frequencies; Fleet size and management constraints, the formula is: ； in, The maximum fleet size that each bus route’s fleet size cannot exceed; For bus routes Length of line; is the average operating speed of vehicles on bus route l; Start taking values from 1 and take them in sequence , Represents the total number of bus routes; Line demand satisfaction and adjustment constraints, the formula is: ； in, For bus routes the maximum passenger capacity of the vehicle; for right No. Passenger flow demand of bus routes, The meaning is the starting point and end point of the trip; An index variable representing the requirement set Each element in the set Covers all OD pairs consisting of travel origins and destinations in the city; each Represents a set of travel origin and destination combinations, corresponding to the travel demand from one starting point to one destination; is the maximum number of bus routes; Line coverage guarantee constraints , the formula is: ； in, is the total passenger demand of the bus network; is the demand coverage coefficient; The grid density optimization constraint is: ； in, is the area of the wire mesh 𝑅; are the upper and lower limits of wire mesh density respectively; For the network The length of the bus route.

5. The method for optimizing an urban area bus network according to claim 1, wherein: The bus network data specifically includes: Card data, road geographic information data, subway operation real-time data, traffic flow monitoring data and resident travel survey data.

6. An optimization system for urban area bus network, characterized in that: include: Data collection module, used to collect bus network data within the urban area; The backbone route construction module is used to evaluate the importance score of bus network data using a three-index system based on the dynamic interaction index of the road section, the spatiotemporal demand index of passenger flow, and the coordinated transfer index of the subway; Set the weight of each indicator and obtain the evaluation results, and select the backbone line based on the evaluation results; The model building module is used to use bus line operating costs, the average daily full load factor of the line network, the number of passengers transferring to the subway, and the energy efficiency of the bus line network as variables, and uses the hierarchical analysis method and multi-objective optimization algorithm to determine the weights of each variable, and obtain the objective function based on the variables and their weights; The model solving module is used to randomly select a preset number of bus routes and backbone routes to form a bus network, and use multiple bus networks as a population, the objective function value of each bus network as the fitness value, and use a genetic algorithm to obtain the individual with the best fitness value in the population as the optimized bus network.

Images