Rail transit ticket business clearing method and system based on dynamic path selection

Abstract

The invention provides a rail transit ticket business clearing method and system based on dynamic path selection. The method comprises the following steps: screening effective paths of target starting and ending stations to form an effective path set; calculating the path selection probability of the effective path; in combination with the path selection probability, the proportion of the key influence factors and the weight coefficient of the key influence factors dynamically adjusted by a multiple linear regression model, calculating the distribution proportion of the operation main body in each path interval in the effective path; and performing aggregation calculation on the distribution proportion of each path interval of each effective path in the effective path set to obtain the clearing proportion of the operation main body in the ticket revenue of the target start-stop station. And multi-dimensional splitting in the section is carried out to final aggregation of the operation main bodies, so that reasonable clearing of tickets among the operation main bodies is realized, and legitimate rights and interests of all parties can be guaranteed.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, and in particular to a rail transit ticketing clearing method and system based on dynamic route selection. Background Technology

[0002] The urban rail transit ticketing clearing system utilizes computer and information technology to solve the problems of ticket revenue and ticketing clearing management in rail transit operations. It aims to provide services for rail transit operators to efficiently manage ticket revenue, rationally allocate network resources, and manage the operation system.

[0003] With the emergence of metropolitan areas, the transportation networks in some regions are accelerating towards multi-network integration, encompassing various systems such as trunk railways, intercity railways, suburban railways, and urban rail transit.

[0004] Currently, ticketing clearing and allocation mechanisms are mostly designed based on a single operating entity and a single fare system, making it difficult to adapt to the diversification of passenger travel routes and the complex interactions between multiple operating entities in the context of multi-network integration. Differences in fiscal subsidies, operating costs, resident income, and pricing models across different regions of metropolitan areas lead to inconsistent rail transit fare systems, resulting in "different prices for the same line." The lack of a unified standard for splitting fares between networks, coupled with disagreements among operating entities due to differences in line attributes, service areas, and cost structures, causes ticketing clearing rules to lag behind the pace of network development. The complexity brought about by multi-network integration makes the ticketing clearing problem between different operating entities increasingly prominent, urgently requiring the establishment of a scientific, fair, transparent, and practical ticketing clearing mechanism to adapt to the complexity of multi-network integration, ensure reasonable allocation for all parties, and promote high-quality development of regional transportation integration. Summary of the Invention

[0005] Based on the above, this invention provides a method and system for clearing tickets for rail transit based on dynamic path selection, aiming to solve the technical problem that ticket clearing under complex operating entities needs to be optimized in the prior art.

[0006] This invention provides a method for clearing and distributing rail transit tickets based on dynamic route selection, comprising:

[0007] Step A1: Construct a rail transit network with identification of the operating entities responsible for each line segment;

[0008] Step A2: Select valid routes from the target origin and destination stations in the rail transit network to form a set of valid routes;

[0009] Step A3: Calculate the path selection probability of the effective paths in the effective path set;

[0010] Step A4: Combine the route selection probability, the proportion of key influencing factors, and the weight coefficients of key influencing factors dynamically adjusted by the multiple linear regression model to calculate the allocation ratio of the operating entity in each route interval in the effective route. Key influencing factors include travel mileage, operating costs, and passenger flow.

[0011] Step A5: Aggregate and calculate the allocation ratio of each path interval in each valid path in the valid path set to obtain the clearing ratio of ticket revenue of the operating entity in the target origin and destination stations.

[0012] Among them, the route interval refers to the line segment formed by stations that are visited continuously without transfers. The multiple linear regression model uses historical ticket revenue as the dependent variable and the factors affecting ticket revenue as independent variables.

[0013] Furthermore, step A2 includes:

[0014] Step A21: Summarize the running paths of the target origin and destination stations to obtain the running path set;

[0015] Step A22: Calculate the path cost of each running path in the running path set;

[0016] Step A23: Based on path cost, filter the running paths from the set of running paths to form a preliminary set of paths;

[0017] Step A24: Calculate the initial probability of path selection for each running path in the preliminary path selection set;

[0018] Step A25: Based on the initial probability of path selection, select effective paths from the preliminary path selection set to form an effective path set;

[0019] In step A3, the path selection probability of the effective path is calculated based on the initial path selection probability and the path cost.

[0020] Furthermore, in step A22, the formula for calculating the path cost is as follows:

[0021]

[0022] Among them, R p The path cost refers to the cost of the execution path;

[0023] m represents the total number of segments in a path divided according to transfer nodes;

[0024] C i This represents the attraction factor of path interval i in the running path;

[0025] T i This represents the running time of path interval i within the running path;

[0026] E xj This represents the transfer factor at the j-th transfer, which increases with the number of transfers.

[0027] T xj This represents the total walking and waiting time during the j-th transfer.

[0028] Furthermore, in step A23, the formula for selecting running paths from the set of running paths based on path cost is as follows:

[0029]

[0030] in, This represents the upper critical value of the path cost;

[0031] The minimum path cost in the set of running paths;

[0032] α1 represents the magnification factor of the minimum path cost;

[0033] α2 represents the increment constant of the minimum path cost;

[0034] In step A23, running paths with path costs less than the upper threshold value are selected from the running path set to form a preliminary path selection set.

[0035] Furthermore, in step A24, the formula for calculating the initial probability of path selection for each running path in the preliminary path selection set is as follows:

[0036]

[0037] Where, p k This represents the initial probability of selecting the k-th running path in the initial path selection set;

[0038] This represents the path cost of the k-th running path in the initial path selection set;

[0039] σ represents the standard deviation of the path cost;

[0040] In step A25, running paths with an initial path selection probability lower than a preset probability threshold are filtered out from the initial path selection set to form an effective path set.

[0041] Furthermore, in step A3, the effective path set S is represented as:

[0042] S = {path 1, path 2, path 3, ..., path n}

[0043] The formula for calculating the path selection probability of an effective path is as follows:

[0044]

[0045] in, The probability of choosing the effective path i;

[0046] p i This represents the initial probability of path selection for the effective path i;

[0047] p 舍弃总和 This represents the sum of the initial probabilities of path selection for the running paths filtered out from the initial path selection set. The calculation formula is as follows:

[0048] p discarded total = ∑ k∈ Discard path p k

[0049] ω i The weight of a valid path i in the set of valid paths is represented by the following formula:

[0050]

[0051] This represents the path cost of the valid path i.

[0052] Furthermore, in step A4, the formula for calculating the allocation ratio of the operating entity to the path interval in the effective path is as follows:

[0053]

[0054] Among them, S i,k,m This represents the allocation ratio of operating entity i in the m-th segment of the k-th valid path in the valid path set;

[0055] This represents the path selection probability of the k-th valid path in the set of valid paths;

[0056] L i,k,m This represents the mileage that operator i is responsible for in the m-th segment of the k-th valid path in the valid path set;

[0057] L k,m This represents the total mileage of the m-th segment of the k-th valid path in the valid path set.

[0058] L k,总 This represents the total mileage of the k-th valid path in the set of valid paths;

[0059] C i,k,m This represents the operating cost of operator i in the m-th segment of the k-th valid path in the valid path set;

[0060] C k,mThis represents the total operating cost of the m-th segment of the k-th valid path in the valid path set.

[0061] F i,k,m This represents the passenger flow of operator i in the m-th segment of the k-th valid path in the valid path set;

[0062] F k,m This represents the passenger flow of the m-th segment of the k-th valid path in the valid path set;

[0063] γ1 represents the weighting coefficient for travel distance as a key influencing factor;

[0064] γ2 represents the weighting coefficient for operating costs as a key influencing factor;

[0065] γ3 represents the weighting coefficient for passenger flow as a key influencing factor;

[0066] In step A5, the formula for calculating the percentage of ticket revenue to be distributed by the operating entity to the target origin and destination stations is as follows:

[0067]

[0068] Among them, S i This indicates the percentage of ticket revenue that operating entity i receives from the target origin and destination stations;

[0069] n k Let k represent the number of transfers on the effective path k, and n represent the number of effective paths in the effective path set.

[0070] Furthermore, in step A4, the adjustment process for the weight coefficients of key influencing factors includes:

[0071] Step A41: Construct a multiple linear regression model. The calculation formula for the multiple linear regression model is as follows:

[0072] Y=β1X1+β2X2+β3X3+β4X4+ε

[0073] Where Y represents historical ticket revenue;

[0074] X1 to X4 represent four factors affecting ticket revenue: travel distance, operating costs, passenger flow, and number of transfers, respectively.

[0075] β1 to β4 represent the weighting coefficients of the four factors affecting ticket revenue: travel distance, operating costs, passenger flow, and number of transfers.

[0076] ε represents a constant;

[0077] Step A42: Obtain historical ticket revenue and use gradient descent to iteratively optimize the multiple linear regression model to determine the weight coefficients of factors affecting ticket revenue.

[0078] Step A43: Adjust the weight coefficients of key influencing factors based on the weight coefficients of factors affecting ticket revenue.

[0079] Furthermore, the rail transit network is divided into multiple sub-regions within a designated area;

[0080] In step A5, when the target origin and destination stations cross sub-regions, the calculation process for the full fare between the target origin and destination stations includes:

[0081] Step A51: Determine the corresponding operating entities within each sub-region;

[0082] Step A52: Calculate the fare for the shortest path segment between two stations within a sub-region based on the shortest path algorithm;

[0083] Step A53: Decompose the target origin and destination stations according to the shortest path segments within the sub-regions to form the shortest path segments of the decomposed sub-regions.

[0084] Step A54: Obtain the fare of the shortest path segment in each sub-region and sum them up to obtain the preliminary fare for the target origin and destination stations across regions.

[0085] Step A55: Use dynamic impact factors to correct the initial ticket prices of the target origin and destination stations to obtain the full ticket prices of the target origin and destination stations.

[0086] This invention also provides a rail transit ticketing clearing system based on dynamic route selection, characterized in that, for executing the aforementioned rail transit ticketing clearing method based on dynamic route selection, it includes:

[0087] The network construction module is used to construct a rail transit network that identifies the operating entities responsible for each line segment;

[0088] The effective path acquisition module and the connecting network construction module are used to filter effective paths to target origin and destination stations based on the rail transit network and form a set of effective paths.

[0089] The path probability calculation module, connected to the effective path acquisition module, is used to calculate the path selection probability of effective paths in the effective path set.

[0090] The module is split and connected to the path probability calculation module. It is used to combine the path selection probability, the proportion of key influencing factors, and the weight coefficients of key influencing factors dynamically adjusted by the multiple linear regression model to calculate the allocation ratio of the operating entity in each path interval in the effective path.

[0091] The aggregation module, which connects the splitting module, is used to aggregate and calculate the allocation ratio of each path interval in the effective path set to obtain the clearing ratio of the ticket revenue of the operating entity in the target origin and destination stations.

[0092] Among them, the route interval refers to the line segment formed by stations that are visited continuously without transfers. The multiple linear regression model uses historical ticket revenue as the dependent variable and the factors affecting ticket revenue as independent variables.

[0093] The beneficial technical effects of this invention are as follows: it proposes a scientific and reasonable ticketing clearing mechanism. This not only helps to solve the prominent problems in the current ticketing allocation, but also enables the reasonable distribution of tickets among various operating entities through multi-dimensional segmentation and final aggregation by the operating entities. This protects the legitimate rights and interests of all parties, stimulates the enthusiasm and creativity of the operating entities, helps to improve service quality and operational efficiency, and enhances the level of regional transportation integration. It will also provide useful lessons and references for the future integrated transportation development of similar metropolitan areas. Attached Figure Description

[0094] Figure 1 This is a flowchart illustrating the overall steps of a rail transit ticketing clearing method based on dynamic path selection according to the present invention.

[0095] Figure 2 This is a flowchart illustrating the steps of an effective route selection method for rail transit ticket clearing based on dynamic route selection, as described in this invention.

[0096] Figure 3 This is a flowchart illustrating the steps of adjusting the weight coefficients in a rail transit ticketing clearing method based on dynamic path selection, as described in this invention.

[0097] Figure 4 This is a flowchart illustrating the steps involved in calculating cross-regional fares in a rail transit fare clearing method based on dynamic path selection, as described in this invention.

[0098] Figure 5 This is a schematic diagram of a rail transit ticketing and clearing system based on dynamic path selection according to the present invention. Detailed Implementation

[0099] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0100] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0101] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0102] See Figure 1 This invention provides a method for clearing and distributing rail transit tickets based on dynamic path selection, comprising:

[0103] Step A1: Construct a rail transit network with identification of the operating entities responsible for each line segment;

[0104] Step A2: Select valid routes from the target origin and destination stations in the rail transit network to form a set of valid routes;

[0105] Step A3: Calculate the path selection probability of the effective paths in the effective path set;

[0106] Step A4: Combine the route selection probability, the proportion of key influencing factors, and the weight coefficients of key influencing factors dynamically adjusted by the multiple linear regression model to calculate the allocation ratio of the operating entity in each route interval in the effective route. Key influencing factors include travel mileage, operating costs, and passenger flow.

[0107] Step A5: Aggregate and calculate the allocation ratio of each path interval in each valid path in the valid path set to obtain the clearing ratio of ticket revenue of the operating entity in the target origin and destination stations.

[0108] Among them, the route interval refers to the line segment formed by stations that are visited continuously without transfers. The multiple linear regression model uses historical ticket revenue as the dependent variable and the factors affecting ticket revenue as independent variables.

[0109] In step A1, a three-dimensional network topology map containing station nodes, station edges, and transfer nodes is constructed, and basic attributes are defined, such as geometric attributes (mileage, travel time, coordinates of transfer nodes), operational attributes (operating entity to which the line belongs, departure interval, maximum passenger capacity), and service attributes (transfer time, congestion index), thereby forming a rail transit network.

[0110] Analysis of the core factors influencing ticket clearing reveals that operating costs determine the clearing benchmark weight, characterized by labor costs, maintenance costs, and management costs. Passenger flow reflects service intensity and contribution, characterized by average daily / peak passenger flow per route. Travel mileage reflects service volume, characterized by the actual travel distance in each segment. Transfer behavior affects route segmentation and revenue allocation, characterized by the number of transfers, waiting time, and walking distance. Passenger route selection determines the allocation ratio in multi-path scenarios, characterized by time sensitivity and convenience preferences. The key influencing factors of this invention include operating costs, passenger flow, and travel mileage. This invention proposes a scientific and reasonable ticket clearing mechanism, which not only helps solve the prominent problems in current ticket clearing and achieve a reasonable distribution of ticket revenue among various operating entities, but also protects the legitimate rights and interests of all parties, stimulates the enthusiasm and creativity of operating entities, helps improve service quality and operational efficiency, and enhances the level of regional transportation integration, but will also provide useful lessons and references for the future integrated transportation development of similar metropolitan areas.

[0111] See Figure 2 Furthermore, step A2 includes:

[0112] Step A21: Summarize the running paths of the target origin and destination stations to obtain the running path set;

[0113] Step A22: Calculate the path cost of each running path in the running path set;

[0114] Step A23: Based on path cost, filter the running paths from the set of running paths to form a preliminary set of paths;

[0115] Step A24: Calculate the initial probability of path selection for each running path in the preliminary path selection set;

[0116] Step A25: Based on the initial probability of path selection, select effective paths from the preliminary path selection set to form an effective path set;

[0117] In step A3, the path selection probability of the effective path is calculated based on the initial path selection probability and the path cost.

[0118] Furthermore, in step A22, the formula for calculating the path cost is as follows:

[0119]

[0120] Among them, R p The path cost refers to the cost of the execution path;

[0121] C i This represents the attraction factor of path interval i in the running path;

[0122] T iThis represents the running time of path interval i within the running path;

[0123] E xj This represents the transfer factor at the j-th transfer, which increases with the number of transfers.

[0124] T xj This represents the total walking and waiting time during the j-th transfer.

[0125] m represents the total number of segments in a route, divided according to transfer nodes.

[0126] Travel time is one of the most important factors for passengers when choosing a travel route. Shorter travel time means higher travel efficiency, so passengers tend to choose routes with shorter travel times. By calculating route costs and filtering based on time factors, passengers generally choose routes with shorter time intervals, i.e., shorter route costs.

[0127] Furthermore, travel distance is directly related to ticket price and is an important factor that passengers need to consider when choosing a route. Longer travel distances usually mean higher ticket costs, so passengers tend to choose shorter routes whenever possible. Incorporating travel distance into the settlement model reflects passengers' focus on travel costs and ensures more reasonable ticket settlement.

[0128] The number of transfers is a key factor affecting passenger travel experience. Frequent transfers not only increase travel time but can also cause inconvenience and uncertainty. Therefore, passengers usually try to avoid too many transfers when choosing a route. Including the number of transfers as a core element reflects passengers' pursuit of travel convenience, making the sorting model more aligned with passengers' actual travel habits.

[0129] The formula for calculating path cost comprehensively reflects these three aspects: path cost R. p It is a core indicator for quantifying passengers’ overall “time cost perception” of a route. Its physical meaning is equivalent travel time (i.e., the comprehensive subjective cost of passengers’ travel time, inconvenience of transfers, etc.), and the unit is consistent with time (e.g., minutes).

[0130] C iThe attractiveness factor reflects the impact of factors such as route comfort and punctuality on passengers' perceived time. It can be obtained by combining passenger ratings of route comfort (e.g., crowding, temperature, noise) and punctuality through questionnaires, assigning different weights to these indicators, and then summing the scores. Alternatively, it can be derived by extracting operational data such as punctuality, equipment failure rate, and passenger density from historical data analysis, and then fitting the relationship between the attractiveness factor and these operational indicators using models such as linear regression. The attractiveness factor can also be obtained through expert evaluation, where experts in transportation planning, operations management, and other fields score the overall service level of the route, and the weighted average is used to form the attractiveness factor.

[0131] To quantify the impact of transfer time and walking distance, actual measurement methods were used to measure the transfer waiting time and walking distance for different routes at each transfer station during different time periods (peak or off-peak) and calculate the average value. A three-dimensional model of the station was constructed using simulation analysis, and parameters such as station layout and passenger walking speed were input to simulate transfer behavior. Time and distance data under different scenarios were calibrated to obtain data. Simultaneously, real transfer experiences were collected through passenger feedback to supplement and improve the measurement and simulation results. Finally, a quantitative index of transfer impact that closely reflects actual travel scenarios was formed, resulting in E. xj and T xj This provides a precise basis for the calculation of the path selection probability model.

[0132] This invention uses "passenger flow allocation" as the core calculation unit when constructing the ticketing clearing model. This design aims to accurately reflect passengers' actual travel choices in a multi-network converged environment. To achieve this goal, the K-short path algorithm is used to generate a set of effective paths. The K-short path algorithm can comprehensively consider various factors such as network topology (divided into m line sections), line travel time, and transfer convenience, providing passengers with multiple feasible travel paths. Compared to a single shortest path algorithm, the K-short path algorithm better reflects the diversity of passenger path choices during travel, thus more accurately simulating actual passenger flow allocation.

[0133] Furthermore, in step A23, the formula for selecting running paths from the set of running paths based on path cost is as follows:

[0134]

[0135] in, This represents the upper critical value of the path cost;

[0136] The minimum path cost in the set of running paths;

[0137] α1 represents the magnification factor of the minimum path cost, which is the minimum cost extension percentage acceptable to passengers;

[0138] α2 represents the increment constant of the minimum path cost, which is the absolute time that passengers can accept beyond the shortest path cost.

[0139] In step A23, running paths with path costs less than the upper threshold value are selected from the running path set to form a preliminary path selection set.

[0140] Perform an initial screening of valid paths, setting two thresholds. Filter invalid paths, select the minimum value as the upper threshold, and filter all running paths, keeping only the invalid ones. The running path. Further, in step A24, assuming the path selection probability follows a normal distribution with a right-tail characteristic, the shortest path cost is used. Let μ be the mean and σ be the standard deviation, representing the passenger's sensitivity to cost (the smaller the σ, the higher the sensitivity). The initial probability of path selection for each operating path is calculated using an exponential decay function, as shown in the following formula:

[0141]

[0142] Where, p k This represents the initial probability of selecting the k-th running path in the initial path selection set;

[0143] This represents the path cost of the k-th running path in the initial path selection set;

[0144] σ represents the standard deviation of the path cost;

[0145] In step A25, running paths with an initial path selection probability lower than a preset probability threshold are filtered out from the initial path selection set to form an effective path set.

[0146] Specifically, the preset probability threshold is 10%.

[0147] Furthermore, if p k If the probability is below a preset threshold, the path is discarded, and its proportion is inversely proportional to the cost of the remaining paths and allocated to the effective paths. The running path with the lower path cost receives more allocation weight, as detailed below.

[0148] In step A3, the valid paths that were not discarded are proportionally redistributed, and the set of valid paths S is represented as:

[0149] S = {path 1, path 2, path 3, ..., path n}

[0150] The formula for calculating the path selection probability of an effective path is as follows:

[0151]

[0152] in, The probability of choosing the effective path i;

[0153] p i This represents the initial probability of path selection for the effective path i;

[0154] p 舍弃总和 This represents the sum of the initial probabilities of path selection for the running paths filtered out from the initial path selection set. The calculation formula is as follows:

[0155] p discarded total = ∑ k∈ Discard path p k

[0156] ω i The weight of a valid path i in the set of valid paths is represented by the following formula:

[0157]

[0158] This represents the path cost of the valid path i.

[0159] After obtaining the set of valid routes, the probability of route selection is calculated using a normal distribution model. The advantage of applying the normal distribution model in this study lies in its ability to effectively describe passengers' choice preferences when faced with multiple feasible routes. By assuming that passengers are influenced by various random factors when choosing a route, and that these factors collectively affect the passenger's decision-making process, the probability of selecting a particular route exhibits a normal distribution. This assumption not only aligns with reality but also facilitates precise mathematical calculations, resulting in a more scientifically sound probability of route selection.

[0160] Furthermore, in step A4, the formula for calculating the allocation ratio of the operating entity within the effective path interval is as follows:

[0161]

[0162] Among them, S i,k,m This represents the allocation ratio of operating entity i in the m-th segment of the k-th valid path in the valid path set;

[0163] This represents the path selection probability of the k-th valid path in the set of valid paths;

[0164] L i,k,m This represents the mileage that operator i is responsible for in the m-th segment of the k-th valid path in the valid path set;

[0165] L k,m This represents the total mileage of the m-th segment of the k-th valid path in the valid path set.

[0166] L k,总 This represents the total mileage of the k-th valid path in the set of valid paths;

[0167] C i,k,m This represents the operating cost of operator i in the m-th segment of the k-th valid path in the valid path set;

[0168] C k,m This represents the total operating cost of the m-th segment of the k-th valid path in the valid path set.

[0169] F i,k,m This represents the passenger flow of operator i in the m-th segment of the k-th valid path in the valid path set;

[0170] F k,m This represents the passenger flow of the m-th segment of the k-th valid path in the valid path set;

[0171] γ1 represents the weighting coefficient for travel distance as a key influencing factor;

[0172] γ2 represents the weighting coefficient for operating costs as a key influencing factor;

[0173] γ3 represents the weighting coefficient for passenger flow as a key influencing factor;

[0174] In step A5, the formula for calculating the percentage of ticket revenue to be distributed by the operating entity to the target origin and destination stations is as follows:

[0175]

[0176] Among them, S i This indicates the percentage of ticket revenue that operating entity i receives from the target origin and destination stations;

[0177] n k Let k represent the number of transfers on the effective path k, and n represent the number of effective paths in the effective path set.

[0178] See Figure 3 Furthermore, in step A4, the adjustment process for the weight coefficients of key influencing factors includes:

[0179] Step A41: Construct a multiple linear regression model. The calculation formula for the multiple linear regression model is as follows:

[0180] Y=β1X1+β2X2+β3X3+β4X4+ε

[0181] Where Y represents historical ticket revenue;

[0182] X1 to X4 represent four factors affecting ticket revenue: travel distance, operating costs, passenger flow, and number of transfers, respectively.

[0183] β1 to β4 represent the weighting coefficients of the four factors affecting ticket revenue: travel distance, operating costs, passenger flow, and number of transfers.

[0184] ε represents a constant, namely, the fluctuation of Y caused by all factors that may affect Y but are not included in the model, in addition to the four independent variables X1 to X4, as well as the specification error of the model itself;

[0185] Step A42: Obtain historical ticket revenue and use gradient descent to iteratively optimize the multiple linear regression model to determine the weight coefficients of factors affecting ticket revenue.

[0186] Step A43: Adjust the weight coefficients of key influencing factors based on the weight coefficients of factors affecting ticket revenue.

[0187] Using historical ticket revenue, the weight coefficients β of each key influencing factor are iteratively adjusted using the gradient descent method. i The goal is to minimize the mean squared error between predicted and actual revenue. K-fold cross-validation is then used to calculate the mean absolute error (MAE), ensuring the model fitting error is controlled within 5% and guaranteeing the reliability of the clearing rules.

[0188] Multiple linear regression models, because they can comprehensively consider the impact of various factors on ticket revenue, set ticket revenue as the dependent variable, while factors such as travel distance, operating costs, passenger flow, and number of transfers are used as independent variables. Travel distance, operating costs, and passenger flow are the key influencing factors. The magnitude of the weight coefficients of each factor affecting ticket revenue in the multiple linear regression model directly reflects the degree of influence of the corresponding factor on ticket revenue Y. If β1 is the largest, it indicates that travel distance has the most significant impact on revenue, and the weight of mileage γ1 can be appropriately increased. If β2 is prominent, it indicates that operating costs have a large impact on revenue, and the weight of cost γ2 needs to be increased. If β3 is dominant, it indicates that passenger flow is a key factor, and the weight of passenger flow γ3 should be increased to encourage the main body to attract more passengers.

[0189] See Figure 4 Furthermore, the rail transit network is divided into multiple sub-regions within the designated area;

[0190] In step A5, when the target origin and destination stations cross sub-regions, the calculation process for the full fare between the target origin and destination stations includes:

[0191] Step A51: Determine the corresponding operating entities within each sub-region;

[0192] Step A52: Calculate the fare for the shortest path segment between two stations within a sub-region based on the shortest path algorithm;

[0193] Step A53: Decompose the target origin and destination stations according to the shortest path segments within the sub-regions to form the shortest path segments of the decomposed sub-regions.

[0194] Step A54: Obtain the fare of the shortest path segment in each sub-region and sum them up to obtain the preliminary fare for the target origin and destination stations across regions.

[0195] Step A55: Use dynamic impact factors to correct the initial ticket prices of the target origin and destination stations to obtain the full ticket prices of the target origin and destination stations.

[0196] In urban rail transit systems that integrate multiple networks, cross-regional travel has become commonplace. Since different operators are responsible for different areas or lines, how to calculate cross-regional fares fairly and transparently has become a pressing issue. Traditional fare calculation methods are often based on single-path or simple summation principles, which are difficult to adapt to complex network structures and passenger travel needs.

[0197] This invention proposes a comprehensive cross-regional fare calculation scheme. Its core lies in calculating fares based on the shortest path within each regional road network. Based on the fare calculations within each sub-region, the complete origin-destination (OD) fare is then calculated by superimposing these calculations. First, shortest path algorithms such as Dijkstra's algorithm are used to calculate the shortest path between stations within a region and the corresponding fare. The fare calculation is based on mileage-based pricing principles, while also considering the combined effects of factors such as the number of transfers and travel time. Based on passenger travel needs, all possible paths from the origin to the destination are identified, including cross-regional paths. These paths may involve multiple regions and operating entities. For each cross-regional path, it is decomposed into multiple shortest path segments within each sub-region. Then, the shortest path fares within each sub-region are superimposed to form the preliminary fare for that cross-regional path. Considering the special characteristics and complexity of cross-regional travel, this study introduces dynamic factors to correct the preliminary fare. These dynamic factors include cross-regional transfer fees and peak-hour surcharges, which can be dynamically adjusted according to actual conditions. This fare reflects both the shortest path cost within each region and the dynamic factors of cross-regional travel.

[0198] See Figure 5 The present invention also provides a rail transit ticketing clearing system based on dynamic route selection, characterized in that, for executing the aforementioned rail transit ticketing clearing method based on dynamic route selection, it includes:

[0199] The network construction module (1) is used to construct a rail transit network that identifies the operating entities responsible for each line segment;

[0200] The effective path acquisition module (2) and the connecting network construction module (1) are used to filter the effective paths of the target starting and ending stations according to the rail transit network and form an effective path set;

[0201] The path probability calculation module (3) is connected to the effective path acquisition module (2) and is used to calculate the path selection probability of the effective paths in the effective path set.

[0202] The split module (4) is connected to the path probability calculation module (3), which is used to combine the path selection probability, the proportion of key influencing factors and the weight coefficients of key influencing factors dynamically adjusted by the multiple linear regression model to calculate the allocation ratio of the operating entity in each path interval in the effective path.

[0203] The aggregation module (5) and the connection splitting module (4) are used to aggregate and calculate the allocation ratio of each path interval of each valid path in the valid path set, so as to obtain the clearing ratio of the ticket revenue of the operating entity in the target origin and destination stations.

[0204] Key influencing factors include travel distance, operating costs, and passenger volume;

[0205] Among them, the route interval refers to the line segment formed by stations that are visited continuously without transfers. The multiple linear regression model uses the ticket revenue of the operating entity as the dependent variable and the key influencing factors as independent variables.

[0206] This invention combines the K-shortest path algorithm and a multiple linear regression model to quantitatively analyze the factors influencing ticket revenue, such as travel distance, operating costs, and passenger flow. It clarifies the mechanism by which each key element affects the clearing process, providing a scientific basis for the weighting and allocation logic in the clearing rules. Based on this, from the comprehensive calculation of cross-regional rail transit fares and multi-dimensional segmentation within sections to the final aggregation by operating entities, the system constructs multi-dimensional clearing rules to ensure that the allocation both aligns with the actual operational structure of the rail network and conforms to the revenue distribution logic under the influence of multiple factors.

[0207] The above are merely preferred embodiments of the present invention and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should recognize that any equivalent substitutions and obvious changes made based on the description and illustrations of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for clearing and distributing tickets for rail transit based on dynamic path selection, characterized in that, include: Step A1: Construct a rail transit network with identification of the operating entities responsible for each line segment; Step A2: Select valid routes from the rail transit network to the target origin and destination stations to form a set of valid routes; Step A3: Calculate the path selection probability of the effective paths in the effective path set; Step A4: Combining the path selection probability, the proportion of key influencing factors, and the weight coefficients of the key influencing factors dynamically adjusted by the multiple linear regression model, calculate the allocation ratio of the operating entity in each path interval of the effective path. The key influencing factors include travel mileage, operating costs, and passenger flow. Step A5: Aggregate and calculate the allocation ratio of each path interval of each valid path in the valid path set to obtain the clearing ratio of the ticket revenue of the operating entity in the target origin and destination stations. The path interval refers to a line segment formed by stations that are visited consecutively without transfers. The multiple linear regression model uses historical ticket revenue as the dependent variable and the factors affecting ticket revenue as independent variables.

2. The rail transit ticketing clearing method based on dynamic path selection as described in claim 1, characterized in that, Step A2 includes: Step A21: Summarize the running paths of the target origin and destination stations to obtain a running path set; Step A22: Calculate the path cost of each running path in the running path set; Step A23: Based on the path cost, filter the running paths from the set of running paths to form a preliminary set of paths; Step A24: Calculate the initial path selection probability of each running path in the preliminary path selection set; Step A25: Based on the initial probability of path selection, select effective paths from the preliminary path selection set to form the effective path set; In step A3, the path selection probability of the effective path is calculated based on the initial path selection probability and the path cost.

3. The rail transit ticketing clearing method based on dynamic path selection as described in claim 2, characterized in that, In step A22, the formula for calculating the path cost is as follows: Among them, R p The path cost of the aforementioned running path; m represents the total number of segments in a path divided according to transfer nodes; C i This represents the attraction factor of path interval i in the running path; T i This represents the running time of path interval i within the running path; E xj This represents the transfer factor at the j-th transfer, which increases with the number of transfers. T xj This represents the total walking and waiting time during the j-th transfer.

4. The rail transit ticketing clearing method based on dynamic path selection as described in claim 3, characterized in that, In step A23, the formula for filtering running paths from the set of running paths based on the path cost is as follows: in, This represents the upper critical value of the path cost; The minimum path cost in the set of running paths; α1 represents the magnification factor of the minimum path cost; α2 represents the increment constant of the minimum path cost; In step A23, running paths with a path cost less than the upper threshold value are selected from the set of running paths to form the preliminary set of paths.

5. The rail transit ticketing clearing method based on dynamic path selection as described in claim 4, characterized in that, In step A24, the formula for calculating the initial probability of path selection for each running path in the preliminary path selection set is as follows: Where, p k This represents the initial probability of path selection for the k-th running path in the preliminary path selection set; This represents the path cost of the k-th running path in the initial path selection set; σ represents the standard deviation of the path cost; In step A25, the running paths whose initial path selection probability is lower than a preset probability threshold are filtered out from the preliminary path selection set to form the effective path set.

6. The rail transit ticketing clearing method based on dynamic path selection as described in claim 5, characterized in that, In step A3, the effective path set S is represented as: S = {path 1, path 2, path 3, ..., path n} The formula for calculating the path selection probability of the effective path is as follows: in, The path selection probability of the effective path i; p i This represents the initial probability of path selection for the effective path i; p 舍弃总和 This represents the sum of the initial probabilities of path selection for the running paths filtered out from the initial path selection set, calculated using the following formula: p discarded total = ∑ k∈ Discard path p k ω i The weight of the effective path i in the effective path set is represented by the following formula: This represents the path cost of the effective path i.

7. The rail transit ticketing clearing method based on dynamic path selection as described in claim 1, characterized in that, In step A4, the formula for calculating the allocation ratio of the path intervals in the effective path by the operating entity is as follows: Among them, S i,k,m This represents the allocation ratio of the operating entity i to the m-th segment of the k-th valid path in the valid path set; This represents the path selection probability of the k-th valid path in the set of valid paths; L i,k,m This represents the mileage that the operating entity i is responsible for in the m-th segment of the k-th valid path in the valid path set; L k,m This represents the total mileage of the m-th segment of the k-th valid path in the set of valid paths; L k,总 This represents the total mileage of the k-th valid path in the set of valid paths; C i,k,m This represents the operating cost of the operating entity i in the m-th segment of the k-th valid path in the valid path set; C k,m This represents the total operating cost of the m-th segment of the k-th valid path in the set of valid paths; F i,k,m This represents the passenger flow of the operating entity i in the m-th segment of the k-th valid path in the valid path set; F k,m This represents the passenger flow of the m-th segment of the k-th valid path in the set of valid paths; γ1 represents the weighting coefficient for travel distance as a key influencing factor; γ2 represents the weighting coefficient for operating costs as a key influencing factor; γ3 represents the weighting coefficient for passenger flow as a key influencing factor; In step A5, the formula for calculating the percentage of ticket revenue that the operating entity receives from the target origin and destination stations is as follows: Among them, S i This represents the percentage of ticket revenue that the operating entity i receives from the target origin and destination stations; n k The number of transfers for the effective path k is represented by , and n represents the number of effective paths in the effective path set.

8. The rail transit ticketing clearing method based on dynamic path selection as described in claim 7, characterized in that, In step A4, the adjustment process for the weight coefficients of the key influencing factors includes: Step A41: Construct a multiple linear regression model. The calculation formula for the multiple linear regression model is as follows: Y=β1X1+β2X2+β3X3+β4X4+ε Where Y represents the historical ticket revenue; X1 to X4 represent four factors affecting ticket revenue: travel distance, operating costs, passenger flow, and number of transfers, respectively. β1 to β4 represent the weighting coefficients of the four factors affecting ticket revenue: travel distance, operating costs, passenger flow, and number of transfers. ε represents a constant; Step A42: Obtain historical ticket revenue and use gradient descent to iteratively optimize the multiple linear regression model to determine the weight coefficients of the factors affecting ticket revenue; Step A43: Adjust the weight coefficients of the key influencing factors based on the weight coefficients of the factors affecting ticket revenue.

9. A method for clearing and distributing rail transit tickets based on dynamic path selection as described in claim 1, characterized in that, The rail transit network is located in a predetermined area, which is divided into multiple sub-areas; In step A5, when the target origin and destination stations cross sub-regions, the calculation process for the full fare between the target origin and destination stations includes: Step A51: Determine the corresponding operating entity within each of the sub-regions; Step A52: Calculate the fare for the shortest path segment between two stations within the sub-region based on the shortest path algorithm; Step A53: Decompose the target origin and destination stations according to the shortest path segments within the sub-regions to form the shortest path segments of the decomposed sub-regions; Step A54: Obtain the ticket price of the shortest path segment of each of the decomposed sub-regions and sum them up to obtain the preliminary ticket price of the target origin and destination stations across regions. Step A55: Use dynamic impact factors to correct the initial ticket prices of the target origin and destination stations to obtain the full ticket prices of the target origin and destination stations.

10. A rail transit ticketing and clearing system based on dynamic path selection, characterized in that, A method for implementing a rail transit ticketing clearing method based on dynamic route selection as described in any one of claims 1-9, comprising: The network construction module is used to construct a rail transit network that identifies the operating entities responsible for each line segment; The effective path acquisition module, connected to the network construction module, is used to filter effective paths to target origin and destination stations based on the rail transit network and form an effective path set. The path probability calculation module, connected to the effective path acquisition module, is used to calculate the path selection probability of the effective paths in the effective path set; The split module, connected to the path probability calculation module, is used to calculate the allocation ratio of the operating entity in each path interval of the effective path by combining the path selection probability, the proportion of key influencing factors, and the weight coefficients of key influencing factors dynamically adjusted by the multiple linear regression model. The aggregation module, connected to the splitting module, is used to aggregate and calculate the allocation ratio of each path interval of each valid path in the valid path set, so as to obtain the clearing ratio of the ticket revenue of the operating entity in the target origin and destination stations. The key influencing factors include travel mileage, operating costs, and passenger flow; the route segment refers to a line segment formed by consecutive stations without transfers; the multiple linear regression model uses historical ticket revenue as the dependent variable and the ticket revenue influencing factors as independent variables.

Images