Method for determining bus route midway control station based on station service comprehensive index

Abstract

This invention discloses a method for determining intermediate control points on bus routes based on a comprehensive suitability index. This method relies on multi-source public transport big data, including historical transaction data and static route and station data. After data cleaning, matching, and key field extraction, it constructs and calculates a bus stop importance index and a bus arrival time stability index, thereby obtaining a comprehensive suitability evaluation index to comprehensively measure the suitability of each stop as an intermediate control point. A reasonable number of intermediate control points is determined through appropriate rules, generating all control point schemes. The comprehensive suitability evaluation index for each scheme is calculated, and the optimal control point scheme is selected. The suggested arrival time of vehicles at the control point under this scheme is obtained using historical arrival time data. This invention helps improve the punctuality rate of bus routes, avoids problems such as adjacent buses running in parallel or large intervals between buses, and improves the reliability and service quality of bus routes.

Description

Technical Field

[0001] This invention relates to a method for selecting and determining intermediate control stations on ground bus routes based on bus card swiping data and bus vehicle positioning data. This method helps improve the on-time performance of bus routes and belongs to the field of public transportation data mining applications and operation optimization. Background Technology

[0002] Surface public transport is a crucial component of urban public transportation. Efficient and reliable surface public transport services are key to alleviating urban traffic congestion, ensuring residents' normal lives, and supporting and promoting sustainable urban development. Currently, surface public transport is transitioning from a phase of increasing public transport service coverage to a more refined development phase focused on improving passenger service quality. Public transport punctuality is one of the core indicators reflecting the reliability of public transport services; improving punctuality is an important measure to enhance public transport service quality and increase passenger satisfaction. However, surface public transport operations are often affected by factors such as morning and evening rush hour traffic congestion, unexpected road incidents, and intersection delays, resulting in low punctuality rates and low passenger satisfaction. The root cause is that public transport systems typically employ a "two-terminal checkpoint" timetable design method during actual operation and scheduling, meaning that only the departure time at the originating station and the arrival time at the destination station are specified. This traditional dispatching method lacks a defined arrival time for buses at each stop along the route and suffers from a lack of comprehensive analysis and adjustment of the intermediate processes of bus route operation. Drivers struggle to find scientific basis for controlling speed and stop times, leading to instances where drivers accelerate or decelerate near the route's end to achieve punctuality. This easily results in delays, missed buses, and other issues that lower the service level of the public transportation system. These phenomena disrupt established driving plans, increase the difficulty of station and bus organization, and objectively reduce the punctuality rate and service reliability of public transportation. In response to this, some cities, led by Beijing, have adopted a series of measures, including promoting the construction of a "punctuality project" for public transportation. These measures improve the punctuality rate of ground public transportation services by strengthening timetable management, introducing fixed-departure routes, and implementing centralized regional dispatching.

[0003] To facilitate timely and dynamic adjustments by bus drivers, optimize route timetable compilation and management, and improve the punctuality rate of bus routes, bus operators often implement measures such as setting up intermediate control stations along the route. By controlling the punctuality rate of buses arriving at each control station, the overall punctuality of the route is improved. Intermediate control stations on a bus route are important stations selected from all stops on a bus route for controlling vehicle movement. Setting up reasonable intermediate control stations is crucial for developing bus route scheduling plans and improving the reliability of bus services.

[0004] In both domestic and international bus route operations, intermediate control points are mostly determined based on the practical experience of operation managers. Operation departments typically select control points based on passenger flow at each stop, choosing those with high passenger volume for time control. However, this approach fails to consider the impact of road conditions at different times, i.e., the speed of traffic between stops, on the selection of control points. This can lead to significant fluctuations in travel time between control points, making it impossible to generate an optimal time control point setting scheme, resulting in reduced vehicle punctuality and route operational efficiency. Chinese invention patent CN106971532A discloses a method for setting time control points in bus route vehicle scheduling. This patent generates multiple alternative time control point schemes, calculates the sum of the average variances of travel time between adjacent time control points for all vehicles within a given operating period for each scheme, finds the evaluation index with the smallest value, and selects the corresponding time control point as the time control point for the given operating period. This patent takes into account the stability of bus arrival times and solves the shortcomings of existing time control point settings that do not consider the impact on bus punctuality and total vehicle travel time. However, it does not comprehensively consider factors such as station size, location and passenger flow when determining control points, and still has certain limitations in practical applications.

[0005] With the rapid development of intelligent transportation technology and public transportation information infrastructure, advanced vehicle positioning technology and electronic payment systems for public transportation have enabled the recording of the entire operation process of public transportation routes. This data allows for the extraction of accurate information on bus route operations, vehicle operation, and passenger flow. Given the massive amounts of public transportation data available, how to leverage this big data to reduce the uncertainty of bus travel times, ensure on-time arrival rates, and improve service levels and operational efficiency are pressing issues that need to be addressed in the field of public transportation operation optimization. Summary of the Invention

[0006] This invention aims to propose a method for selecting and determining intermediate control stations based on a comprehensive station service index. This method selects and determines the optimal location and number of intermediate control stations, and breaks down the punctuality rate benchmark for the bus terminal into punctuality rate benchmarks for reaching several intermediate control points. If a bus arrives at an intermediate control station earlier than the scheduled time, it can improve the punctuality rate by making a brief stop at the control station or appropriately slowing down during subsequent journeys. If a bus arrives later than the scheduled time, it can increase its speed, and, if road conditions permit, immediately depart after serving passengers at subsequent stops, to ensure punctuality at the next control station, thereby guaranteeing the punctuality rate of the entire route. Utilizing these intermediate control points allows for relatively stable control of the vehicle's speed, position, and condition during operation, helping to improve the punctuality rate of bus routes, avoid problems such as overlapping buses or large intervals between buses, and improve the reliability and service quality of bus routes.

[0007] This invention is based on multi-source ground public transport data, including bus arrival data and bus card / QR code swipe data. First, it adds intermediate control points to the existing control at the terminal stations of bus routes. It proposes principles for setting these intermediate control points and determines comprehensive evaluation indicators to screen stations, thereby identifying the optimal control points. Finally, by combining historical arrival time data, it generates planned arrival times for vehicles at each intermediate control point, providing bus operators and drivers with more accurate references.

[0008] To achieve the above objectives, the present invention adopts the following technical solution.

[0009] A method for determining intermediate control stops on bus routes based on a comprehensive station service index, the method comprising the following steps:

[0010] Step 1, Ground Public Transport Data Preprocessing

[0011] The multi-source data for ground public transportation proposed in this patent mainly includes dynamic operation data of ground public transportation collected through electronic devices and static basic data from existing digital management systems. The dynamic operation data of public transportation includes ground public transportation card swiping data, ground public transportation QR code swiping data, and ground public transportation arrival time data, while the static basic data mainly refers to static bus route and station data.

[0012] Urban buses are equipped with public transport IC cards and electronic QR code readers. The card and QR code data obtained from these transactions is imported into the data center's database for unified storage in real time. IC card transaction data includes information such as the user's card number, transaction time, route number, vehicle number, boarding and alighting station numbers, and boarding / alighting indicators, as shown in Table 1. QR code data includes information such as the QR code ID, card number, boarding / alighting stations, and boarding / alighting times, as shown in Table 2. Transaction data is generally used to cover and represent different types of electronic payment behavior data.

[0013] Table 1 Fields of Ground Public Transport IC Card Swiping Data Table

[0014]

[0015]

[0016] Table 2. Fields of QR code scanning data for ground public transportation

[0017]

[0018] Real-time arrival time data for ground public transportation is important for extracting bus arrival intervals and waiting times at bus stops. It mainly includes eight valid fields: route name, direction, stop name, stop number, vehicle number, distance, arrival date and time, as shown in Table 3.

[0019] Table 3 Example of Fields in Ground Public Transportation Arrival Time Data Table

[0020]

[0021] Static bus route and station data is crucial for matching and associating card-swipe data. Based on this data, individual travel stage information can be accurately extracted. Ground-level bus static route and station data mainly includes route direction, name, station name, and station latitude and longitude, as shown in Table 4.

[0022] Table 4 Example of Bus Static Route and Station Data Fields

[0023]

[0024] The ground public transport IC card swipe data only records the boarding and alighting station numbers, not the station names. Therefore, it is necessary to first use the station number as a common field to correlate and match the IC card swipe data with the static bus route and station data to obtain the station name for each transaction. Secondly, due to the large number of bus card readers, short route adjustment cycles, and data feedback mechanisms, some public transport transaction data is of poor quality. Therefore, data filtering is also required to remove erroneous data such as null values, duplicate values, identical boarding and alighting stations, and boarding times later than alighting times. Based on the filtered ground public transport transaction data, historical transaction data within the research period is selected, and data from special dates and times such as major events and holidays is removed. Finally, the effective transaction data fields required for calculating passenger boarding and alighting volumes at each bus station are selected from the historical transaction data fields. Key transaction data fields should include: IC card number, route number, vehicle number, device number, boarding and alighting station numbers, boarding and alighting indicators, and transaction time.

[0025] Step 2: Extraction and averaging of bus stop boarding and alighting data.

[0026] Using passenger transaction data such as bus card or QR code payments, and based on vehicle number, boarding / alighting station number, and transaction time, combined with bus arrival data, the boarding / alighting volume at stations with the same vehicle number and boarding / alighting station number is calculated within a transaction time difference of no more than 1 minute. Among multiple data records meeting these conditions, the median transaction time is extracted as the corresponding time for calculating the boarding / alighting volume at that station. This allows us to obtain the boarding / alighting volume of each station on a given bus route within the study area over a continuous 5-day period t. Assume that on weekday a (a = 1, 2, 3, 4, 5), the number of passengers boarding at station i during time period t is Q. b i,t Number of people getting off the bus Q a i,t The number of arrivals and departures at the station during that period. The average station entry and exit points for the same time period on the same workday within the analysis period are calculated by averaging the data. Depending on the specific scenario and required level of precision, time granularities such as 15 minutes, 30 minutes, 1 hour, and 1 day can be selected for statistical analysis.

[0027]

[0028] In the formula, is the login / drop volume of site i during the t-th time period on weekday a; a is the a-th weekday.

[0029] Table 5 shows an example of preprocessed historical transaction data for public transport passengers. The number of passengers boarding and alighting at ground bus stops can be obtained by statistical analysis, as shown in Table 6.

[0030] Table 5. Examples of Historical Transaction Data for Ground Public Transport Passengers

[0031]

[0032]

[0033] Table 6 Example of Ground Public Transport Station Entry and Exit Data

[0034]

[0035] Step 3: Calculate the comprehensive evaluation index of bus stop services.

[0036] The comprehensive service index of public transport stops consists of two parts: the importance index of stops and the stability index of bus arrival time.

[0037] Step 3.1, calculate the site importance index.

[0038] The importance of a bus stop is not only related to passenger traffic volume, but also influenced by factors such as the type of land use around the stop and the characteristics of the urban location. This patent proposes a method for calculating the importance index of a bus stop by combining location characteristics and the proportion of passenger traffic. The location characteristics of a stop are reflected by the type and density of POIs (Points of Interest) around the stop. Therefore, the more numerous and prevalent the shopping malls, schools, hospitals, office areas, and other facilities within an 800-meter radius of the stop, the more prominent the relative importance of the stop's location. A location importance correction coefficient P is set. i This is placed before the percentage of arrivals and departures at a given site to form a site importance index. The importance index M of site i within time period t. i The calculation method is as follows:

[0039]

[0040] in, Let P be the average number of stops for the t-th time period; n is the number of stops on the line; i This is a site location importance correction coefficient, reflecting the proportion of POIs considering weight. It characterizes the locational attributes around a site and can more objectively measure the importance of a site along the entire line. Its calculation method is as follows:

[0041]

[0042] In the formula, H represents the number of POIs of type f surrounding site i; f λ represents the number of POIs of type f around the entire line; F is the total number of POI categories, which is taken as 8 in this patent; f The weighting coefficient for POI of type f is determined based on the POI category, land area, and the importance of public transport services. The value can be found in Table 7.

[0043] Table 7 Weighting coefficients for different categories of POIs

[0044]

[0045] Step 3.2, calculate the stability index of bus arrival time.

[0046] Existing time control point setting methods mostly consider the interval travel time stability between adjacent stations or adjacent control points, but do not consider the overall travel time stability of the vehicle from the starting station to a certain intermediate station. However, in the actual operation of a bus route, the overall travel time stability of the route has a greater impact on the overall passenger satisfaction along the route. Therefore, this patent uses the time granularity selected in step 2 as a basis, and calculates the travel time T of the vehicle to the i-th station by subtracting the time when the vehicle arrives at the i-th station within time period t on a weekday a from the vehicle's departure time from the starting station. ai1 T ai1 This includes the vehicle's travel time between stations before station i, as well as the dwell time at each station. It comprehensively considers road conditions and the time spent boarding and alighting at stations along the route, providing a more complete reflection of the overall operation of the bus on the road. The stability of arrival time is represented by the coefficient of variation of travel time. The smaller the fluctuation in arrival time, the smaller the coefficient of variation, and the higher the stability. Therefore, the stability index ST of arrival time at station i during weekday time periods a and t is... t ai The calculation method is shown in equation (4), where N is the total number of stations on the line. This represents the average travel time.

[0047]

[0048] For ST t ai After averaging, the average arrival time stability index of station i during the time period t on a weekday is obtained.

[0049]

[0050] In the formula, ST t ai It is the stability index of arrival time of site i during time period t on weekday a; a is the a-th weekday.

[0051] Step 3.3, Indicator Normalization and Calculation of Comprehensive Indicator Value

[0052] In the comprehensive evaluation of bus stop services, the different physical meanings represented by each indicator result in dimensional differences. This dimensionality discrepancy is a significant factor affecting the overall evaluation of the system. Therefore, before evaluation, the evaluation indicators must be uniformly converted into quantitative values ​​within the range of [0, 1]. This patent normalizes the importance index and arrival time stability index of each stop along the bus route and obtains the comprehensive evaluation index of the stop through the product of the two indicators.

[0053] First, the metrics are normalized. Normalized metrics for site importance:

[0054]

[0055] In the formula, M min It is the smallest site importance metric value, M max It is the most important site importance indicator.

[0056] Station arrival time stability normalization index:

[0057]

[0058] In the formula, It is the smallest arrival time stability index value. It is the highest stability index value for arrival time.

[0059] Time control points should be set at stations with high importance and strong arrival time stability. Based on this, the concept of a comprehensive service evaluation index for public transport stations is proposed, which is reflected as the product of the station importance index and the reciprocal of the station arrival time stability index. The higher the comprehensive service index, the more suitable station i is to be selected as a time control point. The comprehensive service index F of each station... i The calculation formula is:

[0060]

[0061] Step 4: Determine the number of bus stop stops along the route.

[0062] Based on the principles for setting up control points for bus routes, the starting and ending stations are first determined as the first and last control points. For intermediate stations, all control point plans are generated according to the selection and quantity principles.

[0063] Selection principles for bus route control points:

[0064] (1) The starting station and the terminal station of the line are the default control points. The time of the two stations plays a role in determining the departure time of the line and aligning with the on-time arrival time of the terminal station.

[0065] (2) Control points along the route should be selected at stations with important locations, such as important transportation hubs, passageway nodes, and stations with high passenger flow such as shopping malls, schools, hospitals, and office areas. These stations can provide more passengers with reliable and efficient public transportation services.

[0066] (3) To ensure that the arrival time at the control point is as punctual and reliable as possible, the station is set at the station with a small degree of variation in arrival time. This makes the arrival time of the vehicle more controllable after the arrival time at the control point is specified.

[0067] (4) The distance between control points should be kept within a reasonable range. The interval between control points should not be too close, so that the vehicle has time to make dynamic adjustments and can ensure that it arrives at the next control point on time. The interval between control points should not be too large. If the interval is too large, the vehicle's operation between control points will be too random, the stability of the vehicle during operation will not be effectively controlled, and the passengers waiting at the station will feel that the line is punctual and reliable.

[0068] Principles regarding the number of control points:

[0069] (1) Based on a survey of public transport operators and bus drivers, the number of control points should be 20%-25% of the total number of stops on the entire route.

[0070] (2) The time interval between adjacent control points should ideally be 6-12 minutes. This allows drivers to adopt appropriate strategies to ensure punctuality at the next control point while also maintaining a good overall punctuality rate for the route. Assuming an average bus speed of 15 km / h and a distance of 500-800 m between each stop, the distance between stops would be 3-6, meaning there are usually at least 2 passing stops between two control points and a maximum of 5 passing stops.

[0071] Step 5: Select the optimal control point scheme and calculate the suggested time for vehicles to reach the control point under this scheme. Step 5.1: Selection of the optimal control point scheme.

[0072] Based on the importance index and arrival time stability index of each station calculated in step 3, calculate the comprehensive evaluation index F of the station service of the kth intermediate control point in each scheme. k The overall service evaluation index F of this plan is:

[0073]

[0074] Where l is the total number of intermediate control points, F k Let F be the station service evaluation index for the kth intermediate control point in this scheme. The scheme that maximizes the comprehensive service index value F is the recommended optimal scheme for the control point of this line, as shown in Table 8.

[0075] Table 8. Examples of Recommended Mid-Route Control Point Schemes

[0076]

[0077] Step 5.2: Calculate the arrival time of the vehicles at each control station.

[0078] Based on the time granularity selected in step 2, the arithmetic mean of the historical arrival times of vehicles from the first station to each control point in the t-th time period of the same week within the analysis cycle is calculated to generate the arrival times of vehicles departing from each control point in each time period, which serve as a reference for vehicle travel, as shown in Equation 10. Where PT k To control the suggested arrival time of station k, T kj T is the arrival time of the vehicle at the k-th control station during the j-th trip. lj is the starting station and departure time of the j-th trip, and r is the total number of trips.

[0079]

[0080] The arithmetic mean of the calculated arrival times of vehicles at each control station is used as the suggested arrival time for each control station, as shown in Table 9, where the control station is labeled K.

[0081] Table 9 Examples of suggested arrival times for a specific train to various control points

[0082] Attached Figure Description

[0083] Figure 1 Schematic diagram of control points along bus routes

[0084] Figure 2 Flowchart of bus route control point setup Detailed Implementation

[0085] Implementation Plan Analysis Case: Beijing Bus Route 605

[0086] Step 1, Ground Public Transport Data Preprocessing

[0087] This case study focuses on Beijing Bus Route 605 (Hongyan Bus Terminal - Shigezhuang direction), covering the period from 5:00 AM to 11:00 PM daily. Data collected from January 14th to 18th, 2019, including bus IC card swipe data, real-time bus arrival data, and static bus route data, were used as raw data for analysis. Data preprocessing was performed, including data cleaning, data association and matching, and effective field filtering. This resulted in core fields such as IC card number, route number, vehicle number, equipment number, boarding / alighting station number, boarding / alighting identification, and transaction time. The route has 30 stops during the study period, as shown in Table 10.

[0088] Table 10. Stop Names for Beijing Bus Route 605

[0089]

[0090]

[0091] Step 2: Extraction and averaging of station entry and exit points

[0092] Using the matched public transport transaction data, calculate the total number of passengers Q boarding at each bus stop (excluding the first and last stops) between 7:30 and 8:30 from January 14th to 18th. b i，t Total number of people getting off Q a i，t This allows us to obtain the arrival and departure times at each station along the route. The average values ​​were then calculated, as shown in Table 11.

[0093] Table 11 Calculation Results of Boarding and Alighting Volume at Beijing Bus Route 605 Stations

[0094]

[0095] Step 3: Calculate the comprehensive evaluation index of bus stop services.

[0096] First, based on the passenger boarding and alighting data of each stop on bus route 605, and combined with the number of POIs in the surrounding area, the passenger flow correction coefficient P for different stops was calculated. i The importance index of the bus stops was calculated according to equation (2). Next, based on historical arrival time data, the coefficient of variation of vehicle travel time was calculated according to equations (4) and (5), serving as the stability index of bus arrival time. The importance index and the stability index of arrival time were normalized before calculating the comprehensive evaluation index of bus stop services, as shown in Table 12.

[0097] Table 12 Calculation Results of the Comprehensive Evaluation Index of Service at Beijing Bus Route 605 Stations

[0098]

[0099]

[0100] Step 4: Determine the number of bus stop stops along the route.

[0101] According to the principles for setting control points, the first and last stops are designated as control points. Secondly, based on the evaluation indicators, the total number of control points must be 20%-25% of the total number of stops, with a time interval of 6-12 minutes between adjacent control points and a distance of 3-6 stops between stops. Bus route 605 has 30 stops; considering all factors, 9 control points are needed, meaning 7 control points are required excluding the first and last stops, with 2-5 stops between adjacent intermediate control points.

[0102] Step 5: Select the optimal control point scheme and calculate the time for vehicles to arrive at each control station.

[0103] A program for calculating the corresponding comprehensive service index was designed using Python. First, based on the principles of 30 stops, 9 control stops, and 3-6 station spacing determined in step 4, all feasible schemes were enumerated, resulting in 728 different control point selection schemes. Second, the station boarding and alighting volume M was calculated using public transport transaction data. i,up +M i,off Calculate the arrival time T based on bus arrival data. i After normalization, the comprehensive service evaluation index of the control station is calculated according to formula (9); finally, the scheme with the largest evaluation index value is selected as the optimal control point selection scheme. The feasible schemes are sorted according to the comprehensive service evaluation index value as shown in Table 13. As can be seen from Table 13, the scheme with the largest index value is the 616th scheme, with a comprehensive evaluation index value of 0.502, and the corresponding control point schemes are (1,6,9,12,15,18,21,24,30).

[0104] Table 13 Feasible solutions for control points of Beijing Bus Route 605

[0105]

[0106] After obtaining the scheme with the highest comprehensive evaluation index value (scheme 616), taking the 7:30 departure bus as an example, the operating period is divided into 1-hour units. The 7:30 departure bus falls within the 7:30-8:30 time period. The average arrival time data of the five consecutive working days within this operating period is used to calculate the suggested arrival time of the vehicle at each control point. Table 14 shows the planned arrival time of bus route 605 departing at 7:30 to each control point.

[0107] Table 147:30 shows the arrival times of trains at various control points.

[0108]

[0109] The mid-route station control strategy first selects stations that are important in location, have high passenger volume, and have good arrival time stability, making it easier to control arrival time. The selected control points can make more passengers feel the stability of the line's operation.

[0110] By providing the arrival time of vehicles at control points, drivers can improve the on-time rate of vehicles arriving at each control point by adjusting the operating speed and stopping time at each station during the operation. This ensures that vehicles arrive at the terminal station on time, improves the stability of the running time between stations and the on-time rate of the line, and thus improves the operational efficiency of the line.

Claims

1. A method for determining a control site of a bus route based on a site service comprehensive index, characterized in that, Includes the following steps: Step 1: Preprocessing of ground public transport data; Preprocessing is performed on multi-source data, including data association and matching, data cleaning and filtering, and key field extraction, to obtain ground public transport transaction data containing complete travel information of travelers. Step 2: Extracting and averaging bus stop boarding and alighting data; Using bus card and QR code swiping data, based on vehicle number, boarding / alighting station number, and transaction time, and combined with bus arrival data, the boarding / alighting volume of stations with the same vehicle number and boarding / alighting station number is counted within a transaction time difference of no more than 1 minute. Among multiple data records that meet the conditions, the median of the transaction time is extracted as the corresponding time for the statistical boarding / alighting volume of that station. By selecting a time granularity, the average number of stops for the same route in the same week and time period within the analysis period is calculated by averaging the number of stops for the same route. Step 3: Calculate the comprehensive evaluation index of bus stop services; Step 3.1: Calculate the site importance index; set the location importance correction coefficient. The importance index of a site is formed by placing it before the percentage of arrivals and departures at each site. The importance index of site i within time period t is... ; Step 3.2, calculate the stability index of bus arrival time; using weekdays. Vehicles arrive at the destination within time period t. The difference between the time at each station and the vehicle's departure time from the starting station is used to calculate the vehicle's arrival time at the next station. Travel time to each station The stability of arrival time is represented by the coefficient of variation of travel time. The smaller the fluctuation in arrival time and the smaller the coefficient of variation, the higher the stability. (Weekdays) middle Stations during the time period The arrival time stability index is ; Step 3.3, index normalization and comprehensive index value calculation: The importance index and arrival time stability index of each stop along the bus route are normalized to calculate the comprehensive service evaluation index of the stops, including the station importance normalization index. Station arrival time stability normalization index and the overall service index of each station ; Step 4: Determine the number of control stops along the bus route; First, determine the starting station and the ending station as the first and last station control points; Intermediate stations are selected according to the selection and quantity principles of control points, thereby generating all control point schemes; Step 5: Select the optimal control point scheme and calculate the suggested time for the vehicle to reach the control point under this scheme; Step 5.1, Optimal control point selection; Assuming that k control points are selected from n bus stations along the route, the service comprehensive evaluation index of the scheme is calculated : ； In the formula, l represents the total number of intermediate control points. Let be the station service evaluation index for the kth intermediate control point in this scheme; select Let the comprehensive service evaluation index value of the scheme be . The largest possible solution is the recommended optimal solution for the control points of this line; Step 5.2: Calculate the arrival time of the vehicle at each control station; Calculate the arithmetic mean of the historical arrival times of vehicles from the first station to each control point in the t-th time period of the same week within the analysis cycle, and generate the arrival times of vehicles departing from each control point in each time period: ； In the formula, PT k To control the suggested arrival time of station k, T kj T is the arrival time of the vehicle at the k-th control station during the j-th trip. 1j is the starting station and departure time of the j-th trip, and r is the total number of trips.

2. The method of claim 1, wherein the method is characterized by, The rules for filtering and removing the original bus card or QR code swipe data in step 1 are as follows: (1) Remove null and duplicate data; (2) Remove transaction data where the boarding and alighting times are not on the same day; (3) Remove transaction data where the boarding time is later than the alighting time; (4) Remove records in the card swipe data where the entry and exit stations are the same; (5) Remove transaction data for certain major events, holidays, and special dates and periods; The key transaction fields for ground public transportation extracted in step 1 include: smart card number, route number, vehicle number, equipment number, boarding / alighting station number, boarding / alighting identification, and transaction time.

3. The method for determining intermediate control stops on bus routes based on a comprehensive station service index according to claim 1, characterized in that, In step 2, during the extraction of station arrival and departure data, time granularities of 15 min, 30 min, 1 h, and 1 d are selected for statistical analysis based on different scenarios and different accuracy requirements.

4. The method for determining intermediate control points of bus routes based on the comprehensive station service index according to claim 1, wherein the index normalization method used in step 3 integrates the original evaluation indicators into a comprehensive evaluation index—the comprehensive station service index, and uses the comprehensive service index to assess the suitability of a station as an intermediate control point for bus routes.

5. The method for determining intermediate control points of bus routes based on a comprehensive station service index according to claim 1, wherein the selection principles for bus route control points in step 4 include: (1) The starting station and the terminal station of the line are the control points by default; (2) Control points along the route should be selected at stations with relatively important locations; (3) To ensure that the arrival time at the control point is as punctual and reliable as possible, it should be set at a station with a small degree of variation in arrival time; (4) The spatial intervals between control points should be kept within a reasonable range; The principles for the number of bus route control points include: (1) The number of control points shall be 20%-25% of the total number of stations along the entire line; (2) The time interval between adjacent control points is 6-12 minutes.

6. The method for determining intermediate control stops on bus routes based on a comprehensive station service index according to claim 1, characterized in that, Step 3.1 specifically includes: the more numerous and prevalent the shopping malls, schools, hospitals, and office facilities within an 800-meter radius of the site, the more prominent the relative importance of the site's location; setting a location importance correction coefficient. The importance index of a site is formed by placing it before the percentage of arrivals and departures at each site; the importance index of site i within time period t. The calculation method is as follows: ； in, is the average number of stops for the t-th time period; n is the total number of stops on the bus route; This is a site location importance correction coefficient, reflecting the proportion of POIs that take weight into account, used to measure the importance of a site along the entire line; its calculation method is as follows: ； In the formula, For the site The number of surrounding POIs of type f; F represents the number of POIs of type f around the entire line; F is the total number of POI categories. The weighting coefficient for the f-th type of POI is determined based on the POI's category, land area, and the importance of public transport services.

7. The method of claim 6, wherein the method further comprises: determining a number of stops of the bus route; and determining a number of stops of the bus route that are within a predetermined distance from the candidate stop. Step 3.2 specifically includes: using working days Vehicles arrive at the destination within time period t. The difference between the time at each station and the vehicle's departure time from the starting station is used to calculate the vehicle's arrival time at the next station. Travel time to each station The stability of arrival time is represented by the coefficient of variation of travel time. The smaller the fluctuation in arrival time and the smaller the coefficient of variation, the higher the stability. (Workdays) middle Stations during the time period Arrival time stability index Calculation method: ； wherein is the total number of stations for the line, is the average travel time; To make the average processing, get the average arrival time stability index of the station in the working day t period : ； In the formula, Is isite on weekdays middle Arrival time stability indicators within a given time period; It is the first One working day.

8. The method for determining intermediate control stops on bus routes based on a comprehensive station service index according to claim 7, characterized in that, Step 3.3 specifically includes: normalizing the importance index and arrival time stability index of each station along the bus route, and calculating the comprehensive service evaluation index of the station; Site Importance Normalization Index : ； In the formula, It is the lowest site importance metric value. It is the highest site importance metric value; Site-to-site time stability normalization index : ； In the formula, is the minimum arrival time stability index value, is the maximum arrival time stability index value; Time control points should be set at stations with high importance and strong arrival time stability. Based on this, the concept of a comprehensive service evaluation index for public transport stations is proposed, which is reflected as the product of the station importance index and the reciprocal of the station arrival time stability index. The comprehensive service index for each station... The calculation formula is: 。

Images