Bus route time control point determination method considering delay propagation influence

Abstract

The invention discloses a bus route time control point determination method considering delay propagation influence, which belongs to the field of public traffic data mining application and mainly comprises the following steps of: calculating a delay value of each bus station based on bus arrival time data; determining the propagation range of delay generated by each bus number at each bus stop and the number of times of delay recovery at each bus stop; identifying a delay generation station based on a K-means algorithm; and determining a final time control point set based on the delay generation station. According to the method, on the basis of considering the fluctuation of the bus travel time, the delay generation intensity, the propagation range and the recovery capability of the station are measured by constructing three indexes including the delay reduction value, the delay propagation length and the recovery delay times from the angle of delay propagation, so that the accurate identification of the delay generation station is realized; by screening delay generation stations as time control points, delay generation can be reduced from the source, further propagation of delay is prevented, and reliability and operation efficiency are improved from the perspective of overall operation of bus lines.

Description

Technical Field

[0001] The present invention relates to a method for determining time control points of bus lines considering the influence of delay propagation, and belongs to the field of public transportation data mining applications. Background Art

[0002] Giving priority to the development of public transportation is an important means to achieve carbon emissions reduction in the transportation industry and alleviate urban traffic congestion. Since buses are subject to various internal and external factors during operation, vehicles often experience delays and the on-time rate decreases, resulting in an extended waiting time for passengers at stations and reducing passengers' willingness to choose bus travel.

[0003] Existing studies usually select several stations on a bus line as time control points to improve the reliability of bus operation by ensuring that buses arrive at the time control points on time. How to select time control points is particularly important. For example, "A Method for Setting Time Control Points in Bus Line Vehicle Scheduling" (Application No.: 201710305436.5) discloses a method for setting time control points. According to the total number of time control points and the number of stations on the bus line, alternative time control point schemes are generated, and then the scheme with the smallest sum of the average variances of travel times between all adjacent time control points is selected to set the time control points of the line. "A Method and Device for Generating an Autonomous Driving Bus Operation Plan Based on Variable Time Control Points" (Application No.: 2024 + 10795562.3) fits the probability distribution function of the travel time of bus vehicles between stations on the bus line based on the travel time between adjacent stations of the bus vehicle and the stopping time at each station, and then determines the key stations of the target bus line according to the historical passenger flow data of the bus vehicle and the line characteristic information of the bus vehicle to generate a set of time control points.

[0004] The above methods mainly select stations as time control points based on the volatility of bus travel times, but ignore the characteristics that the delays generated by buses will accumulate and spread during the operation process. If the source of the delay can be accurately located and appropriate scheduling strategies can be taken to recover the delay in time and alleviate the further spread of the delay, it will help to improve the reliability and operation efficiency of bus operation. Summary of the Invention

[0005] Aiming at the problems and improvement requirements of the existing technology, the present invention proposes a method for determining time control points of bus lines considering the influence of delay propagation, which can comprehensively consider the delay intensity, propagation range and recovery ability generated by stations to accurately identify the stations where delays occur, and then generate a set of time control points, so as to provide support for bus operation and in-transit operation control.

[0006] Step 1: Calculate the delay values of each bus station based on the bus arrival time data.

[0007] Step 1.1: The arrival time of train number n (n ∈ [1, N]) of a certain line at bus stop i (i ∈ [1, I]) is Using the arrival times of two adjacent trains at each stop and Calculate the arrival interval of the stop

[0008]

[0009] Step 1.2: Subtract the arrival interval of train number n (n ≥ 2) at each stop from the departure interval of this train number respectively, and take it as the delay value of this stop If is less than 0, set it to 0. The delay value of the starting stop is set to 0.

[0010] Step 1.3: Statistically calculate the delay increase value (i ≥ 2) and the delay decrease value (i ≥ 2) of each stop respectively. Subtract the delay value of stop i of train number n from the delay value of stop i - 1. When , it means that a new delay occurs for train number n at bus stop i, and the delay increase value is equal to the absolute value of, and at the same time set the delay decrease value to 0; when , it means that the delay of train number n is restored at bus stop i, and the delay decrease value is the absolute value of, and at the same time set the delay increase value to 0.

[0011] Step 2: Determine the propagation range of the delay generated by each train at each bus stop and the number of times the delay is restored at each stop.

[0012] Step 2.1: During the bus operation, if the delay generated by a train at a stop has not been fully restored and a new delay occurs at a subsequent stop, the delays should be matched and offset in the order of their occurrence. For the stop i with the delay value , start traversing from the (i + 1)-th stop, and sum up the delay decrease values from stop i + 1 to stop J (J ∈ [i + 1, I]).

[0013] Step 2.2: Determine the propagation range of the delay generated by each train at each bus stop. When is satisfied, it indicates that the delay generated at stop i has been restored at stop J. The propagation range of the delay generated at stop i is [i + 1, J], and the delay propagation length is J - i.

[0014] When J = I is satisfied, and at this time, it indicates that the delay generated by station i is restored at the terminal station, and the propagation range of the delay generated by station i is the interval [i + 1, I], and the delay propagation length is I - i.

[0015] Step 2.3: Determine the number of times each station restores the delay. For the stations in the propagation range [i + 1, J] of station i where the delay reduction value is greater than 0, increase the number of times it restores the delay by 1.

[0016] Step 2.4: Set the delay reduction values from station i + 1 to station J - 1 to 0, calculate the remaining delay reduction value of station J, which is used as the new delay reduction value of station J. Then process the next delay value of the station.

[0017] Step 3: Identify the stations where delays occur based on the K - means algorithm.

[0018] Step 3.1: Sum up the delay reduction values, the number of times of restoring the delay, and the delay propagation lengths of each station (excluding the starting station and the terminal station) within a day, and record them as the cumulative delay reduction value cumulative delay propagation length and cumulative number of times of restoring the delay The calculation formulas for the three indicators are as follows:

[0019]

[0020]

[0021] [[ID=4I]]

[0022] Step 3.2: Sum up the cumulative delay reduction values, cumulative delay propagation lengths, and cumulative numbers of times of restoring the delay of each station within multiple consecutive working days, and use the K - means clustering algorithm to cluster the stations into 3 categories. According to the values of the clustering centers of each category, define the stations as delay - occurring stations, delay - propagating stations, and delay - dissipating stations respectively.

[0023] Step 4: Determine the final set of time control points based on the delay - occurring stations.

[0024] Step 4.1: Add the starting station and the terminal station of the line to the set of time control points. Step 4.2: Two adjacent bus stops should be avoided from being set as time control points simultaneously. If there are multiple adjacent bus stops that are all delay occurrence stations, select the delay occurrence station with the largest cumulative delay propagation length among them to be added to the time control point set. If the station number of the delay occurrence station with the largest cumulative delay propagation length is 2 or I-1, exclude it and then re-select the delay occurrence station with the largest cumulative delay propagation length from the remaining delay occurrence stations to be added to the time control point set.

[0025] The present invention has the following beneficial effects compared with the prior art: On the basis of considering the volatility of bus travel time, from the perspective of delay propagation, three indicators, namely the delay reduction value, the delay propagation length, and the number of restored delays, are constructed to measure the intensity, propagation range, and restoration ability of the delays generated at the stations, so as to accurately identify the stations where delays occur; By screening the stations where delays occur as time control points, it is beneficial to reduce the generation of delays from the source and prevent the further propagation of delays, thereby improving the reliability and operation efficiency from the perspective of the overall operation of the bus line. Brief Description of the Drawings

[0026] Figure 1 It is a flowchart of the method described in the present invention. Detailed Embodiments

[0027] The present invention will be further described below in conjunction with the drawings and embodiments.

[0028] Refer to Figure 1 The present invention provides a method for determining time control points of a bus line considering the influence of delay propagation, including the following steps: Step 1: Calculate the delay values of each bus stop based on the bus arrival time data. Step 2: Determine the propagation range of the delays generated at each bus stop for each bus trip and the number of times each stop restores the delays. Step 3: Identify the delay occurrence stations based on the K-means algorithm. Step 4: Determine the final time control point set based on the delay occurrence stations.

[0029] Refer to the detailed embodiments as follows:

[0030] Step 1: Calculate the delay values of each bus stop based on the bus arrival time data.

[0031] Step 1.1: The bus arrival time data includes the time when each bus arrives at each bus stop. Taking the bus arrival time of Bus No. 7 in Beijing (from the Zoo Hub Station to the Wujianlou direction) on December 17, 2024 as an example, sort the vehicle numbers in ascending order of the departure time, and sort the stations in ascending order of the station numbers. The data sample after sorting the bus arrival time data is as follows:

[0032] Sample of bus arrival time data in Table 1

[0033] Record the arrival time of the bus number n (n ∈ [1, N]) of the bus line at the bus stop i (i ∈ [1, I]) as Using the arrival times of two adjacent bus numbers at each stop and Calculate the arrival interval of the stop

[0034]

[0035] Step 1.2: Subtract the arrival interval of the bus number n (n ≥ 2) at each stop from the departure interval of this bus number respectively as the delay value of this stop If is less than 0, set it to 0. The delay value of the starting stop is set to 0.

[0036] Step 1.3: Statistically calculate the delay increase value [[ID=3�]]and delay reduction value Subtract the delay value of stop i of bus number n from the delay value of stop i - 1. When , it means that a new delay occurs at bus stop i for bus number n, and the delay increase value is equal to the absolute value of, and at the same time set the delay reduction value to 0; when , it means that the delay is restored at bus stop i for bus number n, and the delay reduction value is the absolute value of, and at the same time set the delay increase value to 0. Taking the first bus number and the second bus number as an example, the calculation results are shown in Table 2. The interval between the first bus number and the second bus number at the Zoo Hub Station is 1204 seconds, that is, the departure interval is 1204 seconds. The interval between the two bus numbers at the Xizhimenwai Station is 1213 seconds. Then the delay value of the second bus number at the Xizhimenwai Station is 9 seconds, the delay increase value is 9 seconds, and the delay reduction value is 0.

[0037] Sample of delay value calculation results in Table 2

[0038] Step 2: Determine the propagation range of the delay generated by each bus number at each bus stop and the number of times the delay is restored at each stop.

[0039] Step 2.1: For the delay value For station i, starting from the (i + 1)-th station, sum up the delay reduction values from station (i + 1) to station J (J ∈ [i + 1, I]). Sum them up.

[0040] Step 2.2: When is satisfied, it indicates that the delay generated at station i has been restored at station j. The propagation range of the delay generated at station i is [i + 1, J], and the propagation station length is J - i.

[0041] When J = I and is satisfied, it indicates that the delay generated at station i has been restored at the terminal station, and the propagation range of the delay generated at station i is the interval [i + 1, I], and the delay propagation length is I - i.

[0042] Step 2.3: For the stations with delay reduction values greater than 0 in the propagation range [i + 1, J] of station i, increment the number of times their delays are restored by 1.

[0043] Step 2.4: Set the delay reduction values from station (i + 1) to station (J - 1) to 0, calculate the remaining delay reduction value of station J, which is used as the new delay reduction value of station J. Then process the next delay value of the station. of the station.

[0044] The calculation results of the delay propagation length and the number of times the delay is restored are shown in Table 3. Taking Train No. 2 and Train No. 3 as examples, the delay increase value of Train No. 3 at the "Baocan Hutong" station is 6 seconds. Start searching from the "Zuojiajie" station and sum up the delay reduction values of each station; when searching to the "Baozi Hutong" station, the sum of the delay reduction values is 71, which is greater than the delay increase value of the "Baocan Hutong" station, so the search ends; set the delay propagation length of the "Baocan Hutong" station to 2 and the number of times the delay is restored to 0; since the delay generated at the "Baocan Hutong" station has not been fully restored at the "Zuojiajie" station and a new delay is generated, with a delay increase value of 71 seconds, the delay reduction value of the "Baozi Hutong" station is first used to restore the delay generated at the "Baocan Hutong" station, and then the remaining delay reduction value of the "Baozi Hutong" station is updated to 71 - (6 - 0) = 65, which is used to restore the delay generated at the "Zuojiajie" station.

[0045] Table 3 Example of the calculation results of the delay propagation length and the number of times the delay is restored

[0046] Step 3: Identify the stations where delays occur based on the K-means algorithm.

[0047] Step 3.1: Sum up the delay reduction values, the number of restored delays, and the delay propagation lengths of all train trips at the stations (excluding the originating and terminating stations) on December 17, 2024, and denote them as the cumulative delay reduction value Cumulative delay propagation length and the cumulative number of restored delays The calculation formulas for the three indicators are as follows:

[0048]

[0049]

[0050]

[0051] Table 4 Sample calculation results of the three indicators

[0052] Step 3.2: Sum up the cumulative delay reduction values, the cumulative delay propagation lengths, and the cumulative number of restored delays at the stations for a total of 5 working days from December 16, 2024, to December 20, 2024, and use the K-means clustering algorithm to cluster the stations into 3 categories. The clustering results are shown in Table 5

[0053] Table 5 Clustering results

[0054] Statistically calculate the values of the clustering centers for each category, and sort them in descending order with reference to the central values of the cumulative delay reduction values. The results are shown in Table 6. Define the stations in categories 0, 1, and 2 as delay dissipation stations, delay propagation stations, and delay occurrence stations respectively. Delay dissipation stations have relatively large cumulative delay reduction values and cumulative delay restoration times, indicating strong delay restoration capabilities. Although delay propagation stations have relatively large delay propagation lengths, their cumulative delay reduction values and cumulative delay restoration times are also relatively high, indicating relatively strong self-restoration capabilities. Delay occurrence stations have relatively large central values of cumulative delay propagation lengths and relatively small central values of cumulative delay reduction values, indicating that the delays generated by buses at delay occurrence stations are intense and the delay propagation range is wide, and the delay restoration capabilities are poor. Operational strategies such as shuttle buses and limited-stop express buses, as well as operation control strategies such as speed control and bus signal priority, need to be used to mitigate the generation and propagation of delays

[0055] Table 6 Clustering center values

[0056] Step 4: Determine the final set of time control points based on the delay occurrence stations

[0057] Step 4.1: Add the starting station and the terminal station of the line to the set of time control points.

[0058] Step 4.2: Two adjacent bus stops should be avoided from being set as time control points simultaneously. If there are multiple adjacent bus stops that are all delay occurrence stations, then select the delay occurrence station with the largest cumulative delay propagation length among them and add it to the set of time control points; if the station number of the delay occurrence station with the largest cumulative delay propagation length is 2 or I - 1, then exclude it and re - select the delay occurrence station with the largest cumulative delay propagation length among the remaining delay occurrence stations and add it to the set of time control points. The time control point setting results for Bus Route 7 (from the Zoo Hub Station to Wujianlou) are shown in Table 7. The four adjacent stations of Xizhimenwai, Xizhimennei, Xinkai Hutong, and Baochan Hutong are all delay occurrence stations. The cumulative delay propagation length of the Baochan Hutong station is the largest at 880, so Baochan Hutong is selected as the time control point.

[0059] Table 7 Time Control Point Setting Results for Bus Route 7 (from the Zoo Hub Station to Wujianlou)

Claims

1. A method for determining the time control points of bus lines considering the impact of delay propagation, characterized in that, It includes the following steps: Step 1: Calculate the delay values of each bus stop based on the bus arrival time data; Step 2: Based on the delay values of each bus stop obtained in Step 1, determine the propagation range of the delay generated by each bus trip at each bus stop and the number of times each stop recovers from the delay; Step 3: Identify the delay occurrence stations for the data determined in Step 2 based on the K-means algorithm; Step 4: Determine the final set of time control points based on the delay occurrence stations obtained in Step 3.

2. The method for determining the time control points of a bus line considering the impact of delay propagation according to claim 1, characterized in that The step of calculating the delay values of each bus stop based on the bus arrival time data described in Step 1 specifically includes: Step 1.1: For the train number n of a certain line, where n ∈ [1, N], the arrival time at bus stop i is i ∈ [1, I]. Use the arrival times of two adjacent trains at each stop and calculate the arrival interval of the stop n ≥ 2; Step 1.2: The arrival intervals of train number n at each station are respectively subtracted from the departure interval of this train number with n≥2, as the delay value of this station If it is less than 0, then set it to 0; the delay value of the starting station is set to 0; Step 1.3: Statistically calculate the increased delay value and the decreased delay value for each station and the decreased delay value For i ≥ 2, subtract the delay value of stop i of train n from the delay value of stop i - 1 of train n. When , it indicates that a new delay is generated at bus stop i for train n, and the increased delay value equals the absolute value of , and at the same time, set the decreased delay value to 0. When it indicates that the delay of train n is restored at bus stop i, and the decreased delay value is the absolute value of , and at the same time, set the increased delay value to 0.

3. The method for determining the time control points of a bus line considering the impact of delay propagation according to claim 1, characterized in that In Step 2, determining the propagation range of the delay generated by each bus trip at each bus stop and the number of times each stop recovers from the delay specifically includes: Step 2.1: During the operation of a bus, if the delay generated at a station for a bus trip has not been fully recovered and a new delay occurs at a subsequent station, the delays are sequentially matched and offset according to the order of occurrence; for the station i with a delay value , starting from the (i + 1)-th station, traverse and sum up the delay reduction values from station (i + 1) to station J, where J ∈ [i + 1, I]; ​ Step 2.2: Determine the propagation range of the delay generated by each train number at each bus stop; When meeting it indicates that the delay generated at stop i has been restored at stop J, the propagation range of the delay generated at stop i is [i + 1, J], and the propagation stop length is J - i; When meeting J = I, and it indicates that the delay generated at stop i has been restored at the terminal stop, and the propagation range of the delay generated at stop i is the interval [i + 1, I], and the delay propagation length is I - i; Step 2.3: Determine the number of times of recovery delay for each site; for the sites in the propagation range [i + 1, J] of site i where the delay reduction value is greater than 0, increment the number of times of their recovery delay by 1; Step 2.4: Set the delay reduction value from site i + 1 to site J - 1 to 0, and calculate the remaining delay reduction value of site J as the new delay reduction value of site J; then process the next site with a delay value as described. ​ 4. The method for determining the time control points of a bus line considering the impact of delay propagation according to claim 1, characterized in that, The step of identifying the delay occurrence stations based on the K-means algorithm described in Step 3 specifically includes: Step 3.1: Sum up the delay reduction value, the number of restored delays, and the delay propagation length for each station excluding the origin and destination stations within a day, and record them as the cumulative delay reduction value Cumulative delay propagation length And the cumulative number of restored delays The calculation formulas for the three indicators are And Step 3.2: Sum up the cumulative delay reduction value, the cumulative delay propagation length, and the cumulative number of times of recovering from the delay of each stop within multiple consecutive working days, and use the K-means clustering algorithm to cluster the stops into 3 categories. According to the values of the clustering centers of each category, define the stops as delay occurrence stations, delay propagation stations, and delay dissipation stations respectively.

5. The method for determining the time control points of a bus line considering the impact of delay propagation according to claim 1, wherein The step of determining the final set of time control points based on the delay occurrence stations described in Step 4 specifically includes: Step 4.1: Add the starting station and the terminal station of the line to the set of time control points; Step 4.2: Two adjacent bus stops should be avoided from being set as time control points simultaneously. If there are multiple adjacent bus stops that are all delay occurrence stations, then select the delay occurrence station with the largest cumulative delay propagation length among them to be added to the set of time control points; if the station number of the delay occurrence station with the largest cumulative delay propagation length is 2 or I-1, then exclude it and re-select the delay occurrence station with the largest cumulative delay propagation length among the remaining delay occurrence stations to be added to the set of time control points.

Images