A rail transit feeder bus path and scheduling optimization method based on shared bicycle travel influence

Abstract

The application discloses an optimization method for rail transit feeder bus path and scheduling based on shared bicycle travel influence, belongs to the field of public traffic planning management, and comprises the following steps: firstly, selecting a to-be-optimized rail transit station, delimiting an optimization region, extracting information of all bus stations and shared bicycle parking points in the research region, and corresponding the shared bicycle parking points and bus stations in the same community travel selection with the shortest distance; then, acquiring the current operation mode of the selected rail transit station and traffic data of passengers from the residence to the nearest bus station and the nearest shared bicycle parking point, constructing a target function model with the minimum total cost of the feeder bus and the shared bicycle connection, determining a Logit model of the selection probability, determining the constraint condition, and solving the model to obtain a planning path; and through the optimization of the feeder bus path and the scheduling time, the overall cost of the passengers riding the bus is reduced, and the selection probability of the passengers to the feeder bus is increased.

Description

Technical Field

[0001] This invention relates to the technical field of public transportation planning and management, specifically to an optimization method for rail transit connection bus routes and scheduling based on the impact of shared bicycle travel. Background Technology

[0002] As a crucial component of the urban transportation system, rail transit is favored by the public due to its advantages of speed, punctuality, safety, and independence from road traffic conditions. However, as the "main artery" of urban transportation, rail transit has a long construction period and high costs. Economic and environmental factors are considered during the design and planning stages, resulting in some rail transit stations being located beside open roads, some distance from residential areas, preventing passengers from getting home in one stop. In this context, rail transit shuttle buses, acting as the "capillaries" of urban transportation, provide "last-mile" transportation, penetrating deep into communities and reaching areas not covered by rail transit. Shuttle buses connect residential areas, office buildings, and commercial areas with rail transit stations, facilitating citizens' travel while expanding the service range of rail transit. Furthermore, the rapid development of shared bicycles and shared electric bicycles in recent years has gradually become an important means of connecting citizens' transportation needs.

[0003] Shared bicycles are characterized by low barriers to entry, ease of use, and low cost, making them irreplaceable for short-distance travel. As a minimally carbon-intensive and environmentally friendly mode of transportation, major cities are increasing their investment in shared bicycles, which are now ubiquitous in major plain cities such as Beijing, Shanghai, Shenzhen, Chengdu, and Kunming. Shared bicycles have become the preferred mode of transportation for commuters connecting to rail transit, which has consequently had a significant impact on the passenger flow of urban public transportation.

[0004] Both shared bicycles and connecting buses are ideal modes of transportation connecting to rail transit, and thus, they are inherently in competition. In recent years, the number of shared bicycle trips has increased significantly, while the passenger volume of connecting buses has declined sharply. This loss of passenger volume translates to decreased revenue for buses, leading bus companies to reduce the frequency of connecting bus services and increase intervals to balance expenses and reduce operating costs. However, for the elderly, some citizens who cannot ride bicycles, passengers carrying large luggage, and during inclement weather, connecting buses remain the best mode of transportation and must continue to provide adequate services to these groups. Therefore, considering the impact of shared bicycles, rationally planning the routes and scheduling of rail transit connecting buses can increase passenger volume and is of great practical significance in effectively addressing the "last mile" transportation problem for urban residents. Summary of the Invention

[0005] The purpose of this invention is to optimize the routes and departure schedules of rail transit connecting buses under the influence of shared bicycles, increase the passenger flow share of connecting buses, achieve a balanced operation of connecting buses and shared bicycles, and provide better travel services for citizens.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows:

[0007] A method for optimizing rail transit connection bus routes and scheduling based on the impact of shared bicycle travel includes the following steps:

[0008] Step 1: Select the rail transit stations to be optimized and delineate the optimization area;

[0009] Step 2: Extract information on all bus stops and shared bicycle stops within the study area, obtain the current operating mode of the selected rail transit stations, obtain traffic behavior data of passengers traveling from their residences to the nearest bus stop and the nearest shared bicycle stop, investigate passenger flow at each station, and obtain traffic data in connection behavior;

[0010] Step 3: Construct an objective function model that minimizes the total passenger and operating costs of connecting buses and shared bicycles;

[0011] Step 4: Construct a Logit model of the probability of travel mode selection;

[0012] Step 5: Construct constraints to reflect reality and achieve reasonable solution results;

[0013] Step 6: Solve the model to obtain the optimization results of connecting bus routes and scheduling under the influence of shared bicycle travel.

[0014] Furthermore, the basic assumptions of this model are as follows: each connecting bus route travels at the obtained average speed, without considering delays caused by traffic lights at intersections, traffic congestion, etc., i.e., arriving on time; the arrival time of shared bicycle riders at the rail transit station in this model follows a uniform distribution; passengers of connecting buses in this model arrive at the station according to the timetable, without considering the waiting time at the bus stop; the passenger flow at each station in this model refers to the connecting passenger flow to the rail transit station, without considering the passenger flow between stations; passengers in this model take the next nearest train after arriving at the rail transit station, without considering the situation of not being able to board; the basic parameters in this model are all obtained from field surveys or real-world experience.

[0015] Furthermore, the information on bus stops and shared bicycle parking spots in step two includes their locations, the distance between bus stops, the distance between shared bicycle parking spots and rail transit stations, and the distance between bus stops and shared bicycle parking spots and the nearest residential area.

[0016] Furthermore, the operational methods in step two include, but are not limited to, the arrival times of trains in both directions during the morning rush hour.

[0017] Furthermore, the traffic behavior data in step two includes passengers' walking distance, walking speed, and average transfer time from bus stops and stops to rail transit stations.

[0018] Furthermore, in step two, station passenger flow refers to the passenger flow of people taking public transportation and shared bicycles to transfer to rail transit during the morning rush hour on weekdays.

[0019] Furthermore, the traffic data for the connecting behavior includes the average speed of connecting buses, stop time, average cycling speed of bicycles, bicycle supply and capacity at each stop, passenger time cost for public transportation, passenger time cost for shared bicycles, public transportation fare, and unit fare parameters for shared bicycles.

[0020] Furthermore, in step two, the latitude and longitude of each bus stop, shared bicycle parking spot, and rail transit station are extracted using a map platform. Then, the distance between bus stops and the distance between parking spots and rail transit stations are calculated based on the latitude and longitude. The map platform mentioned is an open map platform, including but not limited to Baidu Maps, Gaode Maps, etc.

[0021] Furthermore, in step two, the morning peak was studied using the two hours with the highest travel volume in the morning.

[0022] Further, in step two, the nearest bus stop to each residential area is found, and the residential areas are mapped to the bus stops. Then, the average walking distance from each bus stop to the corresponding residential area is calculated. Similarly, find the nearest stop to each residential area and calculate the average walking distance between each residential area and its corresponding stop. The system will also link the shared bicycle parking spots chosen by residents of the same community with bus stops.

[0023] Furthermore, in step two, passenger flow for connecting buses is obtained through bus IC card swiping and QR code scanning data, and passenger flow for shared bicycles is obtained by identifying riding data through mobile phone signaling and shared bicycle data. The two are then summed to obtain the total regional rail transit connecting passenger flow.

[0024] Furthermore, the specific steps of step three are as follows:

[0025] (i) The walking time to the station for passengers can be obtained based on the parameter data obtained in step two. , Distance between stations Shared bicycle connection riding distance Number of stops along the route of the connecting bus l and station stop time Train arrival time and stops x bicycle supply And the demand for bicycle rentals at this location. .

[0026] (ii) Based on the distance between stations Connection and shared bicycle riding distance The travel time of the connecting bus can be obtained. and shared bicycle riding time Based on the number of stops along the route of the connecting bus l and station stop time You can get the total stop time. The transfer waiting time between connecting buses and rail transit can be obtained from the train's arrival time, and is expressed as follows: ,in Let be the arrival time of the nth train. This refers to the arrival time of the kth bus in the connecting bus series. This represents the average time it takes for passengers to walk from the moment they alight from the train to the platform level of the rail transit station; the time passengers spend waiting for connecting buses at the bus stop can be expressed as... ,in This indicates the number of shuttle bus departures per hour, which can be found through the shuttle bus departure times. Calculations show that the transfer waiting time for passengers using shared bicycles can be expressed as half the time interval between two adjacent rail transit train services, i.e. .

[0027] (III) Based on the parking points x single-vehicle supply obtained in step two And the demand for bicycle rentals at this location. Define a penalty cost for borrowing and repaying, denoted as ,in It is the penalty coefficient for borrowing and repaying.

[0028] (iv) In actual use of shared bicycles, especially during the morning rush hour, there is often a large influx of commuters returning bicycles to connect with rail transit, leading to insufficient parking capacity and a "difficulty in returning bicycles" situation. Therefore, the scenarios of returning bicycles are divided into three categories:

[0029]

[0030] in These represent the proportion of the total morning rush hour during which the parking spaces at the return point are available in the three categories mentioned above. Additionally, a condition where there are ≤5 remaining parking spaces is defined as a situation where there are relatively few remaining parking spaces.

[0031] The time cost incurred due to insufficient parking space capacity is also divided into three categories. When there are many remaining parking spaces, the rider's return time is... When there are few parking spaces remaining, the cyclist's return time is... If no parking spaces remain, assuming the rider chooses the nearest parking spot to their current location, the return time will be [time period missing]. , This indicates the distance of the cyclist from the nearest rest stop. This represents the average riding speed of passengers and the average return time of shared bicycle riders. .

[0032] (v) The passenger cost of connecting buses is:

[0033] (1)

[0034] This includes passenger walking time to the station, vehicle travel time, stop time, waiting time, transfer time, and ticket cost.

[0035] In equation (1): S represents the set of alternative stops for connecting buses. This represents the cost per unit of time for passengers. This indicates the time it takes for a passenger to walk to the nearest bus stop, i. This represents the shortest travel distance between stop i and stop j for the connecting bus l. Indicates the bus fare. This represents the passenger flow at bus stop i. 0-1 decision variables:

[0036]

[0037] (vi) The passenger cost of shared bicycles is:

[0038] (2)

[0039] This includes passenger walking time to the stop, cycling time, transfer waiting time, capacity penalty costs, borrowing and repayment imbalance penalty costs, and ticket costs.

[0040] In equation (2): This represents the set of alternative parking spots for shared bicycles. The time it takes for a passenger to walk to the nearest shared bicycle stop (x) is represented by the following value: This indicates the imbalance penalty coefficient. Indicates the price of shared bicycles. This represents the passenger flow at shared bicycle parking point x.

[0041] (vii) The operating cost of the connecting bus is:

[0042] (3)

[0043] In formula (3): This indicates the unit cost of the connecting bus operator. This indicates the number of buses departing per hour for bus route l. This represents the shortest travel distance between station i and station j for the connecting bus l.

[0044] (viii) The operating cost of shared bicycles is .

[0045] (ix) The objective function model constructed in this invention is as follows:

[0046] (4)

[0047] Furthermore, in step four, for this type of multi-modal transportation allocation problem, a Logit model needs to be constructed to calculate the probability of transportation mode selection, expressed as:

[0048] (5)

[0049] In equation (5): i = 1, 2, Let represent the probability of passengers choosing between connecting buses and shared bicycles, respectively. Let θ represent the passenger costs for these two connection methods, and θ represent the utility coefficient.

[0050] The calculation results of this model are jointly determined by the passenger costs of the two modes of transportation. Using the model in step three, the passenger travel costs of the two modes of transportation at each station in the current study area are calculated. Calculations are performed, and then the probability of passenger choice for each mode of transportation at each station is calculated using the Logit model. Stations where the probability of choosing connecting buses is much lower than that of choosing shared bicycles are considered to be stations with minimal connecting bus passenger flow, where passengers mainly rely on shared bicycles for transportation; these stations will not be considered for subsequent connecting bus route optimization. The objective function of the lower-level model is established with the goal of maximizing the sum of public transportation travel choice probabilities.

[0051] Furthermore, the constraints in step five are as follows:

[0052] ①

[0053]

[0054]

[0055] ②

[0056] ③

[0057] ④

[0058] ⑤ (6)

[0059] ⑥

[0060] ⑦

[0061] ⑧

[0062] 9

[0063]

[0064] In equation (6): constraint ① is a capacity constraint, where Let L be the vehicle capacity of line l. Let be the average departure frequency of line l; constraint ② is the departure frequency constraint. The interval between two adjacent train services is calculated. , , Let $\mathbf{i}$ and $\mathbf{j}$ represent the minimum and maximum allowed departure frequencies, respectively; constraint ③ ensures that only one bus route serves path $i$ and $j$; constraint ④ ensures that the number of routes entering and departing from each station is the same; constraint ⑤ is a typical Miller-Tucker-Zemlin (MTZ) constraint, the purpose of which is to eliminate possible sub-loops in the TSP problem; constraint ⑥ is the connecting bus route length constraint. , These represent the minimum and maximum lengths of the connecting bus routes, respectively; constraint ⑦ is the constraint on the number of connecting bus stops. , These represent the minimum and maximum number of stops for connecting buses, respectively; constraint ⑧ is the passenger transfer waiting time constraint, ensuring that the passenger transfer waiting time for connecting buses does not exceed the interval between two adjacent trains; constraint ⑨ is the shared bicycle parking station capacity limit, ensuring that the number of passengers renting bicycles at parking station x does not exceed the number of bicycles supplied at the parking station. The number of vehicles parked at the designated stop; constraints As a constraint on decision variables, when path i,j is selected, =1, otherwise 0.

[0065] Furthermore, in Step 6, the genetic algorithm is used to solve the model, and the decision variables are encoded. The real number string encoding method is adopted for the feeder bus route selection.

[0066] Furthermore, the specific steps of Step 6 are as follows:

[0067] (1) Coordinate point import: Import the longitude and latitude information of the bus stops and bike docking points obtained into the Matlab software.

[0068] (2) Basic parameter setting: Input the data of all basic parameters in the objective function model, and set the population size sizepop, the maximum number of iterations maxgen, the generation gap gap, and the crossover probability , and the mutation probability .

[0069] (3) Initial population initialization: Number each bus stop, number the rail transit feeder station as "0", and number the remaining bus stops from 1 to n. Then randomly generate a random array from 1 to n. According to the site scale in the research area and the reasonable number limit of stops on the feeder line, automatically generate the minimum number of feeder bus lines and the maximum number of lines in the area. For different numbers of feeder bus lines , respectively generate random numbers to divide the stop array into several groups, and add "0" before the first number in each group to represent the feeder bus driving route.

[0070] (4) Constraint check: After obtaining the initial solution of the corresponding population size, input the constraint conditions established in Step 5. The solution that satisfies the constraint conditions is the feasible initial solution.

[0071] (5) Start iteration: Input the initial iteration number gen = 0. When gen < maxgen is satisfied, perform iteration. When it is not satisfied, the maximum number of iterations is reached, and the iteration can be ended.

[0072] (6) Calculate fitness: Calculate the objective function value for the initial solution, compare the objective function values under different numbers of feeder bus lines, and retain the optimal solution. The fitness of the solution is the reciprocal of the objective function value.

[0073] (7) Selection: Adopt the roulette wheel method for population selection, and leave the optimal individuals according to the set generation gap to form the offspring population.

[0074] (8) Crossover: Adopt simulated binary crossover for crossover, and update the population after the crossover operation.

[0075] (ix) Mutation: Polynomial mutation is used for the real number encoded part to update the current population.

[0076] (x) After completing one iteration, the iteration number gen+1 is incremented, and the process returns to (v) for the next iteration.

[0077] (xi) After reaching the maximum number of iterations, the connecting bus routes that traverse all current bus stops are obtained. The objective function is to maximize the sum of bus selection probabilities, and the station selection is the decision variable. The result is obtained again through the genetic algorithm.

[0078] (xii) Population initialization: Station selection uses 0-1 binary encoding, where "1" indicates stopping at the station and "0" indicates not stopping. Based on the results of the first genetic algorithm, chromosomes of the same length are initialized.

[0079] (xiii) Constraint check: The number of bus stops after the station selection should meet the constraints, that is, the number of "1"s in the chromosome should be within the constraint range.

[0080] (XIV) Start iteration: The iteration process and steps are the same as the first genetic algorithm.

[0081] (XV) Calculate the objective function and fitness: The objective is to select the stop with the sum of the probabilities of bus selection as the goal, and the fitness is the value of the objective function.

[0082] (xvi) For binary encoding, choose multi-point crossover for the crossover operator and multi-point mutation for the mutation operator.

[0083] (xvii) After reaching the maximum number of iterations, the optimal route of the connecting bus after the stop selection and its departure timetable are obtained.

[0084] Furthermore, the described apparatus includes an acquisition module. The acquisition module includes a site acquisition module, a passenger flow acquisition module, and a parameter acquisition module.

[0085] Furthermore, the acquisition module is used to acquire information on all bus stops and shared bicycle stops within the study area mentioned in step two above, passenger flow information for connecting rail transit during the study period, and basic parameters of passenger travel.

[0086] Furthermore, the site acquisition module is used to acquire information on all bus stops and shared bicycle stops within the study area. This information includes the location of bus stops and shared bicycle stops, the distance between bus stops, the distance of shared bicycle stops from rail transit stations, and the distance of bus stops and shared bicycle stops from the nearest residential area.

[0087] Furthermore, the passenger flow acquisition module is used to identify and acquire passenger flow data for each station using connecting buses and each stop using shared bicycles to transfer to rail transit during the study period.

[0088] Furthermore, the parameter acquisition module is used to obtain basic parameter information such as the average speed of connecting buses, shared bicycles, and walking; the fares of connecting buses and shared bicycles, transfer time, walking distance to the station, passenger unit travel cost, unit operating cost; and the capacity and supply of shared bicycle parking spots.

[0089] Furthermore, a calculation module is provided to import the information and parameters obtained from the acquisition module mentioned above into the model established in steps three through six, and to perform calculation and iterative operations through a genetic algorithm until the iteration ends.

[0090] Furthermore, an output module is provided to determine the optimal route and final departure timetable of the connecting buses obtained through iterative calculation in the calculation module, and output the route map and timetable in an intuitive and visible manner.

[0091] Furthermore, the device also includes a memory, a processor, and a computer code program capable of executing the aforementioned mathematical model and genetic algorithm.

[0092] Furthermore, the memory is used to store the aforementioned computer program. The processor is used to read the computer program from the memory and execute the contents described in steps three through six.

[0093] Furthermore, a computer program is stored therein, and after the program is read by the processor, the contents described in steps three through six are executed.

[0094] Compared with existing technologies, the beneficial effects of this invention are as follows: 1. It proposes a joint optimization method for connecting bus routes and scheduling considering the impact of shared bicycles. By optimizing the routes and scheduling times of connecting buses, it reduces the overall cost for passengers to take public transportation, increases the probability of passengers choosing connecting buses, thereby increasing passenger flow and making the use of connecting buses and shared bicycles tend to be balanced; 2. When analyzing the cost of using shared bicycles, this invention considers the real-world problems of "difficulty in borrowing bikes" caused by untimely vehicle dispatch during peak hours and "difficulty in returning bikes" caused by insufficient parking lot capacity, which has practical significance; 3. Joint optimization of connecting bus routes and scheduling can effectively reduce passengers' waiting time and on-board time, reducing passengers' time costs. It also provides bus companies with optimization ideas on how to leverage the advantages of public transportation, change the vicious cycle in competition with shared bicycles, and provide citizens with better travel services. Attached Figure Description

[0095] Figure 1This is a schematic diagram of the method for optimizing the route and scheduling of rail transit connecting buses based on the impact of shared bicycle travel in this invention;

[0096] Figure 2 The flowchart of the genetic algorithm for solving the model in this invention is shown below;

[0097] Figure 3 This is a schematic diagram of the optimal route for connecting buses optimized by a genetic algorithm according to an embodiment of the present invention;

[0098] Figure 4 This is a departure scheduling timetable optimized by a genetic algorithm according to an embodiment of the present invention;

[0099] Figure 5 This is a schematic diagram of the module structure according to an embodiment of the present invention;

[0100] Figure 6 This is a schematic diagram of the device structure according to an embodiment of the present invention. Detailed Implementation

[0101] The process of the present invention will be further described below with reference to the accompanying drawings. The following embodiments are intended to describe the present invention more completely and in detail, and are also within the scope of protection of the present invention.

[0102] Please see Figure 1-2 This invention relates to a method for optimizing the route and scheduling of rail transit connecting buses based on the impact of shared bicycle travel, comprising the following steps:

[0103] Step 1: Select the rail transit stations to be optimized and delineate the optimization area;

[0104] Step 2: Extract information on all bus stops and shared bicycle stops within the study area, obtain the current operating mode of the selected rail transit stations, obtain traffic behavior data of passengers traveling from their residences to the nearest bus stop and the nearest shared bicycle stop, investigate passenger flow at each station, and obtain traffic data in connection behavior;

[0105] Step 3: Construct an objective function model that minimizes the total passenger and operating costs of connecting buses and shared bicycles;

[0106] Step 4: Construct a Logit model of the probability of travel mode selection;

[0107] Step 5: Construct constraints to reflect reality and achieve reasonable solution results;

[0108] Step 6: Solve the model to obtain the optimization results of connecting bus routes and scheduling under the influence of shared bicycle travel.

[0109] To make it easier to understand, step two is further detailed. The specific steps are as follows:

[0110] (I) Extract information on all bus stops and shared bicycle parking spots within the study area. This information includes the locations of bus stops and shared bicycle parking spots, the distance between bus stops, and the distance of shared bicycle parking spots from rail transit stations. Find the nearest bus stop to each residential area, associate the residential areas with the bus stops, and then calculate the average walking distance from each bus stop to its corresponding residential area. Similarly, find the nearest stop to each residential area and calculate the average walking distance between each residential area and its corresponding stop. This process involves mapping shared bicycle parking spots chosen for travel within the same residential community to bus stops. The steps include: 1) using a map platform to extract the latitude and longitude information of all connecting bus stops and shared bicycle parking spots within the area; 2) calculating the distance between each bus stop based on the latitude and longitude information. And the distance between shared bicycle parking point x and rail transit transfer parking point y. .

[0111] (II) Basic information on the connection process in the survey example. This step includes the following: Content 1, passenger flow of passengers transferring from connecting buses to rail transit at each bus stop in the current survey area; Content 2, passenger flow of passengers transferring from shared bicycles to rail transit in the current survey area; Content 3, survey of rail transit train arrival times.

[0112] Content 1 uses bus IC card swiping and QR code scanning data to obtain passenger numbers at each station. Since the connecting bus routes are relatively short and mainly handle transfer passenger flow, the number of passengers boarding can be assumed to represent the passenger flow for transferring to rail transit. Content 2 uses mobile phone signaling data to obtain passenger flow for returning shared bicycles at rail transit station entrances and exits after borrowing them. Content 3 uses the official rail transit platform to obtain train departure intervals and calculates the train arrival timetable for the research stations.

[0113] (III) Investigate basic physical parameters such as bus speed, bicycle speed, passenger walking speed, walking distance to the station, transfer time, bus stop time, shared bicycle supply, and station capacity. In this step, the speed data is the mainstream-accepted average speed, obtainable from relevant literature and reports. The corresponding information for walking distances between residential areas and shared bicycle stops and bus stops was obtained through a map platform. All other information was obtained through on-site investigation.

[0114] (iv) Based on the average income level of local residents and relevant literature research and analysis, the unit time cost of passenger travel and the unit operating cost of operators are analyzed.

[0115] Step 3 includes the following: Content 1, the inter-station distances obtained from step (i) of step 2. Connection and shared bicycle riding distance The travel time of the connecting bus can be obtained. and shared bicycle riding time .

[0116] Content 2: Based on the train arrival time obtained from step (ii) of step two, the transfer waiting time between connecting buses and rail transit can be obtained, expressed as follows: ,in Let be the arrival time of the nth train. The arrival time of the k-th shuttle bus is one of the decision variables in this model. This represents the average time it takes for passengers to walk from the moment they alight from the train to the platform level of the rail transit station; the time passengers spend waiting for connecting buses at the bus stop can be expressed as... ,in This indicates the number of shuttle bus departures per hour, which can be found through the shuttle bus departure times. Calculations show that the transfer waiting time for passengers using shared bicycles can be expressed as half the time interval between two adjacent rail transit train services, i.e. .

[0117] Content 3: Based on the parameter data obtained in step (iii) of step two, the walking time to the station for passengers can be obtained. .

[0118] Content 4, based on the number of stops along the route of the connecting bus l The station dwell time obtained in step (iii) of step two. You can get the total stop time. .

[0119] Content 5: Based on the parking points x single-vehicle supply obtained from step (iii) of step two. And the demand for bicycle rentals at this location. A penalty cost for borrowing and repaying a debt is defined, denoted as: ,in This is the penalty coefficient for borrowing and returning bikes. This is the passenger cost caused by the "difficulty in borrowing a bike" phenomenon, which may occur during peak hours when the supply of shared bikes at the parking points is mismatched with the demand for bikes and the bikes are not dispatched in a timely manner.

[0120] Content 6: In actual use of shared bicycles, especially during morning rush hour, there is often a large influx of commuters returning bicycles to connect with rail transit, leading to insufficient parking station capacity and a "difficulty in returning bicycles" phenomenon. Therefore, this invention categorizes bicycle return scenarios into three types:

[0121]

[0122] in These represent the proportion of the total morning rush hour during which the parking spaces at the return point are available in the three situations mentioned above. At the same time, it is defined that when there are ≤5 remaining parking spaces, it means that there are relatively few remaining parking spaces.

[0123] The time cost incurred due to insufficient parking space capacity can be categorized into three scenarios. When there are plenty of remaining spaces, the rider's return time is... When there are few parking spaces remaining, the cyclist's return time is... If no parking spaces remain, assuming the rider chooses the nearest parking spot to their current location, the return time will be [time period missing]. . This indicates the distance of the rider from the nearest rest stop. Average return time for shared bicycle riders. .

[0124] Content 7, the passenger cost of connecting buses can be expressed as:

[0125] (1)

[0126] This includes passenger walking time to the station, vehicle travel time, stop time, waiting time, transfer time, and ticket cost.

[0127] in This represents the cost per unit of time for passengers. This indicates the frequency of shuttle bus departures. This refers to the bus fare. In this example, the connecting bus fare is a flat rate, payable upon boarding, and is independent of the travel distance. This represents the passenger flow at bus stop i. 0-1 decision variables:

[0128]

[0129] Content 8: The passenger cost of shared bicycles can be expressed as:

[0130] (2)

[0131] This includes passenger walking time to the stop, cycling time, transfer waiting time, capacity penalty costs, borrowing and repayment imbalance penalty costs, and ticket costs.

[0132] in This indicates the imbalance penalty coefficient. Indicates the price of shared bicycles. This represents the passenger flow at shared bicycle parking point x. Shared bicycle fares include various types of passes such as single-trip tickets, monthly passes, and annual passes, and the fares differ between different operators. Therefore, in this embodiment, the shared bicycle fare is standardized to the average price of a monthly pass from all shared bicycle operators.

[0133] Content 9: The operating cost of the connecting bus can be expressed as follows:

[0134] (3)

[0135] in This indicates the operating cost per kilometer of the shuttle bus service. The number of buses departing per hour for bus route l is represented by the decision variable. Decide.

[0136] Content 10: The operating cost of shared bicycles can be expressed as... .

[0137] Content 11, the objective function model constructed in the embodiment is as follows:

[0138] (4)

[0139] In step four, for this type of multi-modal transportation allocation problem, a Logit model needs to be constructed to calculate the probability of transportation mode selection, expressed as:

[0140] (5)

[0141] Where i = 1, 2, Let represent the probability of passengers choosing between connecting buses and shared bicycles, respectively. Let θ represent the passenger costs for these two connection methods, and θ represent the utility coefficient.

[0142] The calculation results of this model are jointly determined by the passenger costs of the two modes of transportation. Using the models in steps 3 (sections 7 and 8), the passenger travel costs of the two modes of transportation at various stations in the current study area are calculated. The calculations were performed, and then the probability of passengers choosing the two travel modes at each station was calculated based on the Logit model.

[0143] The probability of choice not only represents the passenger flow choosing a particular mode of transportation, but also the degree of dependence of residents in a community on that mode of transportation. For example, if a community has a higher probability of choosing public transportation, it means that the passenger flow choosing public transportation is greater than that choosing shared bicycles, and the cost of public transportation is lower. In this case, even if the total passenger flow of this community is less than that of other communities, its dependence on public transportation is higher, and the bus stops corresponding to this community should be given more consideration in route design. For example, if a community is close to a rail transit station and has a lower probability of choosing public transportation, it means that the passenger flow choosing public transportation is smaller, and the residents of this community have a lower dependence on public transportation. In route design, it is possible to consider not stopping at this station to reduce the travel costs for passengers at other stations. Therefore, the objective function of the lower-level model is established with the goal of maximizing the sum of public transportation choice probabilities. This not only considers the passenger flow of connecting public transportation and increases the operator's revenue, but also takes into account stations with smaller passenger flow but higher dependence on public transportation, which is more realistic than simply aiming to maximize passenger flow.

[0144] Based on the established model, constraints are constructed to ensure that the solution meets the actual requirements. The constraints in step five are as follows:

[0145] ①

[0146]

[0147]

[0148] ②

[0149] ③

[0150] ④

[0151] ⑤ (6)

[0152] ⑥

[0153] ⑦

[0154] ⑧

[0155] 9

[0156]

[0157] In equation (6): constraint ① is a capacity constraint, where Let L be the vehicle capacity of line l. Let be the average departure frequency of line l; constraint ② is the departure frequency constraint. The interval between two adjacent train services is calculated. , , Let $\mathbf{i}$ and $\mathbf{j}$ represent the minimum and maximum allowed departure frequencies, respectively; constraint ③ ensures that only one bus route serves path $i$ and $j$; constraint ④ ensures that the number of routes entering and departing from each station is the same; constraint ⑤ is a typical Miller-Tucker-Zemlin (MTZ) constraint, the purpose of which is to eliminate possible sub-loops in the TSP problem; constraint ⑥ is the connecting bus route length constraint. , These represent the minimum and maximum lengths of the connecting bus routes, respectively; constraint ⑦ is the constraint on the number of connecting bus stops. , These represent the minimum and maximum number of stops for connecting buses, respectively; constraint ⑧ is the passenger transfer waiting time constraint, ensuring that the passenger transfer waiting time for connecting buses does not exceed the interval between two adjacent trains; constraint ⑨ is the shared bicycle parking station capacity limit, ensuring that the number of passengers renting bicycles at parking station x does not exceed the number of bicycles supplied at the parking station. The number of vehicles parked at the designated stop; constraints As a constraint on decision variables, when path i,j is selected, =1, otherwise 0.

[0158] Furthermore, in step six, a genetic algorithm is used to solve the model and encode the decision variables. The selection of connecting bus routes adopts a real number string encoding method.

[0159] Furthermore, the specific steps of step six are as follows:

[0160] (a) Importing coordinates: Import the latitude and longitude information of the bus stops and bicycle parking points into the Matlab software.

[0161] (II) Basic Parameter Settings: Input the data for all basic parameters in the objective function model, and set the population size (sizepop), maximum number of iterations (maxgen), generation gap, and crossover probability for the genetic algorithm. Probability of mutation .

[0162] (III) Population Initialization: Number each bus stop, assign "0" to the rail transit connection station, and number the remaining bus stops from 1 to n. A random array of 1 to n is generated. Based on the station size within the study area and reasonable limits on the number of stops for connecting bus routes, the minimum number of connecting bus routes within the area is automatically generated. and maximum number of lines , for different numbers of connecting bus lines , generate random numbers respectively to divide the station array into several groups, and adding "0" before the first number in each group can represent the driving route of the connecting bus.

[0163] (4) Constraint checking: After obtaining the initial solution of the corresponding population size, input the constraint conditions established in step five. The solution that satisfies the constraint conditions is the feasible initial solution.

[0164] (5) Start iteration: Input the initial iteration number gen = 0. When gen < maxgen is satisfied, perform iteration. When it is not satisfied, the maximum iteration number is reached, and the iteration can be ended.

[0165] (6) Calculate the fitness of the objective function: Determine the departure time of the connecting bus according to the historical passenger flow data and the arrival time of the rail transit train. Calculate the objective function value for the initial solution, compare the objective function values under different numbers of connecting bus lines, and retain the optimal solution. The fitness of the solution is the reciprocal of the objective function value.

[0166] (7) Selection: Use the roulette wheel method to select the population, and leave the optimal individuals according to the set generation gap to form the offspring population.

[0167] (8) Crossover: The present invention adopts simulated binary crossover (Simulated binary crossover) which is more suitable for real number coding. This method has a good effect on local search in chromosomes with real number coding. After performing the crossover operation, update the current population.

[0168] (9) Mutation: Similar to the crossover operation, adopt polynomial mutation (polynomial mutation) with better effect for real number coding to update the current population.

[0169] (10) After completing one iteration process, increment the iteration number gen by 1, and return to (5) for the next iteration.

[0170] (11) After reaching the maximum iteration number, obtain the connecting bus routes that traverse all current bus stops. However, the passenger flow at some stops is relatively low and the bus selection probability is not high. Considering skipping stops can reduce the travel cost of passengers and the operating cost of the bus company. Therefore, it is necessary to take the maximum sum of bus selection probabilities as the objective function, with the stop selection as the decision variable, and obtain the result through the genetic algorithm again.

[0171] (12) Initialize the population: The stop selection adopts 0-1 binary coding, where "1" represents stopping at this stop and "0" represents not stopping. Initialize chromosomes of the same length according to the results of the first genetic algorithm above.

[0172] (xiii) Constraint check: The number of bus stops after the station selection should meet the constraints, that is, the number of "1"s in the chromosome should be within the constraint range.

[0173] (XIV) Start iteration: The iteration process and steps are the same as the first genetic algorithm.

[0174] (XV) Calculate the objective function and fitness: The objective is to select the stop with the sum of the probabilities of bus selection as the goal, and the fitness is the value of the objective function.

[0175] (xvi) The genetic operators and iterative processes are similar to (7)-(10), but for binary encoding, the crossover operator is selected to use multi-point crossover, and the mutation operator is selected to use multi-point mutation.

[0176] (xvii) After reaching the maximum number of iterations, the optimal route of the connecting bus after the stop selection and its departure timetable are obtained.

[0177] Example 1: Please refer to Figure 3-4 In this embodiment of the invention, Yangchangcun Station on Kunming Metro Line 2 is selected as the research object, and the research area is delineated by Hongyun Road, Longquan Road, Linyu Road, and Beijing Road. There are 23 bus stops and 34 shared bicycle parking areas within the area, and latitude and longitude information is extracted from the Gaode Map Open Platform to calculate distances.

[0178] The survey revealed the following values ​​for some indicators in the model: average walking speed. Average speed of connecting buses Average bus stop time Average time for passengers to walk from the train to the rail transit station connecting bus fare Passenger unit time cost Bus operating costs Average cycling speed .

[0179] Based on the limitations of station size and number of stops within the study area, a minimum number of connecting buses is specified for the area. and maximum number For different numbers of lines within the region Generate respectively The station array is divided into several groups using random numbers. There are 23 bus stops in total, and the constraint is that each connecting bus route stops at between 7 and 12 stations. =2,3. Using the randi function to generate the random number "10", the chromosome is truncated from the 10th number, splitting into two connecting bus routes.

[0180] In this embodiment, the line length of the current solution is first calculated based on the distance between stations. Then, based on the arrival times of Kunming Metro Line 2 trains at Yangchangcun Station between 7:00 and 9:00, the reasonable arrival times of connecting bus stations are estimated. Based on the passenger flow data obtained from the survey, the decision to depart is made according to the following rules. After obtaining the departure times, the average waiting time for passengers is calculated, and finally, the objective function value is obtained.

[0181] In this embodiment, the maximum passenger capacity of the connecting bus is 70 people. Scenario 1: If the number of passengers transferring from the first bus to the rail transit exceeds 70, both the first and subsequent buses must depart. Scenario 2: If the number of passengers on the first bus plus the number on the subsequent bus is less than 70, the first bus will not depart, and only the second bus will depart. Scenario 3: If the number of passengers on the first bus plus the number on the subsequent bus exceeds 70, the first bus will depart, and the number of passengers on the second bus will be added to the number on the subsequent bus to determine whether to depart.

[0182] The optimal route for connecting buses and its departure timetable were obtained after final calculation.

[0183] Please see Figure 5-6 The device includes an acquisition module. This module is used to acquire information on all bus stops and shared bicycle parking spots within the study area, passenger flow information for connecting rail transit during the study period, and basic parameters of passenger travel.

[0184] The acquisition modules include a site acquisition module, a visitor flow acquisition module, and a parameter acquisition module.

[0185] The site acquisition module is used to obtain information on all bus stops and shared bicycle stops within the study area. The information includes the location of bus stops and shared bicycle stops, the distance between bus stops, the distance of shared bicycle stops from rail transit stations, and the distance of bus stops and shared bicycle stops from the nearest residential area.

[0186] The passenger flow acquisition module is used to identify and acquire passenger flow data for each station using connecting buses and each stop using shared bicycles to transfer to rail transit during the study period.

[0187] The parameter acquisition module is used to obtain basic parameter information such as the average speed of connecting buses, shared bicycles, and walking; the fare of connecting buses and shared bicycles, transfer time, walking distance to the station, passenger unit travel cost, unit operating cost; and the capacity and supply of shared bicycle parking spots.

[0188] A calculation module is provided to import the information and parameters obtained from the previous acquisition module into the model established in steps three through six, and to perform calculation and iterative operations through a genetic algorithm until the iteration ends.

[0189] An output module is provided to determine the optimal route and final departure timetable of the connecting buses obtained through iterative calculation in the calculation module, and to output the route map and timetable in an intuitive manner.

[0190] The device also includes a memory, a processor, and a computer code program that can execute the aforementioned mathematical models and genetic algorithms.

[0191] The memory is used to store the aforementioned computer program. The processor is used to read the computer program from the memory and execute the contents described in steps three through six of claims.

[0192] To fulfill the above technical requirements, the memory is connected to the processor via a data bus, but may also be connected via, but not limited to, an address bus or a control bus. When the computer is running, the computer program stored in the memory can be read by the processor through computer instructions and the contents described in steps three through six can be executed to optimize the connecting public transportation within the research area. In this embodiment, the computer program is written using Matlab software, but may also be written using, but not limited to, other software or languages ​​such as Python.

[0193] Furthermore, a computer program is stored therein, which, after being read by the processor, executes the contents described in steps three through six of the claims.

[0194] Computer storage media includes volatile and non-volatile storage media. Volatile storage media cannot retain data after the computer is powered off, and mainly include Random Access Memory (RAM), Dynamic Random Access Memory (SDRAM), and Static Random Access Memory (SRAM). Non-volatile storage media can retain data for a long time after the computer is powered off, and mainly include Read Only Memory (ROM), One-Time Programmable Memory (OTPROM), Erasable Programmable Memory (EPROM), and Electrically Erasable Programmable Memory (EEPROM). It also includes optical discs, floppy disks, and hard disk drives.

[0195] The above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. For those skilled in the art, modifications and variations can be made to the content, and such modifications and variations should be included within the protection scope of the present invention.

Claims

1. A method for optimizing rail transit feeder bus path and scheduling based on shared bicycle travel influence, characterized in that, It includes the following steps: Step 1: Select the rail transit station to be optimized and demarcate the optimization area; Step 2: Extract the information of all bus stops and shared bicycle docking points within the research area, obtain the current operation mode of the selected rail transit station, obtain the traffic behavior data of passengers from their residences to the nearest bus stop and the nearest shared bicycle docking point, investigate the passenger flow of each stop, and obtain the traffic data in the transfer behavior; Step 3: Construct an objective function model that minimizes the total cost of passengers and operations for transfer by bus and shared bicycle; Step 4: Construct a Logit model for the probability of travel mode selection; Step 5: Construct constraint conditions to conform to reality and achieve reasonable solution results; Step 6: Solve the model to obtain the optimization results of the transfer bus route and scheduling under the influence of shared bicycle travel; The specific steps of Step 3 are as follows: (1) The time for a passenger to walk to the nearest bus stop i from the station is , the time for a passenger to walk to the nearest shared bicycle stop x from the station is , the distance between bus stops is , the shared bicycle connection riding distance is , the number of stops along the route of the bus l is , the stop time of the station is , the train arrival time at the station, the bicycle supply at the stop x is , and the travel demand for renting a bicycle at the stop is ; (ii) Based on the distance between stations Connection and shared bicycle riding distance You can get a connecting bus. travel time and shared bicycle riding time Based on the number of stops along the route of the connecting bus l and station stop time You can get the total stop time. The transfer waiting time between connecting buses and rail transit can be obtained from the train's arrival time, and is expressed as follows: ,in Let be the arrival time of the nth train. This refers to the arrival time of the kth bus in the connecting bus series. The average time it takes for passengers to walk from disembarkation to the platform level of the rail transit station to wait for their train. The time a passenger spends waiting at a bus stop for a connecting bus can be expressed as: ,in This indicates the number of shuttle bus departures per hour, which can be found through the shuttle bus departure times. Calculations show that the transfer waiting time for passengers using shared bicycles can be expressed as half the time interval between two adjacent rail transit train services, i.e. ; (III) Based on the number of stops x the number of bicycles supplied And the demand for bike rentals at this location. Define a penalty cost for borrowing and repaying, denoted as ,in It is the penalty coefficient for borrowing and repaying; (4). Classify the scenarios of returning a shared bicycle into three categories: ； in These represent the proportion of the total morning rush hour during which the parking spaces at the return point fall into one of the three categories. Additionally, a condition with ≤5 remaining parking spaces is defined as a situation where there are relatively few remaining parking spaces. The time cost incurred due to insufficient parking space capacity is also divided into three categories. When there are many remaining parking spaces, the rider's return time is... When there are few parking spaces remaining, the cyclist's return time is... If no parking spaces remain, assuming the rider chooses the nearest parking spot to their current location, the return time will be [time period missing]. , This indicates the distance of the cyclist from the nearest rest stop. This represents the average riding speed of passengers and the average return time of shared bicycle riders. ; (5). The passenger cost of the transfer bus is: （1） It includes the passenger's walking time to the station, vehicle travel time, stop time, waiting time, transfer time, and fare cost, In equation (1): i and j represent alternative stops for connecting buses, and S represents the set of alternative stops for connecting buses. This represents the cost per unit of time for passengers. This indicates the time it takes for a passenger to walk to the nearest bus stop, i. This represents the shortest travel distance between stop i and stop j for the connecting bus l. Indicates the bus fare. This represents the passenger flow at bus stop i. 0-1 decision variables: ； (6). The passenger cost of the shared bicycle is: （2） It includes the passenger's walking time to the docking point, cycling time, transfer waiting time, capacity penalty cost, borrowing and returning imbalance penalty cost, and fare cost, In equation (2): x and y represent shared bicycle parking stations. This represents the set of alternative parking spots for shared bicycles. This indicates the time it takes for a passenger to walk to the nearest shared bicycle parking spot, x. This indicates the imbalance penalty coefficient. Indicates the price of shared bicycles. This represents the passenger flow at shared bicycle parking point x. (7). The operating cost of the transfer bus is: （3） In formula (3): This indicates the unit cost of the connecting bus operator. This indicates the number of buses departing per hour for bus route l. This represents the shortest travel distance between station i and station j for the connecting bus l; (viii) The operating cost of shared bicycles is ; (9). The constructed objective function model is: （4）。 2. The method for optimizing rail transit connection bus routes and scheduling based on the impact of shared bicycle travel according to claim 1, characterized in that: The information of the bus stops and shared bicycle docking points in Step 2 includes the locations of the bus stops and shared bicycle docking points, the distance between bus stops, the distance of the shared bicycle docking point from the rail transit station, and the distances of the bus stop and shared bicycle docking point from the nearest community; the operation mode in Step 2 is the arrival time of the trains in both directions during the morning peak; the traffic behavior data in Step 2 includes the walking distance and walking speed of passengers, and the average transfer time of passengers from the bus stop and docking point to the rail transit station; The passenger flow of the stops in Step 2 refers to the passenger flow of taking the bus and shared bicycle to transfer to the rail transit during the morning peak on weekdays; the traffic data of the transfer behavior includes the average driving speed of the transfer bus, stop time, average cycling speed of the bicycle, the supply quantity of bicycles at each docking point, capacity, the passenger time cost of bus travel, the passenger time cost of shared bicycle travel, the bus travel fare, and the unit fare parameter of shared bicycle travel; 3. A method for optimizing rail transit connection bus routes and scheduling based on the impact of shared bicycle travel, as described in claim 1 or 2, characterized in that: The distances between bus stops and the distances between bus stops and rail transit stations were calculated. These distances were calculated based on latitude and longitude, which were obtained from a map platform. The morning peak hours were selected as the two hours with the highest travel volume in the morning. The entrance / exit with the highest passenger flow at the passenger's residence was chosen as the starting point, and the actual distance from the starting point to the nearest bus stop i was measured. The starting point is selected from the entrance / exit with the highest passenger flow at the passenger's residence. The actual distance from the starting point to the nearest shared bicycle stop, x, is measured. Passenger flow for connecting buses is obtained through data from swiping bus IC cards and scanning QR codes for boarding. Passenger flow for connecting bicycles is obtained by identifying riding data of shared bicycles through mobile phone signaling and shared bicycle data. The sum of the two is the total passenger flow for connecting rail transit in the area.

4. The method for optimizing rail transit connection bus routes and scheduling based on the impact of shared bicycle travel according to claim 1, characterized in that: The Logit model in Step 4 is: （5） In formula (5): =1,2, Let represent the probability of passengers choosing between connecting buses and shared bicycles, respectively. Let θ represent the passenger costs for these two connection methods, and θ represent the utility coefficient.

5. The method for optimizing rail transit connection bus routes and scheduling based on the impact of shared bicycle travel according to claim 1, characterized in that: The specific steps of Step 6 are as follows: (A) Coordinate point import: Import the longitude and latitude information of the obtained bus stops and bicycle docking points into the Matlab software; (B) Basic parameter settings: Input all basic parameters of the objective function model and set the population size (sizepop), maximum number of iterations (maxgen), generation gap, and crossover probability for the genetic algorithm. Probability of mutation ; (C) Population Initialization: Number each bus stop, number the rail transit connection station as "0", and number the remaining bus stops from 1 to n. A random array of 1 to n is generated. Based on the station size within the study area and a reasonable limit on the number of stops for connection routes, the minimum number of connection bus routes within the area is automatically generated. and maximum number of lines For different numbers of connecting bus routes Generate respectively Divide the station array into several groups using random numbers, and add "0" before the first number in each group to represent the connecting bus route; (D) Constraint check: After obtaining the initial solution of the corresponding population size, input the established constraint conditions, and the solution that satisfies the constraint conditions is the feasible initial solution; (E) Start iteration: Input the initial iteration number gen = 0, and perform iteration when gen < maxgen is satisfied. When it is not satisfied, the maximum iteration number is reached, and the iteration can be ended; (F) Calculate the fitness: Calculate the objective function value for the initial solution, compare the objective function values under different numbers of transfer bus routes, retain the optimal solution, and the fitness of the solution is the reciprocal of the objective function value; (G) Selection: The population is selected using a roulette wheel method, and the best individuals are retained to become the offspring population according to the set generation gap; (H) Crossover: Simulated binary crossover is used, and the population is updated after the crossover operation; (I) Mutation: Polynomial mutation is used for the real number encoded part to update the current population; (J) After completing one iteration, the iteration number gen+1 is incremented, and (E) is returned to perform the next iteration; (K) After reaching the maximum number of iterations, the connecting bus routes that traverse all current bus stops are obtained. The objective function is to maximize the sum of bus selection probabilities, and the station selection is the decision variable. The result is obtained again through the genetic algorithm. (L) Initialize the population: The station selection adopts 0-1 binary encoding, "1" means to stop at the station, "0" means not to stop. Based on the results of the first genetic algorithm, initialize chromosomes of the same length. (M) Constraint test: The number of bus stops after station selection should meet the constraint, that is, the number of "1"s in the chromosome should be within the constraint range. (N) Start iteration: The iteration process and steps are the same as the first genetic algorithm; (O) Calculate the objective function and fitness: Select the stop with the goal of maximizing the sum of the probabilities of choosing public transport, and the fitness is the value of the objective function; (P) For binary encoding, choose multi-point crossover for the crossover operator and multi-point mutation for the mutation operator; (Q) After reaching the maximum number of iterations, the optimal route of the connecting bus after the stop selection and its departure timetable are obtained.

6. A device for optimizing rail transit connection bus routes and scheduling based on the impact of shared bicycle travel, characterized in that, The device includes: The acquisition module is used to acquire information on all bus stops and shared bicycle stops within the study area, passenger flow information of connecting rail transit during the study period, and basic parameters of passenger travel in steps one and two as described in claim 1. The calculation module performs calculations on the information obtained from the acquisition module. The calculation method is as follows: based on the model involved in steps three to six as described in claim 1, and performs calculations and iterative operations through a genetic algorithm. The output module is used to determine the optimal route and final departure timetable of the connecting buses obtained from the iterative calculation in the calculation module, and outputs the route map and timetable in an intuitive way.

7. The rail transit connection bus route and scheduling optimization device based on the impact of shared bicycle travel according to claim 6, characterized in that, The device also includes a memory, a processor, and a computer code program capable of executing the above-described model and genetic algorithm. The memory is used to store the computer program, and the processor is used to read the computer program from the memory and execute steps three through six as described in claim 1.

8. A computer storage medium for optimizing rail transit connection bus routes and scheduling based on the impact of shared bicycle travel, characterized in that, The device contains a computer program that, after being read by a processor, executes the method as described in claim 1.

Images