A method, device and electronic equipment for optimizing bus transfer network

Abstract

The present invention provides a method, device and electronic device for optimizing a bus transfer network. The optimization method comprises: obtaining a basic network and obtaining a planned number of bus routes, generating bus routes of the planned number of bus routes based on the basic network, networking the bus routes to form a bus route network, performing shortest path planning for each user based on the bus route network, and obtaining a single cost of each user's travel under the shortest path planning, obtaining a total cost of all users' travel based on all single costs, taking the total cost as a target value, judging whether the target value converges, if so, obtaining a bus route network, if not, optimizing the bus route network iteratively using a heuristic algorithm until the target value converges. The present invention provides an optimal transfer network solution without significantly increasing the solution space, and is suitable for solving large-scale urban-level bus networks, transportation networks, etc.

Description

Technical Field

[0001] The present invention relates to the technical field of public transportation network planning, and in particular to a method, device and electronic equipment for optimizing a public transportation transfer network. Background Art

[0002] With the continuous acceleration of urbanization and the acceleration of urban motorization, traffic congestion in many large and megacities has become increasingly severe, and traffic congestion has continued to spread to medium-sized cities. Urban public transportation is an important infrastructure closely related to people's lives. Its basic task is to provide passengers with safe, convenient, fast, punctual and comfortable riding conditions. Giving priority to the development of urban public transportation and increasing the proportion of public transportation will help alleviate the pressure faced by urban transportation. However, to increase the attractiveness of public transportation, it is not enough to simply increase the number of bus routes and buses. It is also necessary to optimize the structure of the bus network.

[0003] Research has found that the number of transfers is the first factor that most passengers consider when traveling. In the existing technology, when designing and optimizing the bus network, all nodes may be used as transfer points, that is, the solution to the problem needs to first assume the transfer points and then determine the vehicle's route. This method of pre-assuming transfer points leads to the expansion of the solution space, and the solution space increases exponentially on the basis of the original scale, increasing the difficulty of solving the problem. Summary of the invention

[0004] The problem solved by the present invention is that the transfer network optimization in the prior art has the problem that the solution space is significantly increased and the solution is difficult.

[0005] In order to solve the above problems, the present invention provides a method for optimizing a bus transfer network, comprising:

[0006] Get the basic network and the number of planned bus routes;

[0007] Generate the planned number of bus routes based on the basic network;

[0008] Networking the bus routes to form a bus route network;

[0009] Based on the bus route network, the shortest path planning is performed for each user respectively, and the individual cost of each user's travel under the shortest path planning is obtained, and the total cost of all users' travel is obtained based on all the individual costs;

[0010] Taking the total cost as the target value, determining whether the target value converges, if so, obtaining the bus route network, if not, using a heuristic algorithm to iteratively optimize the bus route network until the target value converges.

[0011] Preferably, obtaining the planned number of bus routes includes: using the following formula to determine the planned number of bus routes:

[0012] R∈(R1, R2), where

[0013] Among them, R is the number of planned bus routes, C w is the width of the city under study, C l is the length of the city under study, C s is the maximum and reasonable walking distance for users, R1 represents the number of lines required to lay a horizontal and vertical orthogonal bus line grid for the studied city under the condition of equal spacing, I is the total budget for the planned line investment, and P i is the average cost of a single line, and R2 represents the maximum number of lines that the planned budget can support.

[0014] Preferably, the basic network is a network having a node-connection structure, the basic network includes a basic node, and the bus line includes a first node;

[0015] The networking of the bus routes to form a bus route network comprises:

[0016] Obtaining a node in the first nodes that coincides with the base node as a first coincident node;

[0017] The first overlapping nodes in the R bus routes that overlap with the basic nodes are merged into one node as a route connection and transfer node.

[0018] Preferably, the obtaining of the individual cost of each user's travel under the shortest path planning comprises: determining the individual cost using the following formula:

[0019] C i =L i + kI,

[0020] Among them, i represents the i-th user, C i represents the individual cost of the ith user's travel under the shortest path planning, L i is the path length cost of the ith user traveling on the shortest road obtained by the shortest road planning, I is the cost of the ith user transferring once on the bus route network, and k is the number of transfers of the ith user on the bus route network.

[0021] Preferably, the obtaining the total cost of all users' travel based on all the individual costs comprises: determining the total cost using the following formula:

[0022] C j =∑C i ;

[0023] Where, j represents the jth bus route network, C j Represents the total cost of all users traveling on the j-th bus route network.

[0024] Preferably, the determining whether the target value converges includes:

[0025] Obtaining the ratio of the decrease value of the target value in the last 1 / 3 stage of iteration to the total decrease value;

[0026] Comparing whether the ratio is less than a set convergence speed threshold;

[0027] If so, determining that the target value converges;

[0028] If not, the bus route network is optimized iteratively until the target value converges.

[0029] Preferably, the bus routes for generating the planned number of bus routes based on the basic network include:

[0030] According to the line generation requirements, obtain the number of sample points;

[0031] According to the number of sample points, calling an adapted heuristic strategy to heuristically generate a plurality of control sample points in the basic network;

[0032] Based on the basic network generated with the plurality of control sample points, solving the acyclic traveling salesman problem, and obtaining the acyclic shortest path passing through the plurality of control sample points;

[0033] The bus route is generated based on the acyclic shortest path.

[0034] Preferably, the loop-free shortest path includes a plurality of the control sample points and a plurality of first network nodes;

[0035] The generating of the bus route based on the acyclic shortest path comprises:

[0036] Connecting at least one of the first network nodes to a node in the basic network so that the loop-free shortest path becomes a line with a node sequence;

[0037] The route with the node sequence is stored in a database or a configuration file to obtain the bus route.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The present invention first generates a single bus route based on the acquired basic network, and the generation of the planned number of bus routes can be carried out in parallel, thereby improving the speed and efficiency of route planning. The present invention optimizes and evaluates the generated planned number of bus routes after networking, and evaluates them with the total cost of user travel as the target value, which reflects the overall performance of the line network, and there is no need to evaluate each single line separately, but to evaluate with transfers on the overall network, and through continuous optimization and iteration of the bus route network, the optimal transfer network solution is given without significantly increasing the solution space, which is suitable for solving large-scale urban-level bus networks, transportation networks, etc.

[0040] The present invention also provides a bus transfer network optimization device, comprising:

[0041] The acquisition module is used to obtain the basic network and the number of planned bus routes;

[0042] A route generation module, used for generating bus routes of the planned number of bus routes based on the basic network;

[0043] A line networking module, used for networking the bus lines to form a bus line network;

[0044] A navigation evaluation module, used for performing the shortest path planning for each user based on the bus route network, obtaining the individual cost of each user's travel under the shortest path planning, and obtaining the total cost of all users' travel based on all the individual costs;

[0045] The analysis module is used to take the total cost as the target value, determine whether the target value converges, and if so, obtain the bus route network; if not, use a heuristic algorithm to iteratively optimize the bus route network until the target value converges.

[0046] The present invention also provides an electronic device, comprising a memory storing a computer program and being coupled to a processor, wherein when the computer program is loaded and executed by the processor, the bus transfer network optimization method as described above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is an overall flow chart of the bus transfer network optimization method in an embodiment of the present invention;

[0048] Figure 2 It is a flowchart of a method for optimizing a bus transfer network according to an embodiment of the present invention;

[0049] Figure 3 A schematic diagram of a single-line parallel planning process in an embodiment of the present invention;

[0050] Figure 4 Schematic diagram of single-line networking in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0052] It should be noted that the following abbreviations and key terms mentioned in the present invention have the following meanings:

[0053] Bus routes: refers to bus routes to be generated or optimized;

[0054] Basic network: The basic network is a network with a "node-connection" structure, which is the basis for generating bus routes;

[0055] Bus route network: a set of bus routes that can be connected to each other at nodes for transfer;

[0056] Bus route control sample points: points used to determine routes. Depending on the route planning scenario, they can be either geographic points with longitude and latitude attributes or nodes in the basic network.

[0057] Bus route spline: in the present invention, it specifically refers to the bus route determined by the control sample point fitting;

[0058] OD: Origin & Destination, refers to a one-person trip by public transportation, with a specific departure point and destination.

[0059] See also Figure 1 , 2 As shown, an embodiment of the present invention provides a method for optimizing a bus transfer network, comprising the following steps:

[0060] S1, obtain the basic network and the number of planned bus routes;

[0061] S2, generating bus routes with planned bus route numbers based on the basic network;

[0062] S3, networking the bus routes to form a bus route network;

[0063] S4, based on the bus route network, the shortest path planning is performed for each user, and the individual cost of each user's travel under the shortest path planning is obtained, and the total cost of all users' travel is obtained based on all individual costs;

[0064] S5, taking the total cost as the target value, determines whether the target value converges, if so, obtains the bus route network, if not, uses a heuristic algorithm to iteratively optimize the bus route network until the target value converges.

[0065] This embodiment first generates a single bus line based on the acquired basic network, and the generation of the planned number of bus lines can be performed in parallel, thereby improving the speed and efficiency of line planning. This embodiment optimizes and evaluates the generated planned number of bus lines after networking, and evaluates them with the total cost of user travel as the target value, which reflects the overall performance of the line network. There is no need to evaluate each single line separately, but to evaluate with transfers on the overall network, and through continuous optimization and iteration of the bus line network, the optimal transfer network solution is given without significantly increasing the solution space, which is suitable for solving large-scale urban bus networks, transportation networks, etc.

[0066] Specifically, the basic network can be a road network consisting of road segment connections and road segment nodes, or a network consisting of existing bus routes, consisting of line-to-station connections and station nodes. The basic network can be pre-stored in a database or configuration file.

[0067] The planned number of bus routes is the total number of bus routes R selected according to demand. A reasonable value can be selected and determined between R1 and R2 according to the following formula.

[0068] R∈(R1, R2), where

[0069] In the above formula, R is the number of planned bus routes, which is an integer between R1 and R2. w , is the width of the city under study, C l is the length of the city under study, C s is the maximum and reasonable walking distance for users, R1 represents the number of lines required to lay a horizontal and vertical orthogonal bus line grid for the studied city under the condition of equal spacing, I is the total budget for the planned line investment, and P i is the average cost of a single line, and R2 represents the maximum number of lines that the planned budget can support.

[0070] In this embodiment, when determining the number of bus routes, not only the city boundary range is considered, but also the route budget, cost, etc. are considered. The selected number of bus routes is more reasonable, which is also conducive to the subsequent evaluation of the generated bus network with the total cost as the target value, making it easier to reach convergence, reducing the optimization process and improving planning efficiency.

[0071] According to the present invention, on the basis of the basic network, a single bus route is first generated, i.e., single-line planning, and a preset number of planned bus routes R can be planned in parallel, such as Figure 3 In some implementations, the bus routes of the planned bus routes are generated based on the basic network in S2, including:

[0072] S21, according to the line generation requirement, obtain the number of sample points, illustratively:

[0073] The two bus route sample points are suitable for generating trunk lines that cross the entire urban area. Generally, the starting point and the end point only need to be set. When the basic network is regular, such as the orthogonal pattern of the city, the detour coefficient can be guaranteed to be close to 1.

[0074] The three bus line sample points are suitable for generating trunk lines with tidal passenger flow transport capacity, and generally the maximum detour coefficient does not exceed 1.5.

[0075] A rectangular bus loop can be generated using 5 control sample points: the four vertices of the rectangle, plus the end point that coincides with the first point.

[0076] In other scenarios, the number of sample points can be selected as needed without strict restrictions.

[0077] The number of sample points can be pre-stored in the configuration file according to the line scenario.

[0078] In this embodiment, the number of sample points for generating a route can be flexibly selected according to the optimization scenario of the bus route. The minimum number of sample points can be 2, which can be much smaller than the number of nodes of the final route, such as the number of bus stops or the number of road network nodes.

[0079] S22, calling an adapted heuristic strategy according to the number of sample points, and heuristically generating a plurality of control sample points in the basic network.

[0080] Specifically, different heuristic strategies are called according to the scenario of bus route optimization. New sample points can be generated by initial solution generation strategy or neighborhood solution generation strategy in group algorithms such as genetic algorithm, ant colony algorithm, etc. or individual algorithms such as neighborhood search algorithm, etc. The initial solution generation strategy is suitable for generating new sample points, and the neighborhood solution generation strategy is suitable for selecting adjacent nodes around the node where the current sample point is located as the new location of the sample point.

[0081] Preferably, a plurality of generated control sample points are arranged on the basic network, that is, each control sample point corresponds to a node of the basic network. For the convenience of description, the node corresponding to the control sample point in the basic network is called the basic network node. In this embodiment, the control sample points share the coordinate information and / or label information of the basic network node, that is, in the basic network after the control sample points are generated by inspiration, each control sample point retains the label information and coordinate information of the basic network node corresponding to it. At the same time, for the basic network after inspiration, the label and coordinate information of the control sample point can also be shared with the corresponding basic network node. In this way, the interconnectivity between the networks before and after inspiration is improved, and it is convenient to quickly retrieve the control sample points in the basic networks before and after inspiration.

[0082] S23, based on the basic network after the multiple control sample points are generated, the acyclic traveling salesman problem is solved to obtain the acyclic shortest path passing through the multiple control sample points to form a bus route spline.

[0083] The traveling salesman problem is to find the shortest circuit that passes through only one intermediate point and returns to the starting point given a series of intermediate points and the distances between each intermediate point. The acyclic traveling salesman problem means that there is no need to return to the starting point.

[0084] In this embodiment, an acyclic traveling salesman problem model is established based on the inspired basic network, a preset path planning model is established based on the heuristic algorithm, and the acyclic traveling salesman problem model is solved through the preset path planning model, which simplifies the solution method of the acyclic shortest path and improves planning efficiency.

[0085] S24, generating a bus route based on the loop-free shortest path, wherein the loop-free shortest path includes a plurality of control sample points and a plurality of first network nodes, and at least one first network node is connected to a node in the basic network, so that the loop-free shortest path becomes a route with a node sequence. The route with the node sequence is stored in a database or a configuration file to obtain a bus route.

[0086] In this embodiment, the loop-free shortest path is laid on the network, all the basic network nodes that the bus line spline passes through can be selectively connected or disconnected, all the connected nodes and control sample points are sorted and processed to obtain a line with a node sequence, and the node sequence and the loop-free shortest path are stored to form a bus line. Among them, the node in the bus line spline is not connected to the first network node, which means that the node in the bus line spline only passes through without setting a station. The overlapped station can be an ordinary station or transfer station on the bus network, or an ordinary station or transfer station for logistics transportation. The overlap logic can be implemented according to the needs of generating different lines.

[0087] In the prior art, when generating a new bus route based on a road network or a line network, it is necessary to provide all the stations that the route passes through, that is, a long sequence of stations is required to represent a route. In the case of a single line in a general city, this type of bus route defined by all the stations along the route involves dozens of stations, and the number of stations is k. The solution space is N k However, if there are m bus lines in the network, the solution space can reach N mk , it is difficult for the city to reach convergence within the effective time, and there is a high probability that the station sequence will be unreasonable, resulting in a high detour coefficient of the bus line, or even going back and forth in the direction of OD, resulting in excessive costs and reduced operation or transportation efficiency.

[0088] In this embodiment, new control sample points are generated based on the number of sample points, and the number of sample points is selected according to the line generation requirements, that is, the number of sample points is related to the shape of the line generation. By generating control sample points based on the number of sample points, the generation shape of the line can be better controlled. Furthermore, a loop-free shortest path is generated based on the control sample points through an inspiration strategy, and is connected with the basic network through overlapping to form a loop-free shortest path with a node sequence. Since the loop-free shortest path has the dual properties of being loop-free and shortest, the loop-free shortest path is used as a bus route to avoid excessive detours in public transportation routes.

[0089] According to the present invention, after the single-line planning is completed, R single-line bus lines are obtained, and then single-line networking is performed, and the generated R single-line bus lines are networked and merged to form a bus network, such as Figure 4 In some implementations, networking the generated bus lines in S3 to form a bus line network includes: laying the R bus lines generated in S2 on the basic network, merging the overlapping nodes on the basic network of the R bus lines into one node, and the merging criterion is: the basic nodes where these line nodes are located are the same, which means that these lines are connected at the node and can be transferred.

[0090] Specifically, assuming that there are multiple first nodes in a single bus line, the first node in the single bus line that coincides with the basic node in the basic network is the first coincident node. The first coincident node of each bus line is obtained, and then the first coincident nodes in the R bus lines are merged into one node. Since the first coincident node in each bus line is the same as the basic node, the basic nodes where the first coincident nodes in the R bus lines are located are the same, which can be used as line connection transfer nodes, such as Figure 4 As shown. Thus, the R bus routes are networked to form a new bus route, which is referred to as network G below.

[0091] In some implementation methods, in S4, for the generated bus route network G, the shortest path planning is performed on all ODs on the network G, and navigation evaluation is performed.

[0092] For example, the specific process of performing the shortest path planning for all ODs on the network G and performing navigation evaluation is as follows:

[0093] Perform the following parallel operations on all ODs:

[0094] For an ODi, use the shortest path algorithm such as Dijkstra and A-Star to find its shortest path on the network G, and the path length cost is Li.

[0095] If Ri cannot continue the original route after passing a connected node, that is, a transfer occurs, the transfer cost will be increased. The specific amount is a constant I, which represents the cost consumption caused by the transfer.

[0096] Then the single cost of ODi on network G is:

[0097] C i =L i + kI,

[0098] Among them, i represents the i-th user, C i represents the individual cost of the i-th user's travel under the shortest path planning, L i is the path length cost of the ith user traveling on the shortest path obtained by the shortest path planning, is the cost incurred by the ith user transferring once on the bus network, and k is the number of transfers made by the ith user on the bus network.

[0099] The total cost of all user trips is obtained by adding up the costs of all ODs, which is the total cost of network G serving these trip ODs. The total cost is determined using the following formula:

[0100] C j =∑C i ,

[0101] Where, j represents the jth bus route network, C j represents the total cost of all users traveling on the jth bus route network. It should be understood that C j is the cost performance measure of network G, that is, the target value, the lower the value, the better.

[0102] In some implementation modes, in S5, the method for determining whether the target value has converged is:

[0103] Get the ratio of the target value drop in the last 1 / 3 of the iterations to the total drop;

[0104] Compare whether the ratio is less than a set convergence speed threshold, wherein the set convergence speed threshold can be selected as 0.05 or 0.1 as required;

[0105] If so, it is determined that the target value converges;

[0106] If not, the bus route network is optimized iteratively until the target value converges.

[0107] Specifically, the optimization iteration strategy can use heuristic algorithms to advance to the next iteration. In this embodiment, both group heuristic algorithms such as genetic algorithms and ant colony algorithms and individual heuristic algorithms such as adaptive large neighborhood search can be used.

[0108] In this step, part of the solution of the previous generation (such as part of the R lines) needs to be cancelled, mutated, crossed, or regenerated, while the other part can be retained, depending on the iterative algorithm. Taking the genetic algorithm as an example, the optimization iteration process is described in detail.

[0109] For genetic algorithms, the solution of a network can be obtained by List <List <controlpoint>>Type representation, where ControlPoint is the control point, List <controlpoint>A route is represented by a set of control points, and a solution is a bus network composed of multiple routes. In the evaluation, the bus network will obtain a cost value C. For a genetic population, each individual represents an optional bus network, and the entire population is List <List<List <controlpoint>>>, each individual cost value is denoted as Cj.

[0110] For example, the cancellation strategy can be a combination of target value comparison and random strategy, solving k(List <List <controlpoint>>) is cancelled with a probability of:

[0111]

[0112] That is: the greater the cost of a single network, the greater the probability that it will be cancelled.

[0113] The retention strategy can be obtained according to the following formula, that is, the probability of retention is:

[0114] P k保留 =1-P k取消 ;

[0115] For example, the mutation strategy can select an individual k to be mutated, and then perform the mutation operation according to the individual mutation operator. <List <controlpoint>>), represents a complete candidate transportation network, and the process of mutation operation is as follows:

[0116] Step E2.1: Select the variant line;

[0117] A route is randomly selected from the individual solution k by taking a random number

[0118] r r = Random(0,n-1);

[0119] Where n is the number of individuals, Random(0,n-1) is a random integer from 0 to n-1, that is, an integer is randomly selected from 0 to n-1 as the subscript of the selected mutation line, then the rth r The line is the line to be mutated.

[0120] Step E2.2: Select the variation control points;

[0121] For a route selected in step E2.1 (in the form of List <controlpoint>), select one of its control points, labeled:

[0122] r c = Random(0, c-1);

[0123] Where c is the number of control points, then the rth c The control points are the ones that need to be varied.

[0124] Step E2.3: Shift the control points;

[0125] Step E2.1 and step E2.2 can be cycled and repeated until the selection is made. r The rth line c The target position is any random node on the network that does not meet the linear strategy of the scene (such as straight line type, right-angle line type).

[0126] The above steps E2.1, E2.2, and E2.3 can be cyclically and repeatedly performed to select multiple routes and control points until the number of routes and control points subjected to mutation operations meets the mutation rate (preferably between 5% and 30%).

[0127] Exemplarily, for two individuals in a group, partial lines are exchanged, and the specific steps are as follows:

[0128] Step E3.1: Randomly select individual A and individual B;

[0129] The methods for determining the labels are:

[0130] r i = Random(0, I-1);

[0131] Where I is the number of individuals. If the same number is obtained twice, repeat the process until the numbers are different, then A and B are two different individuals.

[0132] Step E3.2 randomly selects the line to be exchanged among the individuals;

[0133] The specific steps are as follows: for each individual (A or B), the probability of each line is calculated, and the line that meets the following conditions is included in the list of individual lines to be exchanged

[0134] r r = RandomF(0,1) > r cf ;

[0135] where r cf is the crossover probability, which is generally selected from 0.1 to 0.25; RandomF(0,1) is a continuous random floating point number between 0 and 1. So far, two lists of lines to be exchanged, ListA and ListB, are formed. ListA is removed from individual A and added to ListB to form A'; ListB is removed from B and added to ListA to form B'. So far, two new individuals are formed after crossover.

[0136] Another embodiment of the present invention provides a bus transfer network optimization device, including:

[0137] The acquisition module is used to obtain the basic network and the number of planned bus routes;

[0138] A route generation module, for generating bus routes with a planned number of bus routes based on the basic network;

[0139] A line networking module is used to network bus lines to form a bus line network;

[0140] The navigation evaluation module is used to plan the shortest path for each user based on the bus route network, obtain the individual cost of each user's travel under the shortest path planning, and obtain the total cost of all users' travel based on all individual costs;

[0141] The analysis module is used to take the total cost as the target value and determine whether the target value converges. If so, the bus line network is obtained. If not, the heuristic algorithm is used to optimize the bus line network iteratively until the target value converges.

[0142] Another embodiment of the present invention further provides an electronic device, including a memory storing a computer program and being coupled to a processor, and when the computer program is loaded and executed by the processor, the above-mentioned bus transfer network optimization method is implemented.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.< / controlpoint> < / controlpoint> < / controlpoint> < / controlpoint> < / controlpoint> < / controlpoint>

Claims

1. A method for optimizing a bus transfer network, characterized in that: include: Get the basic network and the number of planned bus routes; Generate the planned number of bus routes based on the basic network; Networking the bus routes to form a bus route network; Based on the bus route network, the shortest path planning is performed for each user respectively, and the individual cost of each user's travel under the shortest path planning is obtained, and the total cost of all users' travel is obtained based on all the individual costs; Taking the total cost as the target value, judging whether the target value has converged includes: obtaining the ratio of the decrease value of the target value obtained in the last 1 / 3 stage of iteration to the total decrease value; Comparing whether the ratio is less than a set convergence speed threshold; If so, determining that the target value converges; If not, optimizing the bus route network iteratively until the target value converges; If yes, the bus route network is obtained; if no, a heuristic algorithm is used to iteratively optimize the bus route network until the target value converges.

2. The method for optimizing a bus transfer network according to claim 1, characterized in that: The method of obtaining the number of planned bus routes includes: The following formula is used to determine the number of planned bus routes: R∈(R1,R2), where Where R is the number of planned bus routes, C w is the width of the city under study, C l is the length of the city under study, C s is the maximum and reasonable walking distance for users, R1 represents the number of lines required to lay a horizontal and vertical orthogonal bus line grid for the studied city under the condition of equal spacing, I is the total budget for the planned line investment, P i is the average cost of a single line, and R2 represents the maximum number of lines that the planned budget can support.

3. The method for optimizing a bus transfer network according to claim 1, characterized in that: The basic network is a network with a node-connection structure, the basic network includes a basic node, and the bus line includes a first node; The networking of the bus routes to form a bus route network comprises: Obtaining a node in the first nodes that coincides with the base node as a first coincident node; The first overlapping nodes in the R bus routes that overlap with the basic nodes are merged into one node as a route connection and transfer node.

4. The method for optimizing a bus transfer network according to claim 1, characterized in that: The obtaining of the individual cost of each user's travel under the shortest path planning includes: The individual costs are determined using the following formula: C i =L i +kI; Among them, i represents the i-th user, C i represents the individual cost of the ith user's travel under the shortest path planning, L i is the path length cost of the ith user traveling on the shortest road obtained by the shortest road planning, I is the cost of the ith user transferring once on the bus route network, and k is the number of transfers of the ith user on the bus route network.

5. The method for optimizing a bus transfer network according to claim 4, characterized in that: The obtaining of the total cost of all user trips based on all the individual costs includes: The total cost is determined using the following formula: C j =∑C i ; Where, j represents the jth bus route network, C j Represents the total cost of all users traveling on the j-th bus route network.

6. The method for optimizing a bus transfer network according to claim 1, characterized in that: The bus routes for respectively generating the planned number of bus routes based on the basic network include: According to the line generation requirements, obtain the number of sample points; According to the number of sample points, calling an adapted heuristic strategy to heuristically generate a plurality of control sample points in the basic network; Based on the basic network generated with the plurality of control sample points, solving the acyclic traveling salesman problem, and obtaining the acyclic shortest path passing through the plurality of control sample points; The bus route is generated based on the acyclic shortest path.

7. The method for optimizing a bus transfer network according to claim 6, characterized in that: The loop-free shortest path includes a plurality of control sample points and a plurality of first network nodes; The generating of the bus route based on the acyclic shortest path comprises: Connecting at least one of the first network nodes to a node in the basic network so that the loop-free shortest path becomes a line with a node sequence; The route with the node sequence is stored in a database or a configuration file to obtain the bus route.

8. A bus transfer network optimization device, characterized in that: include: The acquisition module is used to obtain the basic network and the number of planned bus routes; A route generation module, used for generating bus routes of the planned number of bus routes based on the basic network; A line networking module, used for networking the bus lines to form a bus line network; A navigation evaluation module, used for performing the shortest path planning for each user based on the bus route network, obtaining the individual cost of each user's travel under the shortest path planning, and obtaining the total cost of all users' travel based on all the individual costs; The analysis module is used to take the total cost as the target value and determine whether the target value has converged, including: obtaining the ratio of the decrease value of the target value obtained in the iteration of the last 1 / 3 stage to the total decrease value; comparing whether the ratio is less than a set convergence speed threshold; if so, determining that the target value has converged; if not, optimizing and iterating the bus route network until the target value has converged; if so, obtaining the bus route network; if not, using a heuristic algorithm to optimize and iterate the bus route network until the target value has converged.

9. An electronic device, characterized in that: The method comprises a memory storing a computer program and being coupleable to a processor, and when the computer program is loaded and executed by the processor, the method for optimizing a bus transfer network as claimed in any one of claims 1 to 7 is implemented.

Images