A ride comfort considering urban rail transit travel path recommendation method

Abstract

The application discloses a kind of urban rail transit travel path recommendation methods considering ride comfort, comprising the following steps: obtaining urban rail transit line network operation information, and constructing road network topology diagram;According to train timetable and passenger entry and exit station card swiping time, the travel path and ride train number of passenger are calculated;The passenger capacity of train passing through each section is calculated, and the section passenger flow is counted;Predict the section passenger flow in future period, and distribute to the train passing through the section;According to the predicted train passenger capacity, the train ride comfort and the comfort of section are calculated;Based on comfort, travel time, transfer delay and the like, the comprehensive right of way is calculated, and the shortest path set between each station is updated;Recommended travel path is generated.The method proposed in the application quantitatively represents train ride comfort on the basis of section passenger flow prediction, and generates recommended path by comprehensively considering ride comfort and travel efficiency, which has positive significance for improving passenger travel satisfaction and rail transit operation management level.

Description

Technical Field

[0001] This invention relates to the field of urban rail transit operation and management technology, specifically to a method for recommending urban rail transit travel routes that takes into account passenger comfort. Background Technology

[0002] As urban rail transit in major Chinese cities gradually enters a network-based operation phase, the increasing complexity of the network leads to a growing number of alternative routes for passengers, making it difficult for them to make the optimal choice based solely on intuition or experience. Especially during peak hours, congestion leads to longer transfer times and reduced comfort within the train, further complicating route selection. To improve passenger efficiency and satisfaction, it is necessary to provide route recommendations to guide passengers in choosing the optimal route and avoiding congested areas.

[0003] Early research (Dial 1967) suggested that passengers use only one minimum-cost path during their journeys. Subsequent studies found that passengers have a certain tolerance for their path choices, selecting paths within the effective path set according to their probabilities. This led to the development of a multi-path selection probability model (Nguyen 1988), which generates and evaluates the effective path set by constructing a generalized cost function. However, generating the optimal recommended path requires not only considering passengers' subjective path-choice behavior but also aiming to achieve the shortest generalized travel cost or a balanced distribution of network passenger flow. This necessitates generating the optimal path based on a specific passenger flow allocation model or criterion. Currently, rail transit passenger flow allocation models are mainly divided into two categories: those based on departure frequency (Schmocker 2011; Nuzzolo 2012) and those based on train timetables (Hamdouch and Szeto 2014; Shang 2018). In terms of travel cost assessment, the former typically uses deterministic variables or the mean of variables to calculate generalized costs, assuming that train departure frequencies and network conditions are known; the latter assumes that passengers cannot fully know train timetable information and uses the actual values ​​of random variables or factors to measure travel costs. However, research on both of these methods is currently limited to the theoretical level and is generally used for extrapolating passenger flow distribution in rail transit networks.

[0004] Existing methods mostly use travel efficiency (travel time) or toll cost (fare) as impedance when constructing path cost functions, lacking consideration for passengers' subjective feelings (such as ride comfort), and are mostly used for passenger flow allocation modeling, not yet practically applied to travel route recommendation. Although some studies consider train congestion as a factor in evaluating generalized costs, this requires the installation of specialized equipment (such as weighing equipment and video passenger flow meters) to detect the actual passenger load of trains to evaluate the degree of congestion, increasing equipment purchase and operating costs. In addition, there are currently a few research studies on route recommendation, but the application scenarios are generally for the evacuation of people inside subway stations during emergencies, or guiding passengers to choose suitable boarding stations based on station congestion, without involving travel route recommendation at the urban rail transit network level, which is different from the application scope of this invention. In view of this, this invention proposes an urban rail transit travel route recommendation method that considers ride comfort. Based on AFC data and train operating timetables, ride comfort is calculated to generate recommended travel routes, which has positive significance for improving rail transit travel efficiency and operation management level. Summary of the Invention

[0005] The technical problem to be solved by this invention is to provide a method for recommending urban rail transit travel routes that takes into account passenger comfort. This method can effectively utilize urban rail transit operation data, quantify the comfort of train travel based on the passenger flow forecast of the section, and use it as an important factor in calculating the generalized travel cost. When generating recommended routes, the method fully considers the passenger's travel experience and improves the rationality of the route recommendation information.

[0006] To address the aforementioned technical problems, this invention provides a method for recommending urban rail transit travel routes that considers passenger comfort, comprising the following steps:

[0007] S1. Use the geometric characteristic parameters of the urban rail transit network as the network topology information to construct a network topology map;

[0008] S2. Based on the road network topology map, train timetable, and passenger card swiping time at the station, use the K-shortest path algorithm to estimate the passenger's travel route and the train number they are taking.

[0009] S3. Based on the passenger's travel route and the train number information, calculate the passenger volume of each train passing through the section within the time period and count the passenger flow of the section.

[0010] S4. Predict the passenger flow of the section in the future time period, and allocate the passenger flow of the section in the future time period to the specific trains passing through the section to obtain the predicted train passenger capacity.

[0011] S5. Based on the predicted train passenger volume, calculate the train ride comfort and section comfort, wherein the section comfort is the average comfort of all trains passing through the section.

[0012] S6. Calculate the comprehensive right-of-way based on train ride comfort, travel time, and transfer delay, and determine the set of K shortest paths between each station by combining the road network topology map.

[0013] S7. Generate recommended travel routes.

[0014] Furthermore, step S1 includes:

[0015] S11. The urban rail transit information includes rail transit network information and rail transit operation information; the rail transit network information includes rail transit timetable, line length, station location, and train rated passenger capacity; the rail transit operation information includes the timetable for each line and passenger card swiping data of the AFC system.

[0016] S12. Construct a network topology map based on the above urban rail transit information, with stations as network nodes, travel time between adjacent stations as edge weights, and parking time as node weights.

[0017] Furthermore, step S2 includes the following steps:

[0018] S21. Based on the road network topology map, use the K-shortest path algorithm to calculate the feasible paths between each pair of stations and generate a set of feasible paths.

[0019] S22. For each passenger whose departure time or arrival time falls within time period t, determine whether the passenger has completed their trip within time period t based on the time of exiting the station and swiping their card. If the exit time is later than the end time of time period t, it is determined that the passenger has completed their trip; otherwise, the trip has not been completed.

[0020] S23. For passengers who have completed their trip, count the set of feasible paths between the passenger's origin and destination stations, assume that the passenger chooses each path in the set, and then combine the train timetable information to determine the number of trains the passenger has taken on each line and the possible departure time.

[0021] If a passenger travels along path P from station A to station B, with departure and arrival times respectively... , The walking times for passengers to enter and exit the station are as follows: , The starting station, train number, and time are determined according to the following rules: ① At station A, the passenger's boarding time is considered to be... The next train to arrive at the station, train V1, arrives at [time to be filled in]. ,in ② At station B, it is assumed that the passenger boarded earlier than the scheduled time. The adjacent train, v2, arrives at the station at [time to be filled in]. At this point, the total travel time for the passenger along path P is calculated using the following formula.

[0022]

[0023] in, v1 represents the total travel time for passengers along path P. If there are no transfers along path P, then v1 equals v2. If there are multiple transfers along path P, then based on the timetable information, the train number after the transfer is determined sequentially according to the arrival times of the trains before the transfer and the trains on the proposed transfer line at the transfer station.

[0024] S24. Calculate the total travel time and actual travel time for each feasible path. The difference is used to determine the passenger's travel route based on the feasible path that is closest to the real time, and the train number corresponding to that route is used as the passenger's boarding train number.

[0025] S25. For passengers who have not completed their trip within time period t, assuming that the passenger chooses the shortest route, determine the train number to take based on the train timetable, and predict the location to be reached at the end of the time period.

[0026] S26. Based on passenger travel route information, each section of the road network is marked sequentially. For section l, if passenger i travels through section l on train v during time period t, then it is marked as... ,otherwise, .

[0027] Furthermore, step S3 involves calculating passenger flow within a given area, including the following steps:

[0028] S31. For each interval, calculate the passenger volume of each train when it passes through that interval, and the passenger volume of train v when it passes through interval l in time period t. It can be represented as:

[0029]

[0030] S32. For each interval, calculate the sum of the passenger loads of all trains passing through that interval within time period t as the interval passenger flow. Then, calculate the interval passenger flow of interval l within time period t. It can be represented as:

[0031]

[0032] Furthermore, step S4 includes the following steps:

[0033] S41. Combining steps S2-S3 and historical AFC card swipe data, calculate the passenger flow over multiple historical days and construct a passenger flow data time series by arranging the passenger flow data of the same interval in chronological order.

[0034] S42. Construct an ARIMA (p, d, q) model to predict the passenger flow of each section of the road network in future time periods t+1, t+2, ..., as shown in the following formula.

[0035]

[0036] in, This is the original time series of passenger flow in interval l during time period t; express A stationary sequence after d differencing; and For the parameter to be estimated, Let be the zero-mean white noise random error sequence for time period t, where p and q are the order of the model;

[0037] S43. Combining the timetable information, the predicted passenger flow of the section is evenly distributed to the trains passing through the section in the future time period, and the predicted passenger volume of each section of the network is obtained in turn.

[0038] Furthermore, step S5 includes the following steps:

[0039] S51. The following formula is used to calculate the passenger comfort level of the train when passing through section l, based on the train's passenger capacity.

[0040]

[0041] in, Indicates train Comfort level during interval l; Indicates train Passenger capacity during interval l; , These represent the number of seats and the rated passenger capacity of the train, respectively. , These are the parameters to be calibrated;

[0042] S52. The average comfort level of each train on the section is taken as the section comfort level. The interval comfort level during time period t can be calculated as follows:

[0043]

[0044] in, This indicates the ride comfort level within time interval l, t. This represents the number of trains that pass through interval l during time period t.

[0045] Furthermore, step S6 includes the following steps:

[0046] S61. Calculate the comprehensive road weight based on ride comfort, travel time, and transfer delay. Define the comprehensive road weight function for path P as follows:

[0047]

[0048] in, The ride comfort level of route P is equal to the sum of the comfort levels of each interval. These represent the train travel time and transfer delay for route P; , These are the weighting coefficients;

[0049] S62. Update the shortest path set based on the road network topology information and the comprehensive road weight function, and sort the paths according to the road weight.

[0050] Furthermore, step S7, generating travel route recommendation information, includes the following steps:

[0051] Based on the shortest path set calculation results, the shortest and second shortest paths are used as recommended paths to generate route recommendation information, including the origin and destination points, route, transfer stations, travel time, and average comfort level.

[0052] The above proper nouns are explained as follows:

[0053] Section: refers to the operating section between adjacent stations on a rail transit line.

[0054] Section passenger flow: The sum of passenger volume of all trains passing through the section per unit time.

[0055] Sectional comfort: The average riding comfort of all trains passing through a certain section within a unit of time.

[0056] AFC data: Passenger card swiping data provided by the Automatic Fare Collection (AFC) system for rail transit, including passenger ID, card swiping time, location, and other information.

[0057] K-shortest path: The set of the k shortest possible paths between two stations in a transportation network, sorted by a certain right-of-way (such as travel time) from shortest to longest.

[0058] The beneficial effects of this invention are as follows:

[0059] 1. This invention proposes a method for recommending urban rail transit travel routes that takes into account riding comfort. When calculating the shortest path, it not only analyzes travel efficiency parameters such as travel time, but also considers the subjective feelings of passengers—riding comfort.

[0060] 2. By utilizing rail transit operation information, the system enables the prediction of passenger flow and the quantitative representation and estimation of comfort levels within a given section, eliminating the need to install any passenger flow detection equipment on the trains and effectively reducing equipment purchase and operating costs.

[0061] 3. The generated recommended routes can be applied to passenger flow guidance in rail transit, helping passengers avoid congested sections and improving travel experience and traffic efficiency. Attached Figure Description

[0062] Figure 1 This is a schematic diagram of the method flow of the present invention.

[0063] Figure 2 This is a partial topology diagram of a city's rail transit network, as an example.

[0064] Figure 3 The example illustrates the daily variation pattern of actual and predicted passenger flow within a given area.

[0065] Figure 4 The spatial distribution of road network section comfort is shown in the example.

[0066] Figure 5 This is a schematic diagram illustrating the results of generating travel route recommendation information as an example. Detailed Implementation

[0067] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the invention. After reading this invention, any modifications of the invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0068] like Figure 1 As shown, a method for recommending urban rail transit travel routes that considers passenger comfort includes the following steps:

[0069] S1. Use the geometric characteristic parameters of the urban rail transit network as the network topology information to construct a network topology map;

[0070] S2. Based on the road network topology map, train timetable, and passenger card swiping time at the station, use the K-shortest path algorithm to estimate the passenger's travel route and the train number they are taking.

[0071] S3. Based on the passenger's travel route and the train number information, calculate the passenger volume of each train passing through the section within the time period and count the passenger flow of the section.

[0072] S4. Predict the passenger flow of the section in the future time period, and allocate the passenger flow of the section in the future time period to the specific trains passing through the section to obtain the predicted train passenger capacity.

[0073] S5. Based on the predicted train passenger volume, calculate the train ride comfort and section comfort, wherein the section comfort is the average comfort of all trains passing through the section.

[0074] S6. Calculate the comprehensive right-of-way based on train ride comfort, travel time, and transfer delay, and determine the set of K shortest paths between each station by combining the road network topology map.

[0075] S7. Generate recommended travel routes.

[0076] The raw data for the AFC portion of the rail transit system in a certain city in Jiangsu Province in step S1 is shown in Table 1, the partial train timetable information is shown in Table 2, and the rail transit network topology diagram is shown in Table 2. Figure 2 As shown in Table 1, LINEID is the rail transit line number, STATIONID is the station number for card swiping at the exit, ENTRYSTATIONID is the starting station number (i.e., the station for card swiping at the entrance), ENTRYTIME is the time of card swiping at the entrance, and TRANSACTIONTIME is the time of card swiping at the exit. Each record in the table represents one trip. In Table 2, STATIONID is the station number, and TRAINID / ARRIVALTIME indicates the train number and arrival time at the station. 0201XX26, 0201XX27, 0201XX28, and 0201XXNN are train numbers. Due to the large number of trains on the same line within a day, this table only shows the timetable for a portion of the trains. Figure 2 This is a partial network operation map of a city used in the embodiment, including 3 lines and 53 stations (including 3 transfer stations). The station numbers in the map are consistent with those in Tables 1 and 2.

[0077] Table 3 shows an example of the passenger travel route and train number information calculated from the train timetable and the time of card swiping at the station in step S2. This table provides route information for three different origin and destination points. 2 " 5 "" indicates transfer stations in the route, and each record represents one trip.

[0078] The actual value of the interval passenger flow obtained from passenger routes and train numbers in step S3, and the predicted interval passenger flow for future time periods in step S4, are as follows: Figure 3 As shown in the figure, this graph illustrates the changes in passenger flow on a certain section (stations 38-39) of Metro Line 2 within a single day. The solid line represents the actual value calculated from AFC data, while the dashed line represents the predicted value based on the ARIMA model.

[0079] Table 1. Raw data of AFC (Automatic Facilitation) for rail transit in a city in Jiangsu Province

[0080] .

[0081] Table 2. Train Timetable for Line 2 (Upbound) during Certain Time Periods

[0082] .

[0083] Table 3 Examples of estimated travel routes and train schedules

[0084] .

[0085] The average comfort level of the inter-section trains calculated from the predicted passenger flow in step S5 is as follows: Figure 4 As shown in the figure, this graph represents the average comfort level across all sections of the rail transit network during a specific period of the morning rush hour (8:15-8:30). The shade and thickness of the lines indicate the magnitude of the comfort level value; the lower the value, the more comfortable the passengers feel. It can be seen that the spatial distribution of comfort levels is uneven across different lines and in both directions.

[0086] The shortest path k calculated based on the comprehensive right-of-way in step S6, and the path recommendation information generated based on the shortest path in step S7, are as follows: Figure 5 As shown in the diagram, taking a trip from station 3 to station 28 as an example, routes 1 and 2 are the shortest and second shortest routes calculated using the k-shortest path algorithm, respectively. The two routes involve transfers at station 2 and station 9, respectively. The right-hand box in the diagram shows route recommendations generated for this trip. Route 1 is more comfortable than route 2, although its travel time is slightly longer. Therefore, route 1 is recommended first, but the final choice depends on the passenger's personal preference.

[0087] As described above, this invention proposes a method for recommending urban rail transit travel routes that considers passenger comfort. This method effectively utilizes urban rail transit operation data, quantifies the comfort of train journeys within a given section based on passenger flow prediction, and fully considers passenger experience when generating recommended routes, effectively improving the rationality of route recommendation information. It achieves passenger flow prediction and comfort characterization without the need for any passenger flow detection equipment, effectively reducing equipment purchase and operating costs. The recommended routes can be used for passenger flow guidance, which is of positive significance for improving passenger travel experience and the level of rail transit operation management.

[0088] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for recommending urban rail transit travel routes considering passenger comfort, characterized in that, Includes the following steps: S1. Use the geometric feature parameters of urban rail transit information as road network topology information to construct a road network topology map; S2. Based on the road network topology map, train timetable, and passenger card swiping time at the station, use the K-shortest path algorithm to estimate the passenger's travel route and the train number they are taking. S3. Based on the passenger's travel route and the train number information, calculate the passenger volume of each train passing through the section within the time period and count the passenger flow of the section. S4. Predict the passenger flow of the section in the future time period, and allocate the passenger flow of the section in the future time period to the specific trains passing through the section to obtain the predicted train passenger capacity. S5. Based on the predicted train passenger volume, calculate the train ride comfort and section comfort, wherein the section comfort is the average comfort of all trains passing through the section. S6. Calculate the comprehensive right-of-way based on train ride comfort, travel time, and transfer delay, and determine the set of K shortest paths between each station by combining the road network topology map. S7. Generate recommended travel routes; Step S5 includes the following steps: S51. The following formula is used to calculate the passenger comfort level of the train when passing through section l, based on the train's passenger capacity. in, Indicates train Comfort level during interval l; Indicates train Passenger capacity during interval l; , These represent the number of seats and the rated passenger capacity of the train, respectively. , These are the parameters to be calibrated; S52. The average comfort level of each train on the section is taken as the section comfort level. The interval comfort level during time period t can be calculated as follows: in, This indicates the ride comfort level within time interval l, t. This represents the number of trains passing through interval l during time period t. Step S6 includes the following steps: S61. Calculate the comprehensive road weight based on ride comfort, travel time, and transfer delay. Define the comprehensive road weight function for path P as follows: in, The ride comfort level of route P is equal to the sum of the comfort levels of each interval. These represent the train travel time and transfer delay for route P; , These are the weighting coefficients; S62. Update the shortest path set based on the road network topology information and the comprehensive road weight function, and sort the paths according to the road weight.

2. The urban rail transit travel route recommendation method considering passenger comfort as described in claim 1, characterized in that, Step S1 includes: S11. The urban rail transit information includes rail transit network information and rail transit operation information; the rail transit network information includes rail transit timetable, line length, station location, and train rated passenger capacity; the rail transit operation information includes the timetable for each line and passenger card swiping data of the AFC system. S12. Construct a road network topology map based on the above urban rail transit information, with stations as network nodes, travel time between adjacent stations as edge weights, and parking time as node weights.

3. The urban rail transit travel route recommendation method considering passenger comfort as described in claim 1, characterized in that, Step S2 includes the following steps: S21. Based on the road network topology map, use the K-shortest path algorithm to calculate the feasible paths between each pair of stations and generate a set of feasible paths. S22. For each passenger whose departure time or arrival time falls within time period t, determine whether the passenger has completed their trip within time period t based on the time of exiting the station and swiping their card. If the exit time is later than the end time of time period t, it is determined that the passenger has completed their trip; otherwise, the trip has not been completed. S23. For passengers who have completed their trip, count the set of feasible paths between the passenger's origin and destination stations, assume that the passenger chooses each path in the set, and then combine the train timetable information to determine the number of trains the passenger has taken on each line and the possible departure time. If a passenger travels along path P from station A to station B, with departure and arrival times respectively... , The walking times for passengers to enter and exit the station are as follows: , The starting station, train number, and time are determined according to the following rules: ① At station A, the passenger's boarding time is considered to be... The next train to arrive at the station, train V1, arrives at [time to be filled in]. ,in ② At station B, it is assumed that the passenger boarded earlier than the scheduled time. The adjacent train, v2, arrives at the station at [time to be filled in]. At this point, the total travel time for the passenger along path P is calculated using the following formula. in, v1 represents the total travel time for passengers along path P. If there are no transfers along path P, then v1 equals v2. If there are multiple transfers along path P, then based on the timetable information, the train number after the transfer is determined sequentially according to the arrival times of the trains before the transfer and the trains on the proposed transfer line at the transfer station. S24. Calculate the total travel time and actual travel time for each feasible path. The difference is used to determine the passenger's travel route based on the feasible path that is closest to the real time, and the train number corresponding to that route is used as the passenger's boarding train number. S25. For passengers who have not completed their trip within time period t, assuming that the passenger chooses the shortest route, determine the train number to take based on the train timetable, and predict the location to be reached at the end of the time period. S26. Based on passenger travel route information, each section of the road network is marked sequentially. For section l, if passenger i travels through section l on train v during time period t, then it is marked as... ,otherwise, .

4. A method for recommending urban rail transit travel routes considering passenger comfort as described in claim 1 or 3, characterized in that, Step S3 involves calculating passenger flow within a given area, including the following steps: S31. For each interval, calculate the passenger volume of each train when it passes through that interval, and the passenger volume of train v when it passes through interval l in time period t. It can be represented as: S32. For each interval, calculate the sum of the passenger loads of all trains passing through that interval within time period t as the interval passenger flow. Then, calculate the interval passenger flow of interval l within time period t. It can be represented as: 。 5. The urban rail transit travel route recommendation method considering passenger comfort as described in claim 4, characterized in that, Step S4 includes the following steps: S41. Combining steps S2-S3 and historical AFC card swipe data, calculate the passenger flow over multiple historical days and construct a passenger flow data time series by arranging the passenger flow data of the same interval in chronological order. S42. Construct an ARIMA (p, d, q) model to predict the passenger flow of each section of the road network in future time periods t+1, t+2, ..., as shown in the following formula. in, This is the original time series of passenger flow in interval l during time period t; express A stationary sequence after d differencing; and For the parameter to be estimated, Let be the zero-mean white noise random error sequence for time period t, where p and q are the order of the model; S43. Combining the timetable information, the predicted passenger flow of the section is evenly distributed to the trains passing through the section in the future time period, and the predicted passenger volume of each section of the network is obtained in turn.

6. The urban rail transit travel route recommendation method considering passenger comfort as described in claim 1, characterized in that, Step S7, generating travel route recommendation information, includes the following steps: Based on the shortest path set calculation results, the shortest and second shortest paths are used as recommended paths to generate route recommendation information, including the origin and destination points, route, transfer stations, travel time, and average comfort level.

Images