Bus line network optimization method and system

Abstract

The application discloses a kind of bus line network optimization method and system, the method includes: step 1: obtaining city traffic road network data, bus station data, station data, passenger flow OD, line needing adjustment;Step 2: bus station is mapped to city traffic road network;Step 3: update city traffic road network data;Step 4: connect adjacent bus station and form bus traffic network;Step 5: find out the target OD that planning route needs to cover;Step 6: generate candidate line pool;Step 7: calculate the covering relationship of line in candidate line pool and target OD, then through finding out the line set that can cover all target OD and the shortest mileage, realize bus line network optimization.The application can solve the line network optimization adjustment problem of common asymmetric traffic network, greatly improve the applicability of line network optimization method.Especially in the case of current bus operation cost pressure, realize the reduction of operation cost while not reducing service level.

Description

TECHNICAL FIELD

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

[0002] Bus planning includes line network design, line frequency setting, timetable development, vehicle scheduling and driver scheduling. This patent focuses on the first stage. With the continuous development of the city and the integration of various transportation modes, it is necessary to periodically optimize and adjust the bus line network. In particular, the current bus operation is under great pressure, and it is urgent to reduce operating costs while ensuring service levels. For example, a city has nearly a thousand regular bus lines, and the line network is optimized and adjusted once or twice a year, with the number of lines adjusted generally less than 10% each time.

[0003] There are many related studies on bus line network optimization. Currently, the main method for line network optimization is mathematical programming, and in recent years, methods for designing lines through reinforcement learning have emerged.

[0004] There are two types of methods for line optimization in mathematical programming:

[0005] 1) Directly take the edges in the traffic network as decision variables, and use flow balance, line length, station spacing, non-straight line coefficient, etc. as constraints. The optimization goal is to minimize cost or maximize passenger flow. This method can theoretically obtain the optimal solution. However, the model established in this way is large, and the number of decision variables and constraints is large, which is only suitable for very small-scale networks.

[0006] 2) Use heuristic rules to generate a line pool, and then select lines from the line pool through integer programming modeling. The optimization goal is also to minimize cost, and the constraint is passenger flow coverage. A key to this approach is how to generate the line pool. Enumerating all lines will cause combinatorial explosion. Generally, industry knowledge and experience are used to give an appropriate size of the line pool, and then the most cost-effective line is selected from the line pool as the optimization result.

[0007] The reinforcement learning method gradually learns the optimized line network design through trial and error and rewards. This method requires a lot of computing power, and is currently more of an academic discussion, and can only be tested on small-scale networks.

[0008] The existing line network planning methods have the following problems:

[0009] 1. Only applicable to symmetric traffic networks.

[0010] Symmetry means that the topology between two network nodes is symmetric. For example, if edge A→B is in the traffic network, then edge B→A is also in the traffic network, where A and B are nodes (bus stops) of the traffic network. Accordingly, the current research also assumes that the uplink and downlink of a line are also symmetric, i.e., the nodes contained by the uplink and downlink are the same, and the order is reversed. For example, as shown in Figure 1 the traffic network is symmetric, the network nodes are bus stops, and the lengths of the paths between two adjacent stops are the same, and the directions are opposite.

[0011] However, in the actual traffic network, many are asymmetric:

[0012] a. There are asymmetric stops, i.e., there are no same-named stops on the opposite side of the road. For example, Figure 2 In (a), the "XFHA" stop in a city is only on one side of the road and does not exist on the other side. In a city, 10% of the bus stops are asymmetric.

[0013] b. There are topological asymmetries between stops. For example, Figure 2 In (b), there is a straight path plus a right turn from "HFJRG (south)" to "HD (west)", while from "HD (east)" to "HFJRG (north)" requires a straight path, a U-turn, and then a left turn, i.e., the traffic network is not symmetric. In fact, as long as two same-named stops are on both sides of an intersection, such an asymmetric problem can exist, and the number of such stops is not small.

[0014] 2. The influence of the station on the line operation cost is not considered.

[0015] One of the two end stops of a line is usually a station or near a station, otherwise it will definitely cause empty running waste. SUMMARY

[0016] The technical problem to be solved by the embodiments of the present application is to provide a bus line network optimization method and system to give a more practical bus line network optimization adjustment.

[0017] In order to solve the above technical problem, the embodiments of the present application propose a bus line network optimization method, comprising:

[0018] Step 1: obtaining city traffic network data, existing bus stop data of bus lines, station data, passenger flow OD, and lines to be adjusted;

[0019] Step 2: mapping the bus stops of the existing bus lines to the city traffic network;

[0020] Step 3: updating the city traffic network data according to the mapping result;

[0021] Step 4: connecting adjacent bus stops to form a bus traffic network;

[0022] Step 5: find out target ODs that need to be covered by the planned route according to the line to be adjusted;

[0023] Step 6: generate a candidate line pool according to the target ODs to be covered;

[0024] Step 7: calculate the coverage relationship between the lines in the candidate line pool and the target ODs, and then realize the optimization of the bus line network by finding out the line set that can cover all the target ODs and has the shortest mileage.

[0025] Correspondingly, the embodiment of the application also provides a bus line network optimization system, comprising:

[0026] The data preparation module: acquires the urban traffic network data, the bus station data of the existing bus line, the station data, the passenger flow OD and the line to be adjusted;

[0027] The station mapping module: maps the bus stations of the existing bus line to the urban traffic network;

[0028] The network updating module: updates the urban traffic network data according to the mapping result;

[0029] The bus network generation module: connects adjacent bus stations to form a bus network;

[0030] The OD determination module: finds out the target ODs that need to be covered by the planned route according to the line to be adjusted;

[0031] The line pool generation module: generates a candidate line pool according to the target ODs to be covered;

[0032] The line network optimization module: calculates the coverage relationship between the lines in the candidate line pool and the target ODs, and then realizes the optimization of the bus line network by finding out the line set that can cover all the target ODs and has the shortest mileage.

[0033] The application has the advantages that the application can solve the common non-symmetrical traffic network line network optimization adjustment problem, greatly improve the applicability of the line network optimization method, and realize the reduction of the operation cost while not reducing the service level under the current situation of the pressure of the bus operation cost. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 is a schematic diagram of a symmetrical traffic network.

[0035] Figure 2 (a) is a schematic diagram of an asymmetrical station; (b) is a schematic diagram of an asymmetrical traffic network.

[0036] Figure 3 is a flowchart of the bus line network optimization method of the embodiment of the application.

[0037] Figure 4 is a schematic diagram of mapping bus stops to urban traffic road network according to an embodiment of the present application.

[0038] Figure 5 is a schematic diagram of updated urban traffic road network data according to an embodiment of the present application.

[0039] Figure 6 is a schematic diagram of bus traffic network according to an embodiment of the present application.

[0040] Figure 7 is a schematic diagram of extended OD path according to an embodiment of the present application. DETAILED DESCRIPTION

[0041] It should be noted that the embodiments and features of the embodiments in the present application can be combined with each other without conflict, and the present application will be further described in detail below in conjunction with the drawings and specific embodiments.

[0042] In the embodiments of the present application, if there is a directional indication (such as up, down, left, right, front, back, etc.), it is only used to explain the relative position relationship, movement condition, etc. between components in a certain specific posture (as shown in the drawings), and if the specific posture changes, the directional indication will also change accordingly.

[0043] In addition, in the present application, if the description involves "first", "second", etc., it is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include at least one of the features.

[0044] The present application is aimed at the bus line network optimization and adjustment problem of reducing operation cost and maintaining service level faced by the current transportation industry, especially to solve the following practical problems, and to take it as the basis for subsequent bus planning stage:

[0045] 1) The traffic network is asymmetric;

[0046] 2) One end of the line needs to be a station or near a station.

[0047] The present application also adopts a mathematical programming method, which will assume that the existing bus stops do not change, and then try to construct a candidate line pool and select appropriate lines from the pool as the optimization result.

[0048] The objective of mathematical programming is to minimize the operation cost. Since the operation cost that can be reduced by line network optimization is mainly the dynamic operation cost, i.e. the total operation mileage, and the total operation mileage is positively correlated with the total line length. Therefore, the optimization objective here is to minimize the line mileage.

[0049] Service level is mainly reflected by travel time. Assuming that the running speed of the bus does not change much before and after the online network adjustment, then the travel time that can be affected by the network optimization is actually the travel distance and the number of transfers. Here, the number of transfers is required to be no more than once. Since the waiting time is determined by the line frequency stage setting in the second phase of the bus planning, the service level is not reduced in the network design stage of the bus planning, which is to increase the travel distance.

[0050] The main idea of the bus network optimization of the application is: 1) forming a bus network (a traffic network between bus stations. Represented by a directed graph The vertices of the graph are bus stations, and the edges of the graph are paths between stations) according to the input data; 2) finding the target OD (OD, i.e. the origin and destination of the trip, Origin and Destination), and constructing a candidate line pool according to the target OD; 3) selecting a line from the line pool to realize network optimization.

[0051] Please refer to Figure 3 The bus network optimization method of the embodiment of the application comprises steps 1-7.

[0052] Step 1: data preparation, obtaining city traffic network data, existing bus route bus station data, station data, passenger flow OD, and lines to be adjusted.

[0053] City traffic network data: including node and road segment data, road segment connecting adjacent two nodes, node is usually the intersection of two roads. The data is owned by the map company or the land planning department.

[0054] Station data of existing bus routes: stations included by each line, station number, ID, name, latitude and longitude, direction angle of the station and distance between adjacent stations. The data is owned by the bus company or the traffic bureau.

[0055] Station data: ID, name, latitude and longitude of the entrance and exit of the station.

[0056] Passenger flow OD: origin and destination stations, travel distance, and number of transfers of existing bus passenger flow. The data is not directly owned by the bus company or the traffic bureau, but the passenger flow OD can be calculated through the electronic payment data of the bus.

[0057] Lines to be adjusted: planning personnel usually specify lines to be adjusted, which are usually considered to be adjusted because of high repetition, poor efficiency or too winding line shape, etc.

[0058] Step 2: Map the bus stops of the existing bus routes to the urban traffic network (a spatial network infrastructure that organizes and guides the flow of buses, represented by road segments and their two end nodes. Each road segment has a start node snode and an end node enode. If the road's traffic direction is snode->enode, the road segment is a forward road; if the road's traffic direction is enode->snode, the road segment is a reverse road; if the road can be traveled in both directions, the road segment is a bidirectional road). This step establishes the relationship between stops and the road network, and the final information to be provided is: 1) which road segment each stop is projected onto in the road network; 2) the specific projection point coordinates; 3) the direction of the stop and the orientation of the road segment (the stop and the road segment are in the same direction: if the vehicle travels through the stop on the road segment snode->enode, the stop and the road segment are in the same direction, otherwise the stop and the road segment are in opposite directions). As shown in Figure 4 , the larger dots are stops and the smaller dots are road network nodes, and the road network segments connecting the road network nodes.

[0059] Bus stop data and urban road network data usually come from different sources, and simply matching by distance is not suitable for some stops, so a more complex method is needed. Specifically, not only the location of the stop is used, but also the direction of the stop and the distance between adjacent stops. The related algorithm is well known in the art and is not the focus of the present invention, so it will not be described in detail here.

[0060] Step 3: Update the urban traffic network data according to the mapping results. This step is to update the road network data and add the stops to the node set of the road network. The road segment where the stop is located also needs to be split accordingly. This is to prepare for subsequent path search using Dijkstra or A* algorithm. As shown in Figure 5 , the larger dots are the projection points of the stops on the road segments, i.e. they have become nodes of the road segments.

[0061] Step 4: Connect adjacent bus stops to form a bus transportation network. This step is to search for the paths between adjacent bus stops and ultimately form a bus transportation network. This bus transportation network is the basis for building a bus route pool in the subsequent step.

[0062] The specific method is as follows: traverse any two non-identical stops and calculate the distance of the path between them on the urban traffic network. If the distance is less than a certain threshold and there are no other bus stops on the path, then the path is part of the bus transportation network. The distance threshold can be set to 2 km. Because it has been ensured that there are no other bus stops on the path, the actual distance between two adjacent stops is generally not more than 600 meters.

[0063] As shown in Figure 6 , the dots are bus stops and edges may be connected between two bus stops.Figure 6 The sections highlighted in medium to high brightness represent one edge of the public transportation network.

[0064] Step 5: Identify the target origin-destination (OD) areas that the planned route needs to cover, based on the route adjustments required. First, identify the relevant OD areas based on the route adjustments. Then, check if these OD areas can be covered by the remaining public transport network. The remaining public transport network is the existing network formed by removing the routes that need adjustment. OD areas that cannot be covered by the remaining public transport network are the OD areas that the planned route needs to cover. Coverage here means that passengers can travel directly from OD to D via the public transport network, or from OD to D with one transfer.

[0065] In actual calculations, small-volume origin-destination (OD) requests (such as those with less than one person per day) can be removed to arrive at the final target OD. This is because public transportation primarily serves OD requests with certain patterns. OD requests with very small volumes are usually random travel demands and do not need to be specifically considered.

[0066] Step 6: Generate a candidate line pool based on the target OD to be covered.

[0067] First, the target bus routes are all in the route pool, and they will also serve as initial values ​​for subsequent optimization solutions. Then, a set of candidate routes is generated for each OD (Original Design Location). The specific steps are described below:

[0068] Step 6.1: Use Yen's algorithm to calculate the k shortest paths from O to D in the public transport network. The value of k can be chosen based on available computing resources; a larger k will result in better optimization.

[0069] Step 6.2: Extend the path to feasible line endpoints and the nearest depot. If a station name has two corresponding stations, that station can be used as a line endpoint. This step extends the path result from the previous step in both directions to ensure that both ends can be line endpoints, and further extends it to the nearest depot entrance / exit.

[0070] like Figure 7 As shown, The site is The site extends forward, if If the station itself can serve as the endpoint of the line, then no extension is needed. The site is Site. There may be multiple site options; you need to select the site's orientation angle and... The site with the closest similarity should be extended vertically as much as possible. Similar to this... Backwards from the site Site extension. The station continues forward to the nearest station entrance / exit. , The station extends backward to the nearest terminal entrance .

[0071] Step 6.3: Generate alternative complete routes according to the extended routes. For each route in step 6.2 is an up route, is another up route. Search for their corresponding down routes respectively.

[0072] Take for example, is a possible down route. Where and are the same station; and are the same station; if is the entrance of the terminal then is the exit of the terminal, and vice versa. and are the shortest paths in the public transit network. and are the shortest paths in the public transit network.

[0073] Compare and two one-way routes, if the following up-down pairing constraints are met, then they form an alternative complete route:

[0074] 1) The same station accounts for more than 70%;

[0075] 2) The route length difference is no more than 10%.

[0076] Do similar processing for .

[0077] Step 6.4: Generate route pool according to the complete routes obtained. For each complete route (including up and down) generated in step 6.3:

[0078] 1) If the route length is within the threshold range, it is checked for duplication and then added as a route in the route pool.

[0079] 2) Select the passing station between the O station and the D station that can be used as the end point of the route as the splitting point to split the route into two. The purpose of this is to give both OD direct coverage and one-transfer coverage opportunities. Check the up-down pairing constraint and the route length constraint for each splitting case and check for duplication. If it meets the requirements, it is a new route in the route pool.

[0080] 3) If the above routes have a head-to-tail relationship, try to merge them. The purpose of this is to reduce the repeated parts and increase the possibility of minimizing the route mileage. Check the up-and-down pairing constraints and route length constraints for each merging case and do a de-duplication check. If it is satisfied, it is a new route in the route pool.

[0081] Step 7: Calculate the coverage relationship between the routes in the candidate route pool and the target ODs, and then find the route set that can cover all target ODs and has the shortest mileage to achieve bus route network optimization. This step first calculates the coverage relationship between the routes in the route pool and the target ODs, and then finds the route set that can cover all target ODs and has the shortest mileage by solving the operations model.

[0082] Calculate the coverage relationship between the routes in the route pool and the target ODs:

[0083] Let represent the target OD set, and represent one of them. , . For each , find the route set that can directly cover it and the transfer station set that can cover it once. Note that if a passenger can take route from to , then walk to , and take route from to , the transfer station is still recorded as .

[0084] In addition, when calculating the one-time transfer coverage, the remaining bus routes need to be considered. Although the remaining bus routes cannot cover the target OD , the combination of the remaining routes and the routes in the route pool may be able to cover the target OD through one transfer.

[0085] Modeling and solving the bus route network optimization scheme:

[0086] Let be the set of routes in the route pool, where the routes are , , and the length of each route is .

[0087] Let be the set of routes that can directly cover .

[0088] Let​​ Let be the set of lines from to the one-way transfer station .

[0089] Let be the set of lines from to the one-way transfer station . or the set of lines from to the walking transfer station . .

[0090] Let be the set of one-way transfer stations.

[0091] Decision variable is a 0-1 boolean variable indicating whether the line is selected.

[0092] Decision variable is a 0-1 boolean variable indicating whether the station is covered by the line directly.

[0093] Decision variable is a 0-1 boolean variable indicating whether the station is covered by a one-way transfer from .

[0094] The objective function is to minimize the length of the selected lines:

[0095] ;

[0096] Constraint 1 - Each needs to be covered, either directly or by a one-way transfer:

[0097] ;

[0098] Constraint 2 - If is covered directly, then the corresponding line must be selected:

[0099] ;

[0100] Constraint 3 - If is covered by a one-way transfer at station , then a line must be selected that can go from to :

[0101] ​​​ ;

[0102] Constraint 4 - If a transfer is covered by a station , then a route must be selected that can cover the transfer from to or from to :

[0103] ;

[0104] Solving the above 0-1 integer programming problem obtains the optimization scheme of the bus line network.

[0105] The bus line network optimization system of the embodiment of the application comprises:

[0106] a data preparation module configured to obtain city traffic road network data, bus station data of existing bus routes, station data, passenger flow OD, and a route to be adjusted;

[0107] a station mapping module configured to map the bus stations of the existing bus routes to the city traffic road network;

[0108] a road network updating module configured to update the city traffic road network data according to the mapping result;

[0109] a bus traffic network generating module configured to connect adjacent bus stations to form a bus traffic network;

[0110] an OD determining module configured to find out the OD to be covered by the planned route according to the route to be adjusted;

[0111] a route pool generating module configured to generate a candidate route pool according to the OD to be covered;

[0112] a line network optimization module configured to calculate the coverage relationship between the routes in the candidate route pool and the target OD, and then find out a route set that can cover all the target OD and has the shortest mileage, so as to realize the optimization of the bus line network.

[0113] As an implementation mode, the route pool generating module generates the route pool according to the following steps:

[0114] calculate several shortest paths from O to D in the bus traffic network;

[0115] extend the paths to the feasible route endpoints and the nearest stations;

[0116] generate the complete routes according to the extended paths;

[0117] generate the route pool according to the complete routes obtained.

[0118] As an implementation, the line pool generation module searches for each line pair corresponding downlink, if the following uplink and downlink pairing constraint conditions are met, a complete line is formed:

[0119] (1) The same station occupies more than 70%;

[0120] (2) The line length difference is not more than 10%;

[0121] The line pool generation module generates each complete line:

[0122] 1) If the line length is within the threshold range, it is checked after de-duplication and is a line in the line pool.

[0123] 2) Select the passing station between the O station and the D station as the splitting point, split the line into two. Check the uplink and downlink pairing constraint and the line length constraint for each splitting case and do de-duplication check. If it is satisfied, it is a new line in the line pool.

[0124] 3) If the above line has a head-to-tail connection relationship, try to merge, check the uplink and downlink pairing constraint and the line length constraint for each merging case and do de-duplication check. If it is satisfied, it is a new line in the line pool.

[0125] As an implementation, the line network optimization module models and solves the line network optimization scheme:

[0126] Let represent the target set, and let represent one of them , ; for each , find the line set that can directly cover the , and the transfer station set that can cover it once ;

[0127] Let be the set of lines in the line pool, where the line is , , and the length of each line is ; let be the line set that can directly cover ; let be the line set that can transfer from the station of to the transfer station ; let be the line set that can transfer from the transfer station of to ​​ The station's route, or pedestrian transfer station arrive of A collection of routes at each station; (Note: The original text appears to be incomplete and contains several errors. A more accurate translation would for A set of transfer stations; decision variables This is a 0-1 Boolean variable, indicating whether a line is selected. Decision variables A 0-1 Boolean variable, representing Whether it is directly covered by a line; decision variables A 0-1 Boolean variable, representing Whether through the site One-transfer coverage;

[0128] The objective function is to minimize the length of the selected path:

[0129] ;

[0130] Constraint 1 - Each Both need to be covered, either through direct routes or with a single transfer:

[0131] ;

[0132] Constraint 2 - If If the route is directly covered, then you must select the corresponding line:

[0133] ;

[0134] Constraint 3 - If On the site If a single transfer is required, then a route must be selected that connects to... Station to One of the lines at the station:

[0135] ;

[0136] Constraint 4 - If On the site If a single transfer is required, then a route must be selected that connects to... Station to Site or from Station to One of the lines at the station:

[0137] ;

[0138] Solving the above 0-1 integer programming problem yields a network optimization scheme.

[0139] As one embodiment, if the passenger can take a line From the station to then walk to then take a line from the station to the station, still noting the transfer station as .

[0140] While embodiments of the application have been shown and described, it is to be understood that the application is not limited to the details of the embodiments described, since modifications can be made by those skilled in the art, without departing from the spirit and scope of the application, which is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing a public transport network, characterized in that, Comprising: Step 1: Obtain urban traffic network data, bus station data of existing bus routes, station data, passenger flow OD, and routes that need to be adjusted; Step 2: Map the bus stations of the existing bus routes to the urban traffic network; Step 3: Update the urban traffic network data according to the mapping results; Step 4: Connect adjacent bus stations to form a bus transportation network; Step 5: Find the target OD that needs to be covered by the planned route according to the route that needs to be adjusted; Step 6: Generate a candidate route pool according to 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 find the route set that can cover all target ODs and has the shortest mileage to achieve bus route network optimization; Step 6 includes the following steps: Step 6.1: Calculate several shortest paths from O to D in the bus transportation network; Step 6.2: Extend the path to the feasible line endpoints and the nearest station; Step 6.3: Generate a complete candidate route according to the extended path; Step 6.4: Generate a route pool according to the complete route obtained; In step 7, model and solve the route network optimization scheme: Use to represent the target set, use to represent one of them , ; for each find a set of lines that can directly cover the set of transfer sites and a set of transfer sites that can cover once ; ; remember This refers to the collection of lines in the line pool, where the lines are... , The length of each line is ;remember In order to achieve direct coverage The set of routes; In order to be able to from of Station to one transfer station A collection of routes; (Note: The original text appears to be incomplete and contains several grammatical errors. A more accurate translation In order to be able to from One transfer station arrive of The station's route, or pedestrian transfer station arrive of A collection of routes at each station; (Note: The original text appears to be incomplete and contains several errors. A more accurate translation would for A set of transfer stations; decision variables This is a 0-1 Boolean variable, indicating whether a line is selected. Decision variables A 0-1 Boolean variable, representing Whether it is directly covered by a line; decision variables A 0-1 Boolean variable, representing Whether through the site One-transfer coverage; The objective function is to minimize the length of the selected route: ； Constraint 1 - Each All need to be covered, direct coverage or one transfer coverage: ； Constraint 2 - If If the direct coverage is selected, the corresponding line must be chosen: ； Constraint 3 - If is at a station that is covered by a transfer, then one of the routes from the station to the station must be selected: ； Constraint 4 - If is covered by a transfer, then one of the routes must be selected that is able to go from to or from to to . ； Solve the above 0-1 integer programming problem to obtain the route network optimization scheme.

2. The transit line network optimization method of claim 1, wherein, In step 6.3, search for the corresponding downlink for each line, and if the following uplink-downlink pairing constraint conditions are met, a complete candidate route is formed: (1) The proportion of stations with the same name exceeds 70%; (2) The length of the route differs by no more than 10%; In step 6.4, for each complete route generated: 1) If the route length is within the threshold range, it is checked for duplicates and then used as a route in the route pool; 2) Select the passing stations between the O station and the D station as splitting points to split the route into two more routes, and check the uplink-downlink pairing constraint and the route length constraint for each splitting case and check for duplicates. If it meets the condition, it is a new route in the route pool; 3) If the above routes have a head-to-tail connection relationship, attempt to merge, and check the uplink-downlink pairing constraint and the route length constraint for each merging case and check for duplicates. If it meets the condition, it is a new route in the route pool.

3. The transit line network optimization method of claim 1, wherein, In step 7, if the passenger can take the line from the station to then walk to then take the line from the station still remembering the transfer station as 4. A public transit network optimization system, characterized by, Comprising: Data preparation module: Obtain urban traffic network data, bus station data of existing bus routes, station data, passenger flow OD, and routes that need to be adjusted; Station mapping module: Map the bus stations of the existing bus routes to the urban traffic network; Road network update module: Update the urban traffic network data according to the mapping results; Bus transportation network generation module: Connect adjacent bus stations to form a bus transportation network; OD determination module: Find the target OD that needs to be covered by the planned route according to the route that needs to be adjusted; Route pool generation module: Generate a candidate route pool according to the target OD that needs to be covered; Route network optimization module: Calculate the coverage relationship between the routes in the candidate route pool and the target OD, and then find the route set that can cover all target ODs and has the shortest mileage to achieve bus route network optimization; The route pool generation module generates the route pool according to the following steps: Calculate several shortest paths from O to D in the bus transportation network; Extend the path to the feasible line end points and the nearest stations; Generate alternative complete lines according to the extended path; Generate line pool according to the complete lines obtained; Line network optimization module models and solves line network optimization scheme: Use to represent the target set, use to represent one of them , ; for each find a set of lines that can directly cover the set of transfer sites that can cover it once , and a set of transfer sites that can cover it once ; remember This refers to the collection of lines in the line pool, where the lines are... , The length of each line is ;remember In order to achieve direct coverage A collection of routes; (Note) In order to be able to from of Station to one transfer station A collection of routes; (Note: The original text appears to be incomplete and contains several grammatical errors. A more accurate translation In order to be able to from One transfer station arrive of The station's route, or pedestrian transfer station arrive of A collection of routes at each station; (Note: The original text appears to be incomplete and contains several errors. A more accurate translation would for A set of transfer stations; decision variables This is a 0-1 Boolean variable, indicating whether a line is selected. ; decision variable is a 0-1 Boolean variable indicating whether covered by a line direct; decision variable is a 0-1 Boolean variable indicating whether covered by a station one transfer; Objective function is to minimize the length of selected lines: ； Constraint 1 - Each All need to be covered, direct coverage or one transfer coverage: ； Constraint 2 - If If the direct coverage is selected, the corresponding line must be chosen: ； Constraint 3 - If is covered by a transfer, then one of the routes from must be selected that is able to reach from : ； Constraint 4 - If is covered by a transfer, then one of the routes must be selected that is able to go from to or from to to . ； Solve the above 0-1 integer programming problem to obtain the line network optimization scheme.

5. The public transit network optimization system of claim 4, wherein, The line pool generation module searches for the corresponding downlink of each line, and if the following uplink-downlink pairing constraint conditions are met, it constitutes an alternative complete line: (1) The proportion of stations with the same name exceeds 70%; (2) The length of the line does not differ by more than 10%; The line pool generation module checks each complete line generated: 1) If the line length is within the threshold range, it is checked for duplication and used as a line in the line pool; 2) Select the passing station between the O station and the D station as the splitting point, split the line into two, check the uplink-downlink pairing constraint and the line length constraint for each splitting case, and check for duplication. If it meets the requirements, it is a new line in the line pool; 3) If the above lines have a head-to-tail connection relationship, try to merge, check the uplink-downlink pairing constraint and the line length constraint for each merging case, and check for duplication. If it meets the requirements, it is a new line in the line pool.

6. The public transit line network optimization system of claim 4, wherein, If the passenger can take a line from the station to then walk to then take a line from the station to the station, still noting the interchange station as .

Images