Passenger selection behavior-based connection bus line sharing optimization method and equipment

Abstract

The invention discloses a passenger selection behavior-based connection bus line-sharing optimization method and equipment, belongs to the technical field of urban traffic, and is used for solving the problems that existing connection buses are unreasonable in line-sharing station design, the cost of passengers and bus enterprises is difficult to reduce, the service level of the connection buses is not favorably improved, and the service quality of the connection buses is influenced. And the travel time of passengers is difficult to optimize. The method comprises the following steps: determining generalized travel cost based on waiting time cost, in-transit time cost and bus company operation cost; according to the generalized travel cost, carrying out regression analysis on the travel selection behavior influence factors of the passengers, and determining a connection bus route selection probability; carrying out model condition constraint under a related cost objective function on the connection bus to obtain a connection bus common line segment station optimization constraint condition set; constructing a connection bus common line segment station optimization model; and carrying out output calculation on the connection bus line-sharing station optimization model to obtain a connection bus line-sharing station optimization scheme and a departure interval.

Description

Technical Field

[0001] The present application relates to the field of urban transportation, and in particular to a method and device for optimizing a common line segment of a shuttle bus based on passenger selection behavior. Background Art

[0002] The topological structure of rail transit feeder bus routes can be classified according to their shapes. Specifically, if the layout of the bus route is like a ring, then it is a ring bus route; if the layout of the bus route is like a straight line, then it is a linear bus route. On this basis, when there are multiple consecutive transfer stations on a rail transit line, and the shape of the bus routes connected to it is similar to an I-shaped bus line arranged side by side, such bus routes are called "I-shaped" bus routes. In short, "I-shaped" bus routes refer to those feeder bus routes that are similar to the shape of an "I" near multiple consecutive transfer points on the rail transit line.

[0003] The connection mode between subway and shuttle bus is to deliver passengers at bus stops to subway stations and disperse passengers exiting subway stations to their respective destinations. In order to optimize the stops of different bus lines on the same line section and adjust the vehicle departure interval of the "I-shaped" bus line on the same line section, it is necessary to comprehensively consider the subway shuttle stations, bus stops, passenger flow demand and the transfer relationship between stations. The goal is to find the best stop plan for the shuttle line station to minimize the bus operating cost and passenger travel cost. Summary of the invention

[0004] The embodiment of the present application provides a method and device for optimizing the common line segment of a shuttle bus based on passenger selection behavior, which is used to solve the following technical problems: the existing shuttle bus station design on the common line segment is not reasonable enough, it is difficult to reduce the costs of passengers and bus companies, it is not conducive to improving the service level of the shuttle bus, and it is difficult to optimize the travel time of passengers.

[0005] The present application embodiment adopts the following technical solutions: On the one hand, the embodiment of the present application provides a method for optimizing the common line segments of shuttle buses based on passenger selection behavior, including: determining the generalized travel cost based on the waiting time cost, the en route time cost and the bus company operating cost; performing a regression analysis on the factors affecting the passenger's travel selection behavior based on the generalized travel cost and through a preset utility maximization theory to determine the passenger's shuttle bus line selection probability; wherein the shuttle bus line selection probability includes: the passenger's choice of ride plan probability and the transfer station probability; and subjecting the shuttle bus to a model under the relevant cost objective function. Condition constraints are set to obtain the optimization constraint condition set of the common line segment stations of the shuttle bus; wherein the optimization constraint condition set of the common line segment stations of the shuttle bus includes: station stop constraints, competition and cooperation constraints, departure interval constraints, load rate constraints and departure number constraints; based on the optimization constraint condition set of the common line segment stations of the shuttle bus and the selection probability of the shuttle bus line, an optimization model of the common line segment stations of the shuttle bus is constructed; through the preset NSGA-II algorithm, the optimization model of the common line segment stations of the shuttle bus is output and calculated to obtain the optimization plan of the common line segment stations of the shuttle bus and the departure interval.

[0006] The embodiments of the present application help to minimize the commuting time of most passengers by introducing accurate departure times and departure intervals. In addition, the inclusion of transfer penalties and station optimization further reduces the travel time cost, and the probability of commuters choosing shuttle buses is significantly increased. The benefits of these optimization measures attract more passengers to choose shuttle buses, increase the operating income of bus companies, and thus enable bus companies to escape the vicious cycle of declining passenger flow. At the same time, based on the time-varying passenger flow and travel time between stations, considering the changes in passenger flow distribution caused by changes in passenger travel behavior, studying passenger flow demand and competitive relationships, analyzing the changing trend of the attractiveness of shuttle bus lines to passengers, and then proposing a station optimization plan for shuttle buses on co-linear segments, in order to reduce the costs of passengers and bus companies, improve the service level of shuttle bus lines, and better achieve coordination and cooperation with rail transit modes.

[0007] In a feasible implementation, the generalized travel cost is determined based on the waiting time cost, the travel time cost and the bus company operating cost, specifically including: , get the waiting time cost ;in, is the passenger waiting cost per unit time; It represents the average waiting time for passengers and is half of the bus line's departure interval; Representative train number On the shuttle bus The number of passengers waiting at the station; F The departure interval of the connecting bus; , get the transit time cost ;in, for Stand Actual distance between stations; is the speed of the bus; is the number of transfers; is the average vehicle stop time; is the average transfer time; , get the operating cost of the bus company ;in, is the number of vehicles dispatched per day; The cost required for operating a single trip on a single route is the cost required; the waiting time cost, the en route time cost and the bus company operating cost are combined to obtain the generalized travel cost.

[0008] In a possible implementation, according to , get the line Number of flights per day ;in, For line Daily operating hours; For line Half of the interval between departures; , get the cost required to operate a single trip on a single route ;in, is the energy consumption cost per unit mileage; is the energy consumption cost of parking per unit time; For line The number of shifts operated per day; is the actual distance between two adjacent stations; i For the station; is the average vehicle stop time.

[0009] In a feasible implementation, according to the generalized travel cost and through the preset utility maximization theory, a regression analysis is performed on the influencing factors of the passenger's travel selection behavior to determine the passenger's connection bus line selection probability, specifically including: based on the influencing factors of each cost in the generalized travel cost, determining the influencing factors of the riding plan selection behavior and the influencing factors of the transfer station selection behavior; wherein the riding plan selection behavior influencing factors include: waiting time for the vehicle, the congestion level of the optional vehicles, and the estimated riding time factor; the transfer station selection behavior influencing factors include: transfer waiting time factor, the walking distance factor required for transfer, and the congestion level of the current vehicle; according to , get the probability of the passenger choosing the ride plan ;in, Select a vehicle on behalf of the passenger possibility; represents a set of optional vehicles for passengers, and the set of optional vehicles for passengers also represents the two vehicles that the passenger will choose between that arrive first and meet his / her travel needs; is the waiting vehicle time factor; is the optional vehicle congestion level; The estimated ride time factor; , , , They are all parameters to be estimated in the probability of the passenger choosing a ride plan; Refers to the line and train number vehicles; , get the transfer station probability ;in, Indicates that the passenger chooses a transfer station probability; It is a collection of transfer stations; , , , are all parameters to be estimated for the transfer station probability; is the transfer waiting time factor; The walking distance factor required for the transfer; is the current degree of congestion in the vehicle; through a preset binary logit regression analysis algorithm, the generalized travel cost is regressed and fitted with the probability of the passenger choosing a travel plan and the probability of the transfer station to determine the probability of selecting the connecting bus line.

[0010] In a feasible implementation manner, before subjecting the connecting bus to the model condition constraints under the relevant cost objective function to obtain the optimization constraint condition set of the connecting bus co-line segment station, the method further includes: , and obtain the cooperation coefficient ;in, For bus routes The number of sites owned; For bus routes The number of sites owned; For bus routes With bus lines The number of public sites; according to , get bus routes l Passenger flow attraction ;in, Select a route for shared-line stations passenger flow; is the total passenger flow between co-line stations; , and obtain the correction factor ;in, Respectively represent the passenger flow attraction based on Bus routes Passenger flow attraction and bus routes passenger flow attraction; according to , and obtain the competition intensity of the connecting bus co-line section ;in, is the competition coefficient.

[0011] In a feasible implementation manner, before subjecting the connecting bus to the model condition constraints under the relevant cost objective function to obtain the optimization constraint condition set of the connecting bus co-line segment station, the method further includes: , we get the objective function of minimizing passenger travel cost: ;in, The interval between departures; is the passenger waiting cost per unit time; Represents the average waiting time of passengers; From the transfer point To the transfer point Penalty function for too many transfers; It is a congestion function and is used to reflect the passengers’ comfort level when riding. For transfer point To the transfer point The maximum number of passengers that can be carried by the shuttle buses; is the number of connecting bus lines within the common line segment; for Stand Actual distance between stations; is the speed of the bus; for Stand Number of transfers between stations; is the average vehicle stop time; is the average transfer time; Representatives at the transfer point The probability of passengers choosing the ride plan; Representatives at the transfer point The probability of a transfer station; according to , we get the objective function of minimizing the bus operation cost ;in, The cost required to operate a single trip on a single route; is the daily departure volume; wherein the cost objective function includes: the passenger travel cost minimization objective function and the public transportation operation cost minimization objective function.

[0012] In a feasible implementation, the connecting bus is subjected to model condition constraints under the relevant cost objective function to obtain a set of optimization constraint conditions for the connecting bus co-line segment stations, specifically including: , and obtain the degree of competition and cooperation used for the competition and cooperation constraint ;in, is the number of connecting bus lines within the common line segment; is the competition intensity of the shared line section of the shuttle bus; , get the departure interval of the connecting bus line for the departure interval constraint ;in, It refers to the minimum departure interval of connecting bus routes; Refers to the maximum departure interval of the connecting bus line; , the load rate constraint is obtained; wherein, The number of seats on the shuttle bus; is the peak hour; q For the The cross-sectional passenger flow of each station; is the minimum load factor; is the maximum load factor; according to , get the departure train constraint; where, Refers to the minimum number of vehicles required for the feeder bus routes based on the existing operation arrangements; is the total number of vehicle resources owned by the shuttle bus line; based on the condition that the shuttle bus common line segment station includes at least one bus route and the rail and bus public transfer station does not skip a station, the station stop constraint is formed; the station stop constraint, the competitive relationship constraint, the departure interval constraint, the load rate constraint and the departure number constraint are processed as a constraint set to obtain the shuttle bus common line segment station optimization constraint set.

[0013] In a feasible implementation manner, based on the optimization constraint set of the connecting bus co-line segment stations and the connecting bus line selection probability, an optimization model for the connecting bus co-line segment stations is constructed, specifically including: taking the connecting bus line selection probability and the generalized travel cost as input quantities of the optimization model, and taking the optimization constraint set of the connecting bus co-line segment stations as the constraint quantity of the optimization model, and taking both the passenger travel cost minimization objective function and the bus operation cost minimization objective function as the optimization objectives of the optimization model; based on the input quantity, constraint quantity and optimization objective of the optimization model, the optimization model for the connecting bus co-line segment stations is constructed.

[0014] In a feasible implementation manner, the shuttle bus co-linear segment station optimization model is output and calculated by a preset NSGA-II algorithm to obtain the shuttle bus co-linear segment station optimization plan and the departure interval, which specifically includes: using a preset non-dominated sorting genetic algorithm, the population parameters of the shuttle bus co-linear segment station optimization model are calculated for the initial population chromosome encoding under the station stop plan and the departure frequency to obtain an initial solution set; according to the objective function of the shuttle bus co-linear segment station optimization model, the initial solution set is evaluated for adaptability to obtain a fitness value; based on the evaluated fitness value, the shuttle bus co-linear segment station optimization model is iteratively calculated for the optimal solution to generate a new generation of population; and the new generation of population is output processed to obtain the shuttle bus co-linear segment station optimization plan and the departure interval; wherein, the iterative optimal solution calculation includes: selection mechanism calculation, recombination strategy calculation and mutation operation calculation.

[0015] On the other hand, an embodiment of the present application also provides a device for optimizing the common line segments of shuttle buses based on passenger selection behavior, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, so that the at least one processor can execute a method for optimizing the common line segments of shuttle buses based on passenger selection behavior as described in any of the above embodiments.

[0016] The present application provides a method and device for optimizing the common line segment of shuttle buses based on passenger selection behavior. Compared with the prior art, the embodiments of the present application have the following beneficial technical effects: The embodiments of the present application help to minimize the commuting time of most passengers by introducing accurate departure times and departure intervals. In addition, the inclusion of transfer penalties and station optimization further reduces the travel time cost, and the probability of commuters choosing shuttle buses is significantly increased. The benefits of these optimization measures attract more passengers to choose shuttle buses, increase the operating income of bus companies, and thus enable bus companies to escape the vicious cycle of declining passenger flow. At the same time, based on the time-varying passenger flow and travel time between stations, considering the changes in passenger flow distribution caused by changes in passenger travel behavior, studying passenger flow demand and competitive relationships, analyzing the changing trend of the attractiveness of shuttle bus lines to passengers, and then proposing a station optimization plan for shuttle buses on co-linear segments, in order to reduce the costs of passengers and bus companies, improve the service level of shuttle bus lines, and better achieve coordination and cooperation with rail transit modes. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 A flow chart of a method for optimizing a common line segment of a connecting bus based on passenger selection behavior provided in an embodiment of the present application; Figure 2 A NSGA-II algorithm program structure diagram provided in an embodiment of the present application; Figure 3 A diagram of an encoding method based on optimization of common line segments of connecting buses provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of a connecting bus common line segment optimization device based on passenger selection behavior provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0019] It should be noted that when passengers' trips are limited to the common line area, they are faced with multiple bus line choices, and may also choose between two consecutive buses on the same line. Specifically, some passengers will not simply choose the first bus to arrive, but will refer to the real-time status of the subsequent buses and make a comprehensive decision based on this. Such bus selection behavior will lead to changes in passenger flow distribution between different lines within the common line segment, and will also affect the passenger capacity of the previous and next vehicles on the same line.

[0020] When passengers transfer between co-linear sections, they consider multiple factors, including the congestion level of the current vehicle, the environmental conditions of the transfer station, and the real-time information of the vehicle to be transferred. However, since the passenger boarding and alighting behaviors at the transfer station and each station in the co-linear section will affect the vehicle congestion, it is difficult for passengers to accurately grasp the specific congestion level when the transfer vehicle arrives at each transfer station. Therefore, passengers often cannot use this indicator as the main basis for decision-making. Based on this, it can be considered that the transfer choice of passengers in the co-linear section is mainly affected by three factors: the congestion level of the current vehicle, the environmental conditions of the transfer station, and some real-time information of the transfer vehicle (such as the arrival time and location of the vehicle).

[0021] The present application embodiment provides a method for optimizing the common line segments of connecting buses based on passenger selection behavior, such as Figure 1 As shown, the method for optimizing the common line segment of connecting buses based on passenger selection behavior specifically includes steps S101-S105: S101. Determine the generalized travel cost based on the waiting time cost, the travel time cost and the bus company's operating cost.

[0022] It should be noted that, from the passenger side, it is necessary to consider the broad travel costs of passengers throughout the entire journey, which covers the waiting time costs and in-transit time costs of all passengers.

[0023] Specifically, firstly according to , get the waiting time cost .in, is the passenger waiting cost per unit time; It represents the average waiting time for passengers and is half of the bus line's departure interval; Representative train number On the shuttle bus The number of passengers waiting at the station; F It is the departure interval of the connecting bus.

[0024] Further, according to , get the time cost in transit .in, for Stand Actual distance between stations; is the speed of the bus; is the number of transfers; is the average vehicle stop time; is the average transfer time.

[0025] Further, according to , get the bus company operating cost .in, is the number of vehicles dispatched per day; The cost required to operate a single train on a single route.

[0026] As a feasible implementation method, according to , get bus routes Number of flights per day .in, For bus routes Daily operating hours; For bus routes Half of the departure interval. , get the cost required to operate a single trip on a single route ;in, is the energy consumption cost per unit mileage; is the energy consumption cost of parking per unit time; For line The number of shifts operated per day; is the actual distance between two adjacent stations; i For the station; is the average vehicle stop time.

[0027] Furthermore, the waiting time cost, the travel time cost and the bus company operating cost are combined to obtain the generalized travel cost.

[0028] S102. Based on the generalized travel cost and the preset utility maximization theory, a regression analysis is performed on the factors affecting the passenger's travel choice behavior to determine the passenger's connection bus line selection probability. The connection bus line selection probability includes: the passenger's choice of travel plan probability and the transfer station probability.

[0029] Specifically, based on the influencing factors of each cost in the generalized travel cost, the influencing factors of the travel plan selection behavior and the transfer station selection behavior are determined. Among them, the influencing factors of the travel plan selection behavior include: waiting time for the vehicle, the congestion of the available vehicles, and the estimated travel time; the influencing factors of the transfer station selection behavior include: transfer waiting time, the walking distance required for transfer, and the congestion of the current vehicle.

[0030] In one embodiment, the travel patterns of travelers vary from person to person, but for the convenience of research, the behavioral differences between passengers are ignored and it is assumed that the behavior of all passengers is consistent. The main factors affecting the two travel choices of passengers are analyzed as follows. For the behavior of choosing a ride plan, the influencing factors include: the waiting time for the vehicle , the level of congestion of available vehicles and estimated ride time As for the choice of transfer station, the influencing factors are: transfer waiting time , Walking distance required for transfer and the current level of traffic congestion The utility of different options for travelers can be expressed by the utility function: That is, the utility of different options for travelers can be expressed by the utility function: and ,in, Representative vehicle The utility value of Refers to the line and train number vehicles; Indicates transfer station The utility value of .

[0031] Further, according to , get the probability of passengers choosing the ride plan .in, Select a vehicle on behalf of the passenger possibility; represents the set of optional vehicles for passengers, which also means that the passenger will choose between the two vehicles that arrive first and meet his travel needs; The waiting time factor for the vehicle; is the optional vehicle congestion level; The estimated ride time factor; , , , They are all parameters to be estimated in the probability of passengers choosing a ride plan; Refers to the line and train number of vehicles.

[0032] Further, according to , get the transfer station probability .in, Indicates the passenger chooses a transfer station The probability of It is a collection of transfer stations; , , , These are the parameters to be estimated for the probability of transfer stations; The waiting time factor for transfer; Factor in the walking distance required for transfers; The current level of vehicle congestion.

[0033] Furthermore, through the preset binary logit regression analysis algorithm, the generalized travel cost is regressed and fitted with the probability of passengers choosing a travel plan and the probability of transfer stations to determine the probability of selecting the connecting bus line.

[0034] S103, subjecting the shuttle bus to model condition constraints under the relevant cost objective function, and obtaining a set of optimization constraint conditions for the shuttle bus co-line segment stations. The set of optimization constraint conditions for the shuttle bus co-line segment stations includes: station stop constraints, competition and cooperation constraints, departure interval constraints, load rate constraints, and departure number constraints.

[0035] Specifically, according to , and obtain the cooperation coefficient .in, For bus routes The number of sites owned; For bus routes The number of sites owned; For bus routes With bus lines The number of public sites.

[0036] As a feasible implementation method, since the two shuttle buses have an intersection in their service areas, passengers have multiple choices when traveling, which forms collinear competition. By comparing the ratio of the actual number of non-collinear stations on the two shuttle bus lines to their theoretical maximum number of non-collinear stations, collinear cooperation is formed. Competition coefficient of competition intensity: .

[0037] In one embodiment, the competition intensity is used to measure the competition and cooperation relationship of the connecting bus co-line section. Assume that there are two connecting buses in the connecting bus co-line section, and the route have Stations, lines have Stations, two lines public stations (i.e. transfer stations). The sum of the proportion of public stations to the total stations and the cooperation coefficient can be regarded as the competition intensity of the two lines. According to the competition and cooperation relationship mentioned above, if there are more than two transfer stations between the two connecting bus lines, the number of co-linear stations of the two lines is considered to measure their degree of competition. At the same time, the degree of cooperation between the two lines is evaluated by comparing the ratio of the actual number of non-co-linear stations of the two lines to their theoretical maximum number of non-co-linear stations (i.e. the cooperation coefficient). The sum of the proportion of public stations to the total stations and the cooperation coefficient can be regarded as the competition intensity of the two lines. On this basis, combined with the passenger flow factor, it is corrected and the calculation method is as follows: First, according to , get bus routesl Passenger flow attraction .in, Select a route for shared-line stations passenger flow; is the total passenger flow between co-line stations.

[0038] Further, according to , and obtain the correction factor .in, Represents the passenger flow attraction Bus routes Passenger flow attraction and bus routes passenger flow attraction.

[0039] Further, according to , and obtain the competition intensity of the connecting bus co-line section ;in, is the competition coefficient.

[0040] It should be noted that it is necessary to construct a set of optimization constraints for the shared line segment of the connecting bus based on the constraints of the bus stop, competition and cooperation, departure interval, load rate and departure frequency. Then, considering the changes in passenger flow distribution on the shared line caused by the dynamic changes in passenger travel choice behavior, an optimization model for the shared line segment of the connecting bus is established with the minimization of passenger travel costs and bus company operating costs as the optimization goal.

[0041] This application needs to adopt the following in the optimization model stage: The impact of other bus routes outside the research scope on passenger travel is not considered. 2. Passengers are homogeneous and rational, and all passengers have the same sensitivity to travel costs and time.

[0042] 3. Passengers can get real-time information about bus stops.

[0043] 4. No stopping decisions are made at bus-subway transfer stations.

[0044] 5. Both bus routes have only one bus stop berth, and passengers can only get on and off the bus at the stop.

[0045] 6. The order of passengers waiting and boarding follows the first-come-first-out (FIFO) principle. At the same time, each station can only accommodate one bus at a time.

[0046] Further, according to , we get the objective function of minimizing passenger travel cost: .in, The interval between departures; is the passenger waiting cost per unit time; Represents the average waiting time of passengers; From the transfer point To the transfer point Penalty function for too many transfers; It is a congestion function and is used to reflect the passengers’ comfort level when riding. For transfer point To the transfer point The maximum number of passengers that can be carried by the shuttle buses; is the number of connecting bus lines within the common line segment; for Stand Actual distance between stations; is the speed of the bus; for Stand Number of transfers between stations; is the average vehicle stop time; is the average transfer time; Representatives at the transfer point The probability of passengers choosing the ride plan; Representatives at the transfer point The probability of a transfer station.

[0047] Further, according to , we get the objective function of minimizing the bus operation cost .in, The cost required to operate a single trip on a single route; The number of vehicles departing each day.

[0048] As a feasible implementation method, in the formula, is the congestion function, which reflects the passengers' comfort level when riding. The expression is: ,in, For Stand The weighted average of the station's passenger load; is the maximum number of passengers that the vehicle can carry; , is the crowding parameter, and Take 0.15, Take 2. From point To point The penalty function for too many transfers is expressed as: .

[0049] Among them, the cost objective function includes: the passenger travel cost minimization objective function and the bus operation cost minimization objective function.

[0050] Further, according to , and obtain the degree of competition and cooperation used to constrain the competition and cooperation relationship .in, is the number of connecting bus lines within the common line segment; It is the competition intensity of the shared line section of the shuttle buses.

[0051] That is to say, after defining the competition and cooperation intensity calculation of the "I-shaped" shuttle bus line, it is possible to quantify the competition and cooperation relationship between the two lines, and then deeply understand the interaction between them, which provides a theoretical basis for improving the public transportation system. Based on the competition and cooperation intensity calculation results of the two bus lines, this application proposes the concept of competition and cooperation degree, that is, competition and cooperation degree , and the calculated results are used as optimization constraints. At the same time, the competition and cooperation degree of the two lines can also be used as an indicator to evaluate the pros and cons of the station skipping strategy. When the cooperation intensity does not reach 1, the increase in competition and cooperation indicates that the cooperation relationship between the lines is closer.

[0052] Further, according to , get the departure interval of the connecting bus line used for departure interval constraint .in, It refers to the minimum departure interval of connecting bus routes; It refers to the maximum departure interval of the connecting bus line.

[0053] Furthermore, the passenger load factor of shuttle buses has a significant impact on passenger comfort and energy consumption. If the load factor is too high, the bus will become more crowded, thereby reducing passenger comfort; conversely, if the load factor is too low, it will lead to an increase in bus energy consumption. In view of this, the following constraints are imposed on the average load factor of buses: , we get the load rate constraint. Among them, The number of seats on the shuttle bus; is the peak hour; q For the The cross-sectional passenger flow of each station; is the minimum load factor; is the maximum load factor.

[0054] Further, according to , get the departure train constraints. Among them, Refers to the minimum number of vehicles required for the feeder bus routes based on the existing operation arrangements; It is the total number of vehicle resources available on the shuttle bus lines.

[0055] Furthermore, based on the condition that the connecting bus common line segment station includes at least one bus route and the rail and bus public transfer station does not skip a station, the station stop constraint is formed.

[0056] In one embodiment, the connecting bus common line segment station should include at least one bus route, and the rail and bus public transfer station does not skip stations, that is: and .

[0057] Furthermore, the station stop constraints, competition and cooperation constraints, departure interval constraints, load factor constraints and departure number constraints are processed as a set of constraints to obtain the optimization constraint set of the connecting bus shared line section stations.

[0058] S104. Based on the optimization constraint set of the connecting bus common line segment stations and the connecting bus line selection probability, a connecting bus common line segment station optimization model is constructed.

[0059] Specifically, the selection probability of the connecting bus line and the generalized travel cost are used as the input of the optimization model, and the optimization constraint set of the connecting bus co-line segment stations is used as the constraint of the optimization model. In addition, the objective function of minimizing the passenger travel cost and the objective function of minimizing the bus operation cost are both used as the optimization goals of the optimization model.

[0060] Furthermore, based on the input quantity, constraint quantity and optimization target of the optimization model, an optimization model for connecting bus co-line segments is constructed.

[0061] S105. The preset NSGA-II algorithm is used to output and calculate the optimization model of the connecting bus co-line segment station, and obtain the connecting bus co-line segment station optimization plan and departure interval.

[0062] It should be noted that the optimization model of the common line segment stations of the shuttle bus integrates the discrete 0-1 variables about whether the shuttle bus stops and the continuous variables used to determine the optimal departure frequency, so it belongs to a mixed integer nonlinear programming problem. This type of problem is more complex than the traditional vehicle routing problem and belongs to the category of NP-hard problems. Solving such problems is quite challenging, but heuristic algorithms are an effective solution. Such algorithms can find more satisfactory solutions in a shorter time and have faster convergence. In view of this, the present application adopts a non-dominated sorting genetic algorithm to solve this problem. By applying the NSGA-II algorithm, a set of optimized Pareto solutions can be obtained. A 0-1 variable is used to indicate whether each station within the optimization range is selected as a stop for the shuttle bus, and the problem is solved by the NSGA-II algorithm.

[0063] Specifically, the preset non-dominated sorting genetic algorithm is used to calculate the chromosome coding of the initial population under the relevant station stop plan and departure frequency for the population parameters of the connecting bus co-linear segment station optimization model, and the initial solution set is obtained.

[0064] Furthermore, according to the objective function of the optimization model of the connecting bus co-line segment station, the initial solution set is evaluated for adaptability to obtain the adaptability value.

[0065] Furthermore, based on the evaluated adaptability value, the optimal solution of the shuttle bus co-line segment station optimization model is iterated to generate a new generation of population. The new generation of population is output and processed to obtain the optimization plan of the shuttle bus co-line segment station and the departure interval. The iterative optimal solution calculation includes: selection mechanism calculation, recombination strategy calculation and mutation operation calculation.

[0066] As a feasible implementation method, Figure 2 A NSGA-II algorithm program structure diagram provided in the embodiment of the present application is as follows: Figure 2 As shown: Step 1: Parameter setting. Set the relevant parameters of the non-dominated sorting genetic algorithm (NSGA-II), including the population size , maximum number of iterations , crossover probability , mutation probability And basic parameters such as iteration counters.

[0067] Step 2: Generate an initial solution set. Randomly generate The bus stop plans are used as the initial population, each plan consists of a series of 0-1 variables Composition, of which The value of 1 means that a station is set at the candidate station, and 0 means that it is not set. Each scheme can be regarded as a chromosome, and the length of the chromosome is equal to the number of all candidate stations. The last two genes of each chromosome represent the departure interval of the connecting line vehicles. For example, a chromosome Represents a specific station layout plan and vehicle departure interval in generation 0. Figure 3 A coding method based on optimization of the common line segment of connecting buses provided in the embodiment of the present application is as follows: Figure 3 In the coding method shown, the track transfer stations within the co-line segment are still retained without skipping station optimization, and the remaining stations are taken as optimization targets, that is, the output result is as follows Figure 3 As shown in , 1 means stopping at the station, and 0 means not stopping at the station.

[0068] Step 3: Assessing chromosome fitness The objective function of the collinear shuttle bus optimization model is used as the fitness evaluation function of the chromosome, and the generated initial solution set is input into the function to calculate the initial fitness value.

[0069] Step 4: Design genetic operators (1) Selection. A binary tournament selection strategy is used. Each time, a certain number of individuals are drawn from the population and replaced in the sampling. The best individuals are selected to enter the offspring population until the original population size is reached. (2) Crossover: Using the single-point crossover method, two individuals in the parent generation are crossed to produce new individuals.

[0070] (3) Mutation. Mutation is to represent a gene value on a certain chromosome with other alleles and select the basic bit mutation operation.

[0071] For The layout scheme formed by each individual in the generation is evaluated for fitness and ranked according to the evaluation results. A new generation of population is formed according to the selection mechanism, recombination strategy and mutation operation of the NSGA-Ⅱ algorithm. .

[0072] Step 5: Verify the iteration end criteria If the current number of iterations Equal to the preset maximum number of iterations , and the fitness value reaches a stable state and cannot be further improved, the iteration is judged to be over. as an approximate optimal solution; otherwise, proceed to step 4.

[0073] In addition, the embodiment of the present application also provides a connecting bus co-line segment optimization device based on passenger selection behavior, such as Figure 4 As shown, the connecting bus common line segment optimization device 400 based on passenger selection behavior specifically includes: At least one processor 401. And a memory 402 in communication with the at least one processor 401; wherein the memory 402 stores instructions executable by the at least one processor 401, so that the at least one processor 401 can execute: The broad travel cost is determined based on the waiting time cost, the travel time cost and the bus company operating cost.

[0074] According to the generalized travel cost and the preset utility maximization theory, a regression analysis is conducted on the factors affecting passengers' travel choice behavior to determine the probability of passengers choosing the connecting bus line; among which, the connecting bus line selection probability includes: the probability of passengers choosing a ride plan and the probability of transfer stations.

[0075] The shuttle bus is subjected to model condition constraints under the relevant cost objective function to obtain a set of optimization constraint conditions for the shuttle bus common line segment stations; among them, the set of optimization constraint conditions for the shuttle bus common line segment stations includes: station stop constraints, competition and cooperation relationship constraints, departure interval constraints, load rate constraints and departure number constraints.

[0076] Based on the optimization constraint set of feeder bus co-line segment stations and the selection probability of feeder bus routes, an optimization model for feeder bus co-line segment stations is constructed.

[0077] Through the preset NSGA-II algorithm, the optimization model of the common line section stations of the connecting buses is output and calculated, and the optimization plan of the common line section stations of the connecting buses and the departure intervals are obtained.

[0078] The embodiments of the present application help to minimize the commuting time of most passengers by introducing accurate departure times and departure intervals. In addition, the inclusion of transfer penalties and station optimization further reduces the travel time cost, and the probability of commuters choosing shuttle buses is significantly increased. The benefits of these optimization measures attract more passengers to choose shuttle buses, increase the operating income of bus companies, and thus enable bus companies to escape the vicious cycle of declining passenger flow. At the same time, based on the time-varying passenger flow and travel time between stations, considering the changes in passenger flow distribution caused by changes in passenger travel behavior, studying passenger flow demand and competitive relationships, analyzing the changing trend of the attractiveness of shuttle bus lines to passengers, and then proposing a station optimization plan for shuttle buses on co-linear segments, in order to reduce the costs of passengers and bus companies, improve the service level of shuttle bus lines, and better achieve coordination and cooperation with rail transit modes.

[0079] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0080] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the embodiments of the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for optimizing the common line segments of shuttle buses based on passenger selection behavior, characterized in that: The method comprises: Based on the waiting time cost, the travel time cost and the bus company's operating cost, the generalized travel cost is determined; According to the generalized travel cost and through the preset utility maximization theory, a regression analysis is performed on the factors affecting the passenger's travel selection behavior to determine the passenger's connection bus line selection probability; wherein the connection bus line selection probability includes: the passenger's choice of travel plan probability and the transfer station probability; The connecting bus is subjected to model condition constraints under the relevant cost objective function to obtain a set of optimization constraint conditions for the connecting bus co-line segment stations; wherein the set of optimization constraint conditions for the connecting bus co-line segment stations includes: station stop constraints, competition and cooperation relationship constraints, departure interval constraints, load rate constraints and departure number constraints; Based on the optimization constraint set of the connecting bus co-line segment stations and the connecting bus line selection probability, constructing a connecting bus co-line segment station optimization model; The preset NSGA-II algorithm is used to output and calculate the optimization model of the common line segment station of the shuttle bus, and obtain the optimization plan of the common line segment station of the shuttle bus and the departure interval.

2. The method for optimizing the common line segments of shuttle buses based on passenger selection behavior according to claim 1, characterized in that: Based on the waiting time cost, the travel time cost and the bus company's operating cost, the generalized travel cost is determined, including: according to , get the waiting time cost ;in, It represents the average waiting time for passengers and is half of the bus line's departure interval; Representative train number On the shuttle bus The number of passengers waiting at the station; F The departure interval of the connecting bus; is the passenger waiting cost per unit time; according to , get the transit time cost ;in, for Stand Actual distance between stations; is the speed of the bus; is the number of transfers; is the average vehicle stop time; is the average transfer time; according to , get the operating cost of the bus company ;in, is the number of vehicles dispatched per day; The cost required to operate a single trip on a single route; The waiting time cost, the en route time cost and the bus company operating cost are combined to obtain the generalized travel cost.

3. The method for optimizing the common line segments of shuttle buses based on passenger selection behavior according to claim 2, characterized in that: according to , get bus routes Number of flights per day ;in, For bus routes Daily operating hours; For bus routes Half of the departure interval; according to , get the cost required to operate a single trip on a single route ;in, is the energy consumption cost per unit mileage; is the energy consumption cost of parking per unit time; For line The number of shifts operated per day; is the actual distance between two adjacent stations; i For the station; is the average vehicle stop time.

4. The method for optimizing the common line segments of shuttle buses based on passenger selection behavior according to claim 1, characterized in that: According to the generalized travel cost and through the preset utility maximization theory, a regression analysis is conducted on the factors affecting the passenger's travel choice behavior to determine the passenger's connection bus line selection probability, specifically including: Based on the influencing factors of each cost in the generalized travel cost, determining the influencing factors of the travel plan selection behavior and the influencing factors of the transfer station selection behavior; wherein the influencing factors of the travel plan selection behavior include: waiting time factor, the congestion level of the optional vehicles and the estimated travel time factor; the influencing factors of the transfer station selection behavior include: transfer waiting time factor, the walking distance factor required for transfer and the congestion level of the current vehicle; according to , get the probability of the passenger choosing the ride plan ;in, Select a vehicle on behalf of the passenger possibility; represents a set of optional vehicles for passengers, and the set of optional vehicles for passengers also represents the two vehicles that the passenger will choose between that arrive first and meet his / her travel needs; is the waiting vehicle time factor; is the optional vehicle congestion level; The estimated ride time factor; , , , They are all parameters to be estimated in the probability of the passenger choosing a ride plan; Refers to the line and train number vehicles; according to , get the transfer station probability ;in, Indicates the passenger chooses a transfer station probability; It is a collection of transfer stations; , , , are all parameters to be estimated for the transfer station probability; is the transfer waiting time factor; The walking distance factor required for the transfer; The current crowdedness of the vehicle; By using a preset binary logit regression analysis algorithm, the generalized travel cost is regressed and fitted with the probability of the passenger selecting a travel plan and the probability of the transfer station to determine the probability of selecting the connecting bus line.

5. The method for optimizing the common line segments of shuttle buses based on passenger selection behavior according to claim 1, characterized in that: Before subjecting the shuttle bus to model condition constraints under the relevant cost objective function to obtain a set of optimization constraint conditions for the shuttle bus co-line segment stations, the method further includes: according to , and obtain the cooperation coefficient ;in, For bus routes The number of sites owned; For bus routes The number of sites owned; For bus routes With bus lines The number of public sites; according to , get bus routes l Passenger flow attraction ;in, Select a route for shared-line stations passenger flow; is the total passenger flow between co-line stations; according to , and obtain the correction factor ;in, Respectively represent the passenger flow attraction based on Bus routes Passenger flow attraction and bus routes passenger flow attraction; according to , and obtain the competition intensity of the connecting bus co-line section ;in, is the competition coefficient.

6. The method for optimizing the common line segments of shuttle buses based on passenger selection behavior according to claim 1, characterized in that: Before subjecting the shuttle bus to model condition constraints under the relevant cost objective function to obtain a set of optimization constraint conditions for the shuttle bus co-line segment stations, the method further includes: according to , we get the objective function of minimizing passenger travel cost: ;in, The interval between departures; is the passenger waiting cost per unit time; Represents the average waiting time of passengers; From the transfer point To the transfer point Penalty function for too many transfers; It is a crowding function and is used to reflect the passengers’ comfort level when riding. For transfer point To the transfer point The maximum number of passengers that can be carried by the shuttle buses; is the number of connecting bus lines within the common line segment; for Stand Actual distance between stations; is the speed of the bus; for Stand Number of transfers between stations; is the average vehicle stop time; is the average transfer time; Representative at the transfer point The probability of passengers choosing the ride plan; Representatives at the transfer point The probability of a transfer station; according to , we get the objective function of minimizing the bus operation cost ;in, The cost required to operate a single trip on a single route; is the number of vehicles dispatched per day; Wherein, the cost objective function includes: the passenger travel cost minimization objective function and the public transportation operation cost minimization objective function.

7. The method for optimizing the common line segments of shuttle buses based on passenger selection behavior according to claim 5, characterized in that: The connecting bus is subjected to the model condition constraints under the relevant cost objective function, and the optimization constraint set of the connecting bus co-line segment station is obtained, which includes: according to , and obtain the degree of competition and cooperation used for the competition and cooperation constraint ;in, is the number of connecting bus lines within the common line segment; The competition intensity of the shared line section for connecting buses; according to , get the departure interval of the connecting bus line for the departure interval constraint ;in, It refers to the minimum departure interval of connecting bus routes; It refers to the maximum departure interval of the connecting bus line; according to , the load rate constraint is obtained; wherein, The number of seats on the shuttle bus; is the peak hour; q For the The cross-sectional passenger flow of each station; is the minimum load factor; is the maximum load factor; according to , get the departure train constraint; where, Refers to the minimum number of vehicles required for the feeder bus routes based on the existing operation arrangements; is the total number of vehicles available on the feeder bus lines; The station stop constraint is formed based on the condition that the connecting bus common line segment station includes at least one bus route and the rail and bus public transfer station does not skip a station; The station stop constraint, the competitive relationship constraint, the departure interval constraint, the load factor constraint and the departure number constraint are processed as a constraint set to obtain the optimization constraint set of the connecting bus common line segment station.

8. The method for optimizing the common line segments of shuttle buses based on passenger selection behavior according to claim 1, characterized in that: Based on the optimization constraint set of the connecting bus co-line segment station and the connecting bus line selection probability, a connecting bus co-line segment station optimization model is constructed, which specifically includes: The selection probability of the connecting bus route and the generalized travel cost are used as inputs of the optimization model, and the optimization constraint set of the connecting bus co-line segment stations is used as the constraint of the optimization model, and the passenger travel cost minimization objective function and the bus operation cost minimization objective function are both used as optimization objectives of the optimization model; Based on the input quantity, constraint quantity and optimization target of the optimization model, an optimization model of the connecting bus co-line segment station is constructed.

9. The method for optimizing the common line segments of shuttle buses based on passenger selection behavior according to claim 1, characterized in that: The preset NSGA-II algorithm is used to output and calculate the optimization model of the shuttle bus co-line segment station, and obtain the shuttle bus co-line segment station optimization plan and departure interval, which specifically includes: By using a preset non-dominated sorting genetic algorithm, the population parameters of the shuttle bus co-linear segment station optimization model are calculated based on the initial population chromosome encoding under the station stop plan and departure frequency to obtain an initial solution set; According to the objective function of the connecting bus co-line segment station optimization model, the initial solution set is evaluated for adaptability to obtain an adaptability value; Based on the evaluated adaptability value, the optimal solution of the shuttle bus co-line segment station optimization model is iteratively calculated to generate a new generation of population; and the new generation of population is output processed to obtain the shuttle bus co-line segment station optimization plan and the departure interval; wherein the iterative optimal solution calculation includes: selection mechanism calculation, recombination strategy calculation and mutation operation calculation.

10. A device for optimizing the common line segment of shuttle buses based on passenger selection behavior, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, so that the at least one processor can execute the method for optimizing the common line segments of connecting buses based on passenger selection behavior according to any one of claims 1-9.

Images