Method for Estimating Urban Rail Transit In-and-Out Station and Transfer Times by Fusing Card-Swiping Data and Behaviors

Abstract

The present invention provides a method for estimating the in-out station and transfer times of urban rail transit by integrating card-swipe data and behaviors. For any pair of origin and destination stations, based on the tolerance limit of passengers for travel time, an effective path set is generated, and the selection probability of each effective path is calculated using a path selection model. The state transition function is updated based on the parameter values of the last iteration, and the parameter values of the current iteration are sampled using the Gaussian random walk Metropolis-Hastings sampling method. Based on the path selection probability, the parameter values sampled this time, and the travel time observation values, the likelihood function value is updated, the acceptance rate is updated using the likelihood function value, and it is determined whether to terminate the iteration according to the acceptance rate result and the current iteration round. The parameter values recorded when the iteration is terminated are used as the final estimated values of the in-out station time distribution and transfer time distribution parameters of each urban rail transit station. This method estimates the in-out station time distribution of each urban rail transit station and the transfer time distribution of each transfer station by integrating card-swipe data and behavior analysis, which not only ensures the estimation accuracy but also avoids the difficulties of data collection.

Description

Technical Field

[0001] The present invention relates to the technical field of urban rail transit operation management, and particularly to a method for estimating the entry and exit times and transfer times of urban rail transit by integrating card - swiping data and behaviors. Background Art

[0002] The entry and exit times of each station and the transfer times of each transfer station of urban rail transit (hereinafter referred to as urban rail) are important indicators for evaluating the service levels of urban rail transit systems, their stations, and transfer stations. They are also important components of the travel time of urban rail transit, providing important data support for calculating the happiness of urban rail transit travel, exploring passengers' travel choice behaviors, estimating the distribution of travel demands, formulating transportation plans, and providing travel navigation services.

[0003] The entry time refers to the time from swiping the card to enter the station to boarding the train; the exit time refers to the time from getting off the train to swiping the card to exit the station; the transfer time refers to the time from getting off one line to boarding another line.

[0004] Due to the rapid development of the urban rail transit network, the number of stations and transfer relationships is increasing. Especially in megacities such as Beijing, Shanghai, Guangzhou, and Shenzhen, the urban rail transit systems have already formed a networked, large - scale, and complex pattern. The number of ordinary stations and transfer stations is huge. Coupled with the relatively closed nature of the urban rail transit system, as well as the random volatility of the entry and exit times and transfer times, it poses great challenges to the traditional estimation of entry and exit times and transfer times.

[0005] Currently, the calculation methods for the entry and exit times of each station and the transfer times of each transfer station in the prior art of urban rail transit include:

[0006] 1) The model estimation method, that is, the average entry - exit or transfer walking time is obtained by dividing the average distance from the ticket gate to the platform by the average walking speed. However, this method has a large error and cannot reflect the randomness of the walking time.

[0007] 2) The sampling survey method, that is, by following passengers on the spot and recording the entry - exit or transfer times of the surveyed objects to estimate the entry - exit times and transfer times. However, this method consumes a large amount of manpower, material resources, and financial resources.

[0008] The card - swiping data (including mobile phone QR - code scanning data) is the data collected by the existing automatic fare collection system for ticket revenue clearing in urban rail transit, which accurately records information such as the boarding station name, boarding time, alighting station name, alighting time of each passenger every day. The massive card - swiping data contains rich passenger travel information, which makes it possible to estimate the boarding and alighting times of each station and the transfer times of each transfer relationship based on the card - swiping data, and can greatly save manpower, material resources and financial resources. However, due to the networked nature of urban rail transit, there is more than one effective path between many OD (Origin - Destination) pairs, and it is impossible to directly identify which path a certain card - swiping data comes from in the card - swiping data. Based on behavior analysis, the path - selection ratio of passengers can be obtained, which can solve the problem that the card - swiping data cannot directly match the effective path. Summary of the Invention

[0009] An embodiment of the present invention provides a method for estimating boarding, alighting and transfer times in urban rail transit by integrating card - swiping data, so as to effectively improve the management efficiency of urban rail transit.

[0010] To achieve the above object, the present invention adopts the following technical solutions.

[0011] A method for estimating boarding, alighting and transfer times in urban rail transit by integrating card - swiping data and behavior includes:

[0012] Initialize the urban rail transit network, each station parameter and algorithm iteration setting, and extract travel - time observation values based on the card - swiping data. The station parameters include the mean and standard deviation of the normal distribution of the boarding and alighting times of each station, and the mean and standard deviation of the normal distribution of the transfer times of each transfer station in different transfer directions;

[0013] For any OD pair, generate an effective path set based on the tolerance threshold of the travel time of each path relative to the travel time of the shortest path;

[0014] Construct a path - selection model for urban rail transit passengers based on the random utility theory, and use the path - selection model to calculate the selection probabilities of each effective path between each OD pair;

[0015] Update the state - transition function according to the station parameter values of the previous iteration, and randomly sample the station parameter values of this iteration based on the state - transition function using Gaussian random walk M - H sampling;

[0016] Update the likelihood - function value according to the selection probabilities of each effective path, the sampled values of the station parameters and the travel - time observation values, update the acceptance rate using the updated likelihood - function value, determine whether to terminate the iteration according to the result of the acceptance rate and the current iteration round, and use the station parameter values recorded when the iteration is terminated as the estimated values of the final boarding, alighting and transfer - time distribution parameters.

[0017] Preferably, the initialization of the urban rail network, parameters of each station and algorithm are iteratively set, and travel time observations are extracted based on card swiping data. The station parameters include the mean and standard deviation of the normal distribution of entry and exit times of each station, and the mean and standard deviation of the normal distribution of transfer times in different transfer directions at each transfer station, including:

[0018] Load the physical topology and train schedule, initialize the station parameters to be estimated, and form a station parameter combination Where G is the number of parameters. The station parameters include the mean and standard deviation of the normal distribution of the entry and exit times of each station, and the mean and standard deviation of the normal distribution of the transfer times in different transfer directions at each transfer station. At the same time, for any OD pair, a travel time set T is generated based on the card swiping data. The total number of iterations of the method is set to M, which serves as the termination condition of the iterative operation.

[0019] Preferably, for any OD pair, generating a valid path set based on a tolerance threshold of the travel time of each path relative to the travel time of the shortest path includes:

[0020] For any OD pair s, the running time between adjacent stations and the stop time of each station are calculated according to the train schedule as the components of the travel time. The K-shortest path algorithm is used to generate a set of feasible paths for each OD pair based on the travel time of the path. The difference between the travel time of each path and the travel time of the shortest path is calculated. If the difference is less than the passenger's tolerance threshold Y s =α·ln(t s +1), then the corresponding path is a valid path, and all valid paths constitute the valid path set K s , where t s is the shortest travel time under OD pair s, and α is a constant.

[0021] Preferably, the method of constructing a route selection model for urban rail transit passengers based on random utility theory and calculating the selection probability of each valid route between each OD pair using the route selection model includes:

[0022] Based on the random utility theory, a path selection model for urban rail transit passengers is constructed to calculate the selection probability of each effective path between each OD pair and the selection probability of the effective path k between OD pairs s. for:

[0023]

[0024] Where, is the influencing factor of OD on path k between s, and β is the corresponding station parameter;

[0025] The selection probabilities of all valid paths constitute the set P.

[0026] Preferably, update the state transition function according to the station parameter values of the previous iteration, and randomly extract the station parameter values of this iteration by using Gaussian random walk M-H sampling based on the state transition function, including:

[0027] Extract the i-th station parameter obtained from the (m-1)-th iteration

[0028] Assume that the state transition function of the station parameter follows a normal distribution, and use the station parameter Update the state transition function That is where ξ 2 is the proposed variance of the i-th station parameter;

[0029] Based on the updated state transition function, use Gaussian random walk M-H sampling to randomly extract the pending sample values The extracted sample values Together with the first (i-1) station parameters of the m-th iteration and the last (G-i) station parameters of the (m-1)-th iteration, jointly form the pending station parameter combination

[0030] Preferably, update the likelihood function value according to the selection probability of each effective path, the sampled value of the station parameter and the observed value of the travel time, update the acceptance rate by using the updated likelihood function value, and determine whether to terminate the iteration according to the result of the acceptance rate and the current iteration round, and use the station parameter value recorded when terminating the iteration as the estimated value of the final in-out and transfer time distribution parameters, including:

[0031] Construct a likelihood function by combining the path selection probability and the path travel time distribution, and update the likelihood function value according to the station parameter value and the observed value of the travel time:

[0032]

[0033] In the formula, L is the likelihood function, and ψ is the probability density function;

[0034] Based on the station parameter Station parameter And its prior distribution And And the travel time set T and the likelihood function value, update the acceptance rate κ:

[0035]

[0036] The prior distributions of the means of the normal distribution station parameters for the inbound time, outbound time, and transfer time are respectively set as U(1, 30), U(1, 20), U(1, 20), and the prior distributions of the standard deviations are respectively set as U(0, 20), U(0, 15), U(0, 20);

[0037] Random values u are drawn from the uniform distribution U(0, 1). If the value is not greater than the acceptance rate u ≤ κ, the set of station parameters for the m-th iteration is the set of station parameters formed by sampling in step S4, that is, ζ (m) = ζ * ; Otherwise, the set of station parameters for the m-th iteration is the same as the set of station parameters generated after the (m - 1)-th iteration, that is, ζ (m) = ζ (m-1) , if i = G, then set i = 1 and enter the subsequent iteration termination condition judgment process; otherwise, set i = i + 1, transfer to the station parameter sampling process, and continue the subsequent station parameter iterative update process;

[0038] Iteration termination condition judgment:

[0039] If the number of iterations has not reached the total number of iterations M, that is, m <= M, then the iteration mark m = m + 1, transfer to the station parameter sampling process and the station parameter iterative update process; if m > M, stop sampling, and take the average value of the recorded station parameter values at this time as the estimated value of the final inbound / outbound and transfer time distribution parameters.

[0040] As can be seen from the technical solutions provided in the embodiments of the present invention above, the embodiments of the present invention realize the estimation of the inbound / outbound time distribution of each station and the transfer time distribution of each transfer station in urban rail transit through efficient, accurate, and intelligent data mining technology by integrating swipe card data and behavior analysis. This method can estimate the inbound / outbound time distribution of each station and the transfer time distribution of each transfer station in the whole network of urban rail transit by integrating swipe card data and behavior analysis, which not only ensures the estimation accuracy but also avoids the difficulty of data collection.

[0041] Additional aspects and advantages of the present invention will be given in part in the following description, and these will become obvious from the following description or be understood through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0043] Figure 1This is a processing flowchart of a method for estimating the entry / exit and transfer times of urban rail transit by integrating card - swiping data and behaviors provided by an embodiment of the present invention. Detailed implementation manners

[0044] The following details the implementation manners of the present invention. Examples of the implementation manners are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements with the same or similar functions throughout. The implementation manners described below by referring to the accompanying drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation of the present invention.

[0045] Those skilled in the art of the present technology can understand that, unless specifically stated, the singular forms "a", "an", "the", and "said" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present invention means the presence of the described features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or their groups. It should be understood that when we say an element is "connected" or "coupled" to another element, it can be directly connected or coupled to other elements, or there may also be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or coupling. The phrase "and / or" used herein includes any unit and all combinations of one or more of the associated listed items.

[0046] Those skilled in the art of the present technology can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as the general understanding of those of ordinary skill in the technical field to which the present invention belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art and will not be interpreted with an idealized or overly formal meaning unless defined as such here.

[0047] For the convenience of understanding the embodiments of the present invention, the following will further explain with several specific embodiments by combining with the accompanying drawings, and each embodiment does not constitute a limitation to the embodiments of the present invention.

[0048] The processing flow of a method for estimating the entry / exit and transfer times of urban rail transit by integrating card - swiping data and behaviors provided by an embodiment of the present invention is as Figure 1 shown and includes the following processing steps:

[0049] Step S1: Initialization.

[0050] (1) Load basic data such as the physical topology structure and train timetables.

[0051] (2) Initialize the parameters of the stations to be estimated and form a combination of station parameters Among them, G is the number of station parameters, including the mean and standard deviation of the normal distribution of the entry and exit times of each station, and the mean and standard deviation of the normal distribution of the transfer times at different transfer directions of each transfer station. i is the label of the station parameter. For the entry time, exit time, and transfer time, the initial values of the mean and standard deviation are both set to 1 minute.

[0052] It is assumed that the entry and exit times and transfer times all follow independent and uncorrelated normal distributions, and the riding time is a definite value.

[0053] (3) Set the total number of iterations of this method to M = 10000 as the termination condition of the iterative operation, set the iteration label m = 1, and set the number of iterations B = 5000 in the trial calculation stage.

[0054] (4) For any OD (Origin-Destination, starting and ending stations) pair s, randomly select H = 100 pieces of data from the card swiping data of one month to form the travel time set T of the OD pair s s . The travel times of all OD pairs constitute the set T.

[0055] Step S2: Generate the set of valid paths.

[0056] (1) According to the train timetable, calculate the running time between adjacent stations and the station stop time of each station as the components of the riding time.

[0057] (2) Use the K-shortest path algorithm (K = 10) to generate the path set of each OD pair based on the riding time of the path.

[0058] (3) For any OD pair s, calculate the difference between the riding time of each path and the shortest riding time. If this difference is less than the tolerance threshold Y of the passenger s = α·ln(t s +1), then the corresponding path is a valid path, and finally form the set of valid paths K s , where, t s is the shortest riding time under the OD pair s, with the unit of h. Through investigation, the station parameter α = 5.04.

[0059] The above tolerance threshold of the passenger can be estimated in advance through the investigation data.

[0060] Step S3: Calculate the path selection probability:

[0061] Construct a path selection model for urban rail transit passengers based on the random utility theory. This path selection model can be calibrated in advance based on the investigation data.

[0062] Taking the Logit model as an example, calculate the selection proportion of each valid path between each OD pair, that is, the selection probability of path k between OD pairs s is:

[0063]

[0064] Wherein, is the influencing factor of path k between OD pair s, including riding time, transfer times, angle cost, comfort level, etc.; β is the corresponding station parameter.

[0065] Table 1 Calculation Explanation of Influencing Factors and Station Parameter Table

[0066]

[0067]

[0068] The selection probabilities of all valid paths form a set P.

[0069] Step S4: Sampling of station parameters.

[0070] (1) Extract the i-th station parameter obtained in the (m - 1)-th iteration

[0071] (2) Assume that the state transition function of the station parameter follows a normal distribution, and use the station parameter to update the state transition function That is where ξ 2 is the proposed variance of the i-th station parameter. For the inbound time, outbound time, and transfer time, the variances are proposed to be 22 min 2 , 15 min 2 , 15 min 2 .

[0072] (3) Based on the updated state transition function, use Gaussian random walk M-H sampling to randomly extract the pending sample values

[0073] (4) Combine the extracted sample values with the first (i - 1) station parameters of the m-th iteration and the last (G - i) station parameters of the (m - 1)-th iteration to jointly form the set of station parameters to be estimated

[0074] Step S5: Calculation of likelihood function value:

[0075] Combine the path selection probability and the path travel time distribution to construct a likelihood function, and update the likelihood function value according to the station parameter value and the travel time observation value:

[0076]

[0077] Wherein, L is the likelihood function, and ψ is the probability density function

[0078] Step S6, Station parameter update:

[0079] (1) Update the acceptance rate. Based on the station parameters Station parameters and their prior distributions and and the travel time set T and the likelihood function value, update the acceptance rate κ:

[0080]

[0081] Here, the prior distributions of each station parameter can be set to uniform distributions. For the normal distribution station parameters of the inbound time, outbound time, and transfer time, the prior distributions of their means are respectively set to U(1, 30), U(1, 20), U(1, 20), and the prior distributions of the standard deviations are respectively set to U(0, 20), U(0, 15), U(0, 20).

[0082] (2) Update the station parameters. Draw a random value u from the uniform distribution U(0, 1). If the value is not greater than the acceptance rate u ≤ κ, then the set of station parameters for the m-th iteration is the set of station parameters formed by sampling in Step S4, that is, ζ[[ID=E22]] (m) = ζ[[ID=E24]] * ;

[0083] Otherwise, the set of station parameters for the m-th iteration is the same as the set of station parameters generated after the (m - 1)-th iteration, that is, ζ[[ID=E29]] (m) = ζ[[ID=E31]] (m-1) . If i = G, then set i = 1 and enter Step S7. Otherwise, set i = i + 1 and transfer to Step S4.

[0084] Step S7, Termination condition judgment:

[0085] If the number of iterations does not reach the predetermined number of iterations in the trial calculation stage, that is, m <= B, then the iteration mark m = m + 1, and repeat Steps S4 - Step S6; if m > B, then record the station parameter values of this iteration, the iteration mark m = m + 1, and repeat Steps S4 - Step S6; if the number of iterations reaches the predetermined total number of iterations, that is, m > M, then stop sampling, and take the average of the recorded station parameter values to obtain the final estimated value of the station parameters.

[0086] In summary, the method of the embodiment of the present invention utilizes the characteristics of accurately recording the station names of each passenger's entry and exit and the accurate full - journey travel time with the card - swiping data, as well as the advantage of the full - sample of the card - swiping data. At the same time, using behavior analysis, the advantage of obtaining the selection probabilities of multiple effective paths between OD pairs under networked conditions can be obtained. Considering the randomness of the entry and exit times and transfer times, based on Bayesian inference and Markov chain Monte Carlo method, a set of methods for estimating the entry, exit, and transfer time distributions of each station in the urban rail transit network is developed.

[0087] A large amount of card - swiping data can, to a certain extent, correct the deficiencies of prior knowledge and improve the sampling quality; behavior analysis can comprehensively consider the influence of various factors on passengers' path - selection preferences. Through the integration of card - swiping data and behavior analysis, it is possible to conveniently estimate the inbound and outbound time distributions of each station and the transfer - time distributions in each transfer direction at each transfer station, without the need for a large amount of manpower, material resources, and financial resources to investigate the time distributions at each location. This not only saves costs but also makes full use of existing data resources.

[0088] Those of ordinary skill in the art can understand that the attached drawings are only schematic diagrams of an embodiment, and the modules or processes in the attached drawings are not necessarily essential for implementing the present invention.

[0089] From the description of the above - mentioned embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general - purpose hardware platform. Based on this understanding, the technical solution of the present invention, in essence or the part that makes a contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0090] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiments, since they are basically similar to the method embodiments, they are described relatively simply. For the relevant parts, refer to the partial description of the method embodiments. The device and system embodiments described above are only illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

[0091] As described above, the above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for estimating urban rail entry and exit and transfer time by integrating card swiping data and behavior, characterized by: include: Initialize the urban rail network, station parameters, and algorithm iteration settings, and extract travel time observations based on card swipe data. The station parameters include the mean and standard deviation of the normal distribution of entry and exit times at each station, and the mean and standard deviation of the normal distribution of transfer times in different transfer directions at each transfer station. For any origin-destination OD pair, a valid path set is generated based on the tolerance threshold of each path's travel time relative to the shortest path's travel time. Based on random utility theory, a route selection model for urban rail transit passengers is constructed, and the selection probability of each effective route between each OD pair is calculated using the route selection model. Update the state transfer function according to the station parameter values of the previous iteration, and randomly extract the station parameter values of this iteration using Gaussian random walk MH sampling based on the state transfer function; The likelihood function value is updated based on the selection probability of each valid path, the sampled values of station parameters, and the observed values of travel time. The acceptance rate is updated using the updated likelihood function value. The termination of the iteration is determined based on the acceptance rate result and the current iteration round. The station parameter values recorded when the iteration is terminated are used as the final estimates of the entry, exit, and transfer time distribution parameters. For any OD pair, generating a valid path set based on the tolerance threshold of each path's travel time relative to the shortest path's travel time includes: For any OD pair s, the running time between adjacent stations and the stop time of each station are calculated according to the train schedule as the components of the travel time. The K-shortest path algorithm is used to generate a set of feasible paths for each OD pair based on the travel time of the path. The difference between the travel time of each path and the travel time of the shortest path is calculated. If the difference is less than the passenger's tolerance threshold Y s =α·ln(t s +1), then the corresponding path is a valid path, and all valid paths constitute the valid path set K s , where t s is the shortest ride time under OD pair s, α is a constant; The route selection model for urban rail transit passengers based on random utility theory is constructed, and the selection probability of each valid route between each OD pair is calculated using the route selection model, including: Based on the random utility theory, a path selection model for urban rail transit passengers is constructed to calculate the selection probability of each effective path between each OD pair and the selection probability of the effective path k between OD pairs s. for: Where, is the influencing factor of OD on path k between s, and β is the corresponding station parameter; The selection probabilities of all valid paths constitute the set P.

2. The method according to claim 1, characterized in that The initialization of the urban rail network, the parameters of each station and the algorithm are iteratively set, and the travel time observation values are extracted based on the card swiping data. The station parameters include the mean and standard deviation of the normal distribution of the entry and exit times of each station, and the mean and standard deviation of the normal distribution of the transfer time in different transfer directions at each transfer station, including: Load the physical topology and train schedule, initialize the station parameters to be estimated, and form a station parameter combination Where G is the number of parameters. The station parameters include the mean and standard deviation of the normal distribution of the entry and exit times of each station, and the mean and standard deviation of the normal distribution of the transfer times in different transfer directions at each transfer station. At the same time, for any OD pair, a travel time set T is generated based on the card swiping data. The total number of iterations of the method is set to M, which serves as the termination condition of the iterative operation.

3. The method according to claim 2, characterized in that The updating of the state transfer function according to the station parameter value of the previous iteration and the random sampling of the station parameter value of this iteration using Gaussian random walk MH sampling based on the state transfer function include: Extract the parameters of the i-th station obtained in the m-1th iteration Assuming that the state transfer function of the station parameters obeys the normal distribution, the station parameters Update state transition function Right now where ξ 2 is the suggested variance of the parameters of the ith station; Based on the updated state transfer function, Gaussian random walk MH sampling is used to randomly extract the pending sample value The sample value to be drawn Together with the first i-1 station parameters of the m-th iteration and the last Gi station parameters of the m-1-th iteration, they form the station parameter combination to be estimated 4. The method according to claim 3, characterized in that The method of updating the likelihood function value based on the selection probability of each valid path, the sampled station parameter values, and the observed travel time values, updating the acceptance rate using the updated likelihood function value, determining whether to terminate the iteration based on the acceptance rate result and the current iteration round, and using the station parameter values recorded when the iteration is terminated as the final estimated values of the entry, exit, and transfer time distribution parameters includes: The likelihood function is constructed by combining the path selection probability and the path travel time distribution, and the likelihood function value is updated according to the station parameter value and the travel time observation value: Where L is the likelihood function and ψ is the probability density function; Based on station parameters Station parameters and its prior distribution and As well as the travel time set T and the likelihood function value, update the acceptance rate κ: The prior distributions of the means of the normally distributed station parameters for arrival time, departure time, and transfer time are set to U(1,30), U(1,20), and U(1,20), respectively, and the prior distributions of the standard deviations are set to U(0,20), U(0,15), and U(0,20), respectively; A random value u is drawn from the uniform distribution U(0,1). If the value is not greater than the acceptance rate u≤κ, the station parameter set of the mth iteration is the station parameter set constructed by sampling in step S4, that is, ζ (m) =ζ * Otherwise, the station parameter set of the mth iteration is the same as the mth iteration. - The station parameter set generated after 1 iteration is ζ (m) =ζ * Otherwise (m) =ζ (m-1) If i=G, then set i=1 and enter the subsequent iterative termination condition judgment process; otherwise, set i=i+1 and enter the station parameter sampling process, and continue the subsequent station parameter iterative update process; Iteration termination condition judgment: If the number of iterations does not reach the total number of iterations M, that is, m<=M, the iteration mark m=m+1, and the station parameter sampling process and the station parameter iterative update process are entered; if m>M, the sampling is stopped, and the average value of the station parameter values recorded at this time is taken as the final estimated value of the entry and exit and transfer time distribution parameters.

Images