Bus network optimization method and system

Abstract

The invention discloses a public transit network optimization method and system. The method comprises the steps of 1, acquiring urban traffic network data, bus station data, station data, passenger flow OD and lines needing to be adjusted; step 2, mapping bus stops to an urban traffic road network; 3, updating the urban traffic network data; 4, adjacent bus stations are connected to form a bus traffic network; step 5, finding out a target OD which needs to be covered by the planned route; step 6, generating a candidate line pool; and 7, calculating a coverage relationship between the lines in the candidate line pool and the target OD, and then finding out a line set which can cover all the target OD and has the shortest mileage to realize public transit network optimization. According to the method, the problem of network optimization adjustment of a common asymmetric traffic network can be solved, and the applicability of the network optimization method is greatly improved. Particularly, under the condition that the current bus operation cost is high, the service level is not reduced while the operation cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a bus network optimization method and system. Background Art

[0002] Public transportation planning includes five stages: network design, route frequency setting, timetable development, vehicle dispatching, and driver scheduling. This patent focuses on the first stage. With the continuous development of cities and the integration of various modes of transportation, it is necessary to regularly optimize and adjust the public transportation network. Especially now that public transportation operations are under great pressure, there is an urgent need to reduce operating costs while ensuring service levels. For example, a city has nearly a thousand regular bus routes and conducts one or two network optimization adjustments every year, with the number of routes adjusted each time generally less than 10%.

[0003] There is a lot of research on bus network optimization. Currently, mathematical programming is the main method used for network optimization, and recently reinforcement learning has also emerged as a method for route design.

[0004] There are two types of route optimization methods in mathematical programming: 1) Directly treating edges in the transportation network as decision variables, using flow balance, route length, station spacing, and nonlinearity coefficients as constraints. The optimization objective is to minimize cost or maximize passenger flow. This approach can theoretically yield an optimal solution. However, the model created in this way is large, with a large number of decision variables and constraints, making it applicable only to very small networks.

[0005] 2) Use heuristic rules to generate a route pool, then select routes from this pool through integer programming modeling. The optimization objective is also to minimize cost, with passenger flow coverage as the constraint. A key aspect of this approach is how to generate the route pool. Enumerating all routes would result in a combinatorial explosion. Generally, industry knowledge and experience are used to determine an appropriately sized route pool, and then the route with the lowest cost is selected as the optimization result.

[0006] Reinforcement learning, a method that uses trial and error combined with rewards, allows a deep learning network to gradually learn an optimized network design. This method requires significant computing power and is currently primarily an academic study, limited to small-scale network experiments.

[0007] Existing network planning methods all have the following problems: 1. Only applicable to symmetrical traffic networks.

[0008] Symmetry means that the topology between two network nodes is symmetrical. For example, if edge A→B is in the transportation network, then edge B→A is also in the transportation network, where A and B are nodes (bus stops) in the transportation network. Accordingly, current research also assumes that the uplink and downlink of a line are also symmetrical, that is, the uplink and downlink contain the same nodes but in opposite order. Figure 1 The traffic network shown is symmetrical. The network nodes are bus stops. The paths between two adjacent stops have the same length but are in opposite directions.

[0009] But in real traffic networks, many are asymmetric: a. There are asymmetric sites, that is, there is no site with the same name on the opposite side of the road. Figure 2 In (a), a certain city has an "XFHA" bus stop on one side of the road but not on the other. In a certain first-tier city, 10% of bus stops are asymmetric.

[0010] b. There is a topological asymmetry in the paths between sites. Figure 2 In (b), there's a straight-ahead and right-turn route from "HFJRG (South Road)" to "HD (West Road)," while going from "HD (East Road)" to "HFJRG (North Road)" requires going straight, making a U-turn, and then turning left. This indicates an asymmetric traffic network. This asymmetry can occur whenever two stations with the same name are on opposite sides of an intersection, and there are many such stations.

[0011] 2. The impact of stations on line operating costs is not considered.

[0012] One of the stations at both ends of the line must usually be a depot or near a depot, otherwise empty travel will inevitably result. Summary of the Invention

[0013] The technical problem to be solved by the embodiments of the present invention is to provide a bus network optimization method and system to provide a bus network optimization adjustment that is more in line with reality.

[0014] In order to solve the above technical problems, an embodiment of the present invention proposes a bus network optimization method, comprising: Step 1: Obtain urban traffic network data, bus stop data of existing bus routes, bus station data, passenger flow OD, and routes that need to be adjusted; Step 2: Map the bus stops of existing bus routes to the urban transportation network; Step 3: Update urban traffic network data based on the mapping results; Step 4: Connect adjacent bus stops to form a bus transportation network; Step 5: Find the target OD that the planned route needs to cover based on the route that needs to be adjusted; Step 6: Generate a candidate line pool based on the target OD that needs to be covered; Step 7: Calculate the coverage relationship between the routes in the candidate route pool and the target OD, and then optimize the bus network by finding a set of routes that can cover all target ODs and have the shortest mileage.

[0015] Accordingly, an embodiment of the present invention further provides a bus network optimization system, comprising: Data preparation module: obtain urban traffic network data, bus stop data of existing bus routes, station data, passenger flow OD, and routes that need to be adjusted; Station mapping module: maps the bus stops of existing bus routes to the urban transportation network; Road network update module: updates urban traffic road network data according to mapping results; Public transportation network generation module: connects adjacent bus stops to form a public transportation network; OD determination module: finds the target OD that the planned route needs to cover based on the route that needs to be adjusted; Line pool generation module: generates a candidate line pool based on the target OD that needs to be covered; Line network optimization module: Calculates the coverage relationship between the routes in the candidate line pool and the target OD, and then optimizes the bus line network by finding a set of routes that can cover all target ODs and has the shortest mileage.

[0016] The beneficial effects of the present invention are: the present invention can solve the common network optimization and adjustment problems of asymmetric traffic networks, and greatly improve the applicability of network optimization methods; especially under the current situation where public transportation operating costs are under pressure, it can achieve the goal of reducing operating costs without reducing service levels. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a symmetrical transportation network diagram.

[0018] Figure 2 (a) is a schematic diagram of an asymmetric site; (b) is a schematic diagram of an asymmetric transportation network.

[0019] Figure 3 It is a flowchart of the bus network optimization method according to an embodiment of the present invention.

[0020] Figure 4 This is a schematic diagram of mapping bus stops onto an urban traffic network according to an embodiment of the present invention.

[0021] Figure 5 3 is a schematic diagram of updated urban traffic network data according to an embodiment of the present invention.

[0022] Figure 6 Schematic diagram of a public transportation network according to an embodiment of the present invention.

[0023] Figure 7 Schematic diagram of the expanded OD path according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] It should be noted that, unless there is a conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The present invention is further described in detail below with reference to the drawings and specific embodiments.

[0025] 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 and movement status of the various components under a specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0026] In addition, the terms "first," "second," and so on, used in this disclosure are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Therefore, features specified as "first" or "second" may explicitly or implicitly include at least one of these features.

[0027] This invention addresses the current transportation industry challenges of optimizing bus network operations to reduce operating costs while maintaining service levels. Specifically, it addresses the following practical issues, which serve as the foundation for subsequent bus planning phases: 1) The transportation network is asymmetric; 2) One end of the line needs to be a station or near a station.

[0028] The present invention also adopts a mathematical programming method, assuming that existing bus stops do not change, and then tries to build a candidate route pool and select a suitable route from the pool as the optimization result.

[0029] The goal of mathematical programming is to minimize operating costs. Since network optimization can reduce operating costs primarily through dynamic operating costs, i.e., total operating mileage, which is positively correlated with total line length, the optimization goal here is to minimize line mileage.

[0030] The level of service is primarily reflected in travel time. Assuming that bus speeds remain largely unchanged before and after network adjustments, the only factors that network optimization can impact travel time are trip distance and the number of transfers. This requires no more than one transfer. Since the frequency setting in the second phase of bus planning determines waiting time, a consistent level of service during the network design phase of bus planning means no increase in travel distance.

[0031] The main idea of ​​the bus network optimization of the present invention is: 1) forming a bus transportation network (the transportation network between bus stops) based on the input data. To represent the vertices of the graph is a bus stop, the edge of the graph 1) The target OD (origin and destination) is determined, and a candidate route pool is constructed based on the target OD. 2) Routes are selected from the route pool to optimize the network.

[0032] Please refer to Figure 3 The bus network optimization method of the embodiment of the present invention includes steps 1 to 7.

[0033] Step 1: Data preparation: obtain urban traffic network data, bus stop data of existing bus routes, bus station data, passenger flow OD, and routes that need to be adjusted.

[0034] Urban traffic network data: This includes node and segment data. A segment connects two adjacent nodes, and a node is typically the intersection of two roads. Map vendors or land planning departments own this data.

[0035] Existing bus route station data: each route includes the station number, ID, name, longitude and latitude, station direction angle, and the distance between adjacent stations. Bus companies or transportation bureaus have this data.

[0036] Station data: station ID, name, latitude and longitude of entrances and exits.

[0037] Passenger OD: The origin and destination points, travel distance, and number of transfers of existing bus ridership. Bus companies or transportation bureaus do not directly possess this data, but it can be inferred from electronic payment data for buses.

[0038] Routes that need adjustment: Planners usually specify routes that need adjustment. These routes are usually considered to need adjustment due to high duplication, poor efficiency or too winding line shape.

[0039] Step 2: Map the bus stops of existing bus routes to the urban traffic road network (the spatial network infrastructure that organizes and guides bus flows, represented by road sections and the nodes at both ends. Each road section has a start node snode and an end node enode. If the road's direction of travel is snode->enode, then the road section is a forward road; if the road's direction of travel is enode->snode, then the road section is a forward road; if the road can travel in both directions, then the road section is a two-way road). This step is to establish the relationship between the station and the road network. The final information to be given is: 1) which road section of the road network each station is projected onto; 2) the specific coordinates of the projection point; 3) whether the direction of the station is in the same direction or in the opposite direction to the direction of the road section (the station and the road section are in the same direction: if the vehicle passing through the station is snode->enode on the road section, then the station and the road section are in the same direction, otherwise the station and the road section are in the opposite direction). For example Figure 4 As shown, the larger dots are stations, the smaller dots are network nodes, and the network links are network segments.

[0040] Bus stop data and urban road network data often come from different sources. Simple distance matching is not suitable for some stops, requiring a more complex approach. Specifically, this approach utilizes not only the location of the stop but also its orientation and the distance between adjacent stops. The relevant algorithms are common knowledge in the field and are not the focus of this invention, so they will not be described in detail here.

[0041] Step 3: Update the urban traffic network data based on the mapping results. This step is to update the road network data and add the station to the node set of the road network. The road section where the station is located should also be split accordingly. This is to prepare for the subsequent path search using Dijkstra or A* algorithm. Figure 5 As shown, the larger dots are the projection points of the stations on the road segments, that is, they have become nodes of the road segments.

[0042] Step 4: Connect adjacent bus stops to form a bus network. This step searches for pathways between adjacent bus stops and ultimately forms a bus network. This bus network serves as the foundation for the subsequent construction of a bus route pool.

[0043] The specific approach is to traverse any two non-identical bus stops and calculate the distance between them on the urban transportation network. If the distance is less than a certain threshold and there are no other bus stops on the path, the path is considered part of the bus transportation network. The path between two adjacent stops is the section of the subsequent bus route, and a distance threshold of 2 km can be set. Because there are no other bus stops on the path, the actual distance between two adjacent stops is generally no more than 600 meters.

[0044] like Figure 6 As shown, the dots are bus stops, and there may be edges connecting two bus stops. Figure 6 The highlighted road section is an edge of the bus network.

[0045] Step 5: Identify the target ODs that the planned route needs to cover based on the routes that need to be adjusted. First, identify the relevant ODs for the routes that need to be adjusted. Then, check whether these ODs are covered by the remaining bus network. The remaining bus network is the existing bus network minus the routes that need to be adjusted. ODs that are not covered by the remaining bus network are the ODs that the planned route needs to cover. Coverage here means that passengers can travel from O to D directly via the bus network or with a single transfer.

[0046] In the actual calculation process, small ODs (e.g., ODs with an average of less than one passenger per day) can be removed to form the final target OD. This is because public transportation primarily serves those with regular travel ODs. Small ODs are usually random travel demands and do not need to be specifically considered.

[0047] Step 6: Generate a candidate line pool based on the target OD that needs to be covered.

[0048] First, the target bus routes are all in the route pool, which will also serve as the initial values ​​for the subsequent optimization solution. Then, a set of candidate routes will be generated for each OD. The specific steps are described below: Step 6.1: Use the Yen algorithm to calculate the k shortest paths from O to D in the bus network. The value of k here can be selected based on the available computing resources. If k can be selected larger, the corresponding optimization effect will be better.

[0049] Step 6.2: Extend the path to the feasible route endpoints and the nearest station. If a station name has two corresponding stations, that station can be used as a route endpoint. This step extends the path result from the previous step in both directions to ensure that both ends can be route endpoints, and further extends it to the nearest station entrance or exit.

[0050] like Figure 7 As shown, The site is The site extends forward if The station itself can be used as the line endpoint, so there is no need to extend it. The site is site. There may be multiple options for the site. To select the site direction angle and The one with the closest station is extended as straight as possible. Site backward Site extension. The station continues forward to the nearest station entrance and exit , The station extends back to the nearest station entrance and exit .

[0051] Step 6.3: Generate an alternative complete route based on the expanded path. It's an uplink line. is another uplink line. Search for the corresponding downlink lines for these two lines respectively.

[0052] by For example, is the possible line downlink. and The site has the same name; and Sites with the same name; if It is the entrance of the station. It is the exit of the station, and vice versa. and between and and The shortest path between them is the bus network.

[0053] contrast and These two unidirectional lines form a complete alternative line if they meet the following uplink and downlink pairing constraints: 1) Sites with the same name account for more than 70%; 2) The line length difference does not exceed 10%.

[0054] right Do similar processing.

[0055] Step 6.4: Generate a line pool based on the obtained complete lines. For each complete line (including uplink and downlink) generated in step 6.3: 1) If the line length is within the threshold, it is deduplicated and added to the line pool.

[0056] 2) Select a route endpoint between Station O and Station D as a split point, splitting the route into two. This ensures both direct OD coverage and single-transfer coverage. For each split, check the uplink and downlink pairing constraints and route length constraints, performing a deduplication check. If the constraints are met, the route is added to the route pool.

[0057] 3) If any of the above routes overlap, they are merged. This is done to reduce duplication and minimize route length. For each merge, the upstream and downstream pairing constraints and route length constraints are checked for duplicates. If they meet the requirements, the route is added to the route pool.

[0058] Step 7: Calculate the coverage relationship between the routes in the candidate route pool and the target OD, and then optimize the bus network by finding a route set that covers all target ODs with the shortest mileage. This step first calculates the coverage relationship between the routes in the route pool and the target OD, and then solves the operations research model to find a route set that covers all target ODs with the shortest mileage.

[0059] Calculate the coverage relationship between the line pool and the target OD: use Indicates the target Collection, use Indicates one of , For each Find the one that can directly reach the Line collection , and a collection of transfer stations that can be covered by one transfer . Note that if passengers can take the line from Site to Then walk to , then take the line from Site to Station, transfer station is still recorded as .

[0060] In addition, the remaining bus routes need to be considered when calculating the coverage of a transfer. However, the combination of the remaining routes and the routes in the route pool is possible to cover the target with one transfer. of.

[0061] Modeling and solving network optimization solutions: remember is the set of lines in the line pool, where the lines are , , the length of each line is .

[0062] remember To achieve direct coverage A collection of lines.

[0063] remember To be able to of From station to one transfer station A collection of lines.

[0064] remember To be able to A transfer station arrive of The line of the station, or Walking transfer station arrive of A collection of lines for a site.

[0065] remember for A collection of transfer stations.

[0066] Decision variables A 0-1 Boolean variable indicating whether a line is selected .

[0067] Decision variables is a 0-1 Boolean variable, indicating Whether the line is directly covered.

[0068] Decision variables is a 0-1 Boolean variable, indicating Whether to pass the site Covered by one transfer.

[0069] The objective function is to minimize the length of the selected line: ; Constraint 1 - Every All need to be covered, either directly or with one transfer: ; Constraint 2 - If If you are directly covered, you must select the corresponding line: ; Constraint 3 - If Being on the site One transfer coverage, then you must choose a line that can Site to A line of stations: ; Constraint 4 - If Being on the site One transfer coverage, then you must choose a line that can Site to Site or from Site to A line of stations: ; Solving the above 0-1 integer programming problem will yield the network optimization solution.

[0070] The bus network optimization system according to the embodiment of the present invention includes: Data preparation module: obtain urban traffic network data, bus stop data of existing bus routes, station data, passenger flow OD, and routes that need to be adjusted; Station mapping module: maps the bus stops of existing bus routes to the urban transportation network; Road network update module: updates urban traffic road network data according to mapping results; Public transportation network generation module: connects adjacent bus stops to form a public transportation network; OD determination module: finds the OD that the planned route needs to cover based on the route that needs to be adjusted; Line pool generation module: generates a candidate line pool based on the ODs that need to be covered; Line network optimization module: Calculates the coverage relationship between the routes in the candidate line pool and the target OD, and then optimizes the bus line network by finding a set of routes that can cover all target ODs and has the shortest mileage.

[0071] As an implementation method, the line pool generation module generates a line pool according to the following steps: Calculate several shortest paths from O to D in the bus network; Extend the path to the feasible line end point and the nearest station; Generate alternative complete routes based on the expanded path; Generate a line pool based on the obtained complete lines.

[0072] As an implementation method, the line pool generation module searches for the downlink line corresponding to each line respectively, and forms a complete candidate line if the following uplink and downlink pairing constraints are met: (1) Sites with the same name account for more than 70%; (2) The line length difference does not exceed 10%; The line pool generation module generates each complete line: 1) If the line length is within the threshold, it is deduplicated and added to the line pool.

[0073] 2) Select a site between sites O and D that can serve as a line endpoint as a split point, splitting the line into two. For each split, check the uplink and downlink pairing constraints and line length constraints, and perform a deduplication check. If the conditions are met, the line is added to the line pool.

[0074] 3) If any of the above lines are connected end-to-end, they are merged. For each merge scenario, the uplink and downlink pairing constraints and line length constraints are checked and duplicate removal is performed. If they meet the requirements, the line is added to the line pool as a new line.

[0075] As an implementation method, the line network optimization module models and solves the line network optimization solution: use Indicates the target Collection, use Indicates one of , For each Find the one that can directly reach the Line collection , and a collection of transfer stations that can be covered by one transfer ; remember is the set of lines in the line pool, where the lines are , , the length of each line is ;remember To achieve direct coverage The line collection of To be able to of From station to one transfer station A collection of lines; To be able to A transfer station arrive of The line of the station, or Walking transfer station arrive of A collection of lines of stations; for The set of transfer stations; decision variables A 0-1 Boolean variable indicating whether a line is selected ; Decision variables is a 0-1 Boolean variable, indicating Whether it is directly covered by the line; decision variables is a 0-1 Boolean variable, indicating Whether to pass the site One transfer coverage; The objective function is to minimize the length of the selected line: ; Constraint 1 - Every All need to be covered, either directly or with one transfer: ; Constraint 2 - If If you are directly covered, you must select the corresponding line: ; Constraint 3 - If Being on the site One transfer coverage, then you must choose a line that can Site to A line of stations: ; Constraint 4 - If Being on the site One transfer coverage, then you must choose a line that can Site to Site or from Site to A line of stations: ; The above 0-1 integer programming problem is solved to obtain the network optimization solution.

[0076] As an implementation method, if passengers can take the line from Site to Then walk to , then take the line from Site to Station, transfer station is still recorded as .

[0077] While 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 these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A bus network optimization method, characterized in that: include: Step 1: Obtain urban traffic network data, bus stop data of existing bus routes, bus station data, passenger flow OD, and routes that need to be adjusted; Step 2: Map the bus stops of existing bus routes to the urban transportation network; Step 3: Update urban traffic network data based on the mapping results; Step 4: Connect adjacent bus stops to form a bus transportation network; Step 5: Find the target OD that the planned route needs to cover based on the route that needs to be adjusted; Step 6: Generate a candidate line pool based on the target OD that needs to be covered; Step 7: Calculate the coverage relationship between the routes in the candidate route pool and the target OD, and then optimize the bus network by finding a set of routes that can cover all target ODs and have the shortest mileage.

2. The bus network optimization method according to claim 1, characterized in that: Step 6 includes the following steps: Step 6.1: Calculate several shortest paths from O to D in the bus network; Step 6.2: Extend the path to the feasible line endpoints and the nearest station; Step 6.3: Generate an alternative complete route based on the expanded path; Step 6.4: Generate a line pool based on the obtained complete lines.

3. The bus network optimization method according to claim 2, characterized in that: In step 6.3, the downlink routes corresponding to each line are searched separately. If the following uplink and downlink pairing constraints are met, a complete candidate route is formed: (1) Sites with the same name account for more than 70%; (2) The line length difference does not exceed 10%; In step 6.4, for each complete line generated: 1) If the line length is within the threshold, it is deduplicated and added to the line pool. 2) Select a site between sites O and D that can serve as a line endpoint as a split point, splitting the line into two. For each split, check the uplink and downlink pairing constraints and line length constraints and perform a deduplication check. If the constraints are met, the line is added to the line pool. 3) If any of the above lines are connected end-to-end, they are merged. For each merge scenario, the uplink and downlink pairing constraints and line length constraints are checked and duplicate removal is performed. If they meet the requirements, the line is added to the line pool as a new line.

4. The bus network optimization method according to claim 1, wherein: In step 7, model and solve the network optimization solution: use Indicates the target Collection, use Indicates one of , For each Find the one that can directly reach the Line collection , and a collection of transfer stations that can be covered by one transfer ; remember is the set of lines in the line pool, where the lines are , , the length of each line is ;remember To achieve direct coverage The line collection of To be able to of From station to one transfer station A collection of lines; To be able to A transfer station arrive of The line of the station, or Walking transfer station arrive of A collection of lines of stations; for The set of transfer stations; decision variables A 0-1 Boolean variable indicating whether a line is selected ; Decision variables is a 0-1 Boolean variable, indicating Whether it is directly covered by the line; decision variables is a 0-1 Boolean variable, indicating Whether to pass the site One transfer coverage; The objective function is to minimize the length of the selected line: ； Constraint 1 - Every All need to be covered, either directly or with one transfer: ； Constraint 2 - If If you are directly covered, you must select the corresponding line: ； Constraint 3 - If Being on the site One transfer coverage, then you must choose a line that can Site to A line of stations: ； Constraint 4 - If Being on the site One transfer coverage, then you must choose a line that can Site to Site or from Site to A line of stations: ； The above 0-1 integer programming problem is solved to obtain the network optimization solution.

5. The bus network optimization method according to claim 4, characterized in that: In step 7, if the passenger can take the line from Site to Then walk to , then take the line from Site to Station, transfer station is still recorded as .

6. A bus network optimization system, characterized in that: include: Data preparation module: obtain urban traffic network data, bus stop data of existing bus routes, station data, passenger flow OD, and routes that need to be adjusted; Station mapping module: maps the bus stops of existing bus routes to the urban transportation network; Road network update module: updates urban traffic road network data according to mapping results; Public transportation network generation module: connects adjacent bus stops to form a public transportation network; OD determination module: finds the target OD that the planned route needs to cover based on the route that needs to be adjusted; Line pool generation module: generates a candidate line pool based on the target OD that needs to be covered; Line network optimization module: Calculates the coverage relationship between the routes in the candidate line pool and the target OD, and then optimizes the bus line network by finding a set of routes that can cover all target ODs and has the shortest mileage.

7. The bus network optimization system according to claim 6, characterized in that: The line pool generation module generates a line pool according to the following steps: Calculate several shortest paths from O to D in the bus network; Extend the path to the feasible line end point and the nearest station; Generate alternative complete routes based on the expanded path; Generate a line pool based on the obtained complete lines.

8. The bus network optimization system according to claim 7, characterized in that: The line pool generation module searches for the downlink line corresponding to each line. If the following uplink and downlink pairing constraints are met, a complete alternative line is formed: (1) Sites with the same name account for more than 70%; (2) The line length difference does not exceed 10%; The line pool generation module generates each complete line: 1) If the line length is within the threshold, it is deduplicated and added to the line pool. 2) Select a site between sites O and D that can serve as a line endpoint as a split point, splitting the line into two. For each split, check the uplink and downlink pairing constraints and line length constraints and perform a deduplication check. If the constraints are met, the line is added to the line pool. 3) If any of the above lines are connected end-to-end, they are merged. For each merge scenario, the uplink and downlink pairing constraints and line length constraints are checked and duplicate removal is performed. If they meet the requirements, the line is added to the line pool as a new line.

9. The bus network optimization system according to claim 6, characterized in that: The line network optimization module models and solves the line network optimization solution: use Indicates the target Collection, use Indicates one of , For each Find the one that can directly reach the Line collection , and a collection of transfer stations that can be covered by one transfer ; remember is the set of lines in the line pool, where the lines are , , the length of each line is ;remember To achieve direct coverage The line collection of To be able to of From station to one transfer station A collection of lines; To be able to A transfer station arrive of The line of the station, or Walking transfer station arrive of A collection of lines of stations; for The set of transfer stations; decision variables A 0-1 Boolean variable indicating whether a line is selected ; Decision variables is a 0-1 Boolean variable, indicating Whether it is directly covered by the line; decision variables is a 0-1 Boolean variable, indicating Whether to pass the site One transfer coverage; The objective function is to minimize the length of the selected line: ； Constraint 1 - Every All need to be covered, either directly or with one transfer: ； Constraint 2 - If If you are directly covered, you must select the corresponding line: ； Constraint 3 - If Being on the site One transfer coverage, then you must choose a line that can Site to A line of stations: ； Constraint 4 - If Being on the site One transfer coverage, then you must choose a line that can Site to Site or from Site to A line of stations: ； The above 0-1 integer programming problem is solved to obtain the network optimization solution.

10. The bus network optimization system according to claim 9, characterized in that: If passengers can take the line from Site to Then walk to , then take the line from Site to Station, transfer station is still recorded as .

Images