A commuting corridor identification and problem diagnosis method and system for time-poor people

Abstract

The application discloses a commuting corridor identification and problem diagnosis method and system for time-poor people, comprising the following steps: acquiring various data of a target city, screening effective users, and calculating the residence and employment of the effective users; constructing an index system of time-poor people portraits, and screening time-poor people; performing spatial correlation on the time-poor people and land use data through a space-time data integration platform to obtain a time-poor commuting corridor feature set of the target city; identifying a commuting peak of the time-poor people in the target city; constructing a constraint index system and a suggestion index system; performing constraint index examination on the time-poor commuting corridor, and outputting a result if the examination fails; if the examination passes, suggestion index evaluation is continuously performed, and a result is outputted; and the application is helpful for quickly identifying characteristic problems of the current urban traffic situation, and can provide time-dimension research results and basis for urban planning.

Description

TECHNICAL FIELD

[0001] The application relates to a commuting corridor identification and problem diagnosis method and system, in particular to a commuting corridor identification and problem diagnosis method and system for time-poor people, and belongs to the technical field of urban traffic. BACKGROUND

[0002] Job-housing is an important core function of a city, and the commuting behavior of residents based on this is a necessary travel behavior. As an important observation dimension of urban daily traffic behavior, commuting intuitively reflects the effectiveness of urban traffic construction and management. In actual work, the problem diagnosis of the current commuting situation of a city is often focused on the calculation of single-dimensional static indicators, which is difficult to cope with the dynamic changes of urban commuting problems, and cannot identify the urban commuting corridors with potential transformation potential.

[0003] On the other hand, the existing urban commuting problem diagnosis method mainly focuses on finding and solving the universal urban commuting problems, but it is difficult to find the commuting disadvantaged groups in the city and take targeted measures to cope with them, so as to promote social fairness and resident happiness. Based on this, the application introduces the concept of time-poor people. The concept of time poverty was first proposed by economist Vickery in the 1970s, which refers to the minimum time that a family needs in addition to income to maintain a non-poor consumption level. SUMMARY

[0004] The purpose of the application is to provide a commuting corridor identification and problem diagnosis method and system for time-poor people, which can improve the efficiency of urban traffic status problem identification.

[0005] Technical scheme: The application provides a commuting corridor identification and problem diagnosis method for time-poor people, which comprises the following steps:

[0006] S1: Obtain the commuting monitoring big data of a target city, location-based service (LBS) data, land use data, POI data, road network data and administrative division spatial data, perform data preprocessing on the LBS data, filter effective users, and calculate the residence and employment of the effective users, and assign a block number to each residence and employment, and uniformly import the preprocessed data and the latitude and longitude of the residence and employment into a time-space data integration platform of the target city; the commuting monitoring big data comprises average vehicle running speed of each road section in each time period and bus passenger flow and rated load flow of each road section in each time period;

[0007] S2: Construct an index system of time-poor people portrait, calculate the index weight based on the analytic hierarchy process, and construct a time poverty index formula; and combine the time-space data integration platform to calculate the time poverty index of the effective users and filter the time-poor people;

[0008] S3: Spatially correlate the time-poor population with land use data through the spatio-temporal data integration platform, generate shortest commuting paths based on the shortest path algorithm, and generate a buffer zone of the shortest commuting paths to obtain the time-poor commuting corridor feature set of the target city;

[0009] S4: Identify the commuting peak of the time-poor population in the target city, and store the identified commuting peak time in the spatio-temporal data integration platform; build a constraint index system and a suggestion index system as the evaluation and assessment criteria for the diagnosis of the time-poor commuting corridor problem; the suggestion index system includes static indicators and dynamic indicators;

[0010] S5: Test the constraint index evaluation of the time-poor commuting corridor, if the test fails, output the result; if the test passes, continue to evaluate the suggestion index, and output the constraint index evaluation result and the suggestion index evaluation result on the spatio-temporal data integration platform.

[0011] Further, the preprocessing method of the LBS data in step S1 is:

[0012] Identify the stay point, cluster the trajectory points that do not move within a predetermined time or move within a predetermined range in the LBS data, identify the average center of the cluster as the stay point, and generate the stay point information, the stay point information includes user, latitude and longitude, stay duration and stay start and end time; the average center refers to a point returned for each stay, which represents the average center in time and distance, and the stay point information has a time type interval;

[0013] Clean the data and select valid users, specifically: aggregate all identified stay points, count the number of stay days for each user, delete user data with less than half of the total number of stay days, mark the remaining users as valid users, and associate the original all stay point data with the valid users to obtain valid user stay data.

[0014] Further, the method for calculating the residence and employment of the valid user in step S1 is:

[0015] Let the time-space anchoring property of the stay point represent the number of stay days, the duration of the stay, and the stability of the stay period of the user stay point in a period of time, and let the time-space anchoring property index M be greater, representing that the user stay point has stronger time-space anchoring property; the time-space anchoring property index M of the user i in the jth land unit in the land use data ij The calculation formula is as follows:

[0016]

[0017] Where, D ijDij represents the total number of days that user i stays in the jth land unit i T represents the total number of days that user i stays in all land units ij Tij represents the total length of time that user i stays in the jth land unit i T represents the total length of time that user i stays in all land units, stay time entropy q ij Dij represents the discrete degree of user i's daily stay time period in land j, and the specific formula is:

[0018]

[0019] When |x n_ij -x n_ij+1 |≤12, |x n_ij -x n_ij+1 |>12, When |y n_ij -y n_ij+1 |≤12, |y n_ij -y n_ij+1 |>12,

[0020] wherein, Dij represents the Euclidean distance between point P n_ij and P n+1_ij , P n_ij (x n_ij ,y n_ij ) represents the data discrete point of user i in the jth land unit on the nth day, x n_ij is the start time of the longest daily stay period of user i in the jth land unit, y n_ij is the end time of the longest daily stay period of user i in the jth land unit, x n_ij and y n-ij are rounded to the integers in the set [1, 24], and s_ij is the total number of days that user i stays in the jth land unit;

[0021] The residence is identified by the following method: calculating the spatiotemporal anchoring data set M alldays_i of all land units that user i has stayed in, taking the maximum value in M alldays_i and marking the corresponding land as the residence of user i;

[0022] The work place is identified by the following method: calculating the spatiotemporal anchoring data set M workdays_i of all land units that user i has stayed in on weekdays, taking the maximum value in M workdays_i , checking whether the corresponding land is different from the residence, and if so, marking the corresponding land as the work place of user i; if not, taking Mworkdays_i The land marked with the second largest value is the user's working place.

[0023] Further, the spatio-temporal data integration platform in step S1 comprises basic spatial data and spatio-temporal trajectory data, the basic spatial data is obtained from a public website, including land use data, POI data, road network data and administrative division spatial data; the user spatio-temporal trajectory data comprises effective user stay data, user residence and user working place data and public website obtained commuting monitoring big data.

[0024] Further, the index system for constructing the time poverty population portrait in step S2 comprises:

[0025] S21: calculating the basic indexes of the index system for constructing the time poverty population portrait, forming a time poverty population portrait index data set; the basic indexes comprise a commuting cost dimension, a working time length dimension and a facility acquisition dimension;

[0026] The commuting cost dimension comprises commuting time length and commuting distance, the commuting time length is obtained from the effective user stay data, taking the average time length of the effective user commuting between the residence and the working place, denoted as T C , and the commuting distance is obtained from the effective user residence and working place data and land use data, taking the straight line distance between the effective user residence and the working place, denoted as D rw ;

[0027] The working time length dimension comprises single-day working time length and single-week working day number, the single-day working time length is obtained from the effective user stay data, taking the average time length of the effective user staying at the working place, denoted as T w , and the single-week working day number is obtained from the effective user stay data, taking the average number of days of the effective user appearing in the working place stay record and the stay time length being greater than a preset time length per week, denoted as d w ;

[0028] The facility acquisition dimension comprises life facility number, life facility accessibility and life facility diversity, the life facility number is obtained from the format POI data, land use data and user residence and working place data; the life facility classification comprises health management, commercial service and sports fitness, and the total number of each type of life facility within the extended boundary of the effective user residence and working place block boundary expanded by a preset distance is counted, denoted as n f ;

[0029] The life facility accessibility is obtained from the format POI data, land use data, user residence and working place data, and the specific formula is as follows:

[0030]

[0031] wherein, ki is the kth point in the i-type living facility, is the distance value between the central point of the residential or working place and the kth point in the i-type facility, n i is the total number of i-type facilities that can cover the plot, t i is the average straight-line distance between the geometric center of the residential or working place plot and the i-type facility, D a is the living facility accessibility index of the user's residential or working place;

[0032] The living facility diversity is obtained from format POI data, land use data, and user residential and working place data, and the specific formula is as follows:

[0033]

[0034] wherein, P i is the proportion of the total number of living facilities in the user's residential or working place, s is the total number of living facilities covering the user's residential or working place, and H is the living facility diversity index of the user's residential or working place; i

[0035] S22: normalize the indicators of commuting time T C , commuting distance D rw , daily working time T w , weekly working days d w , number of living facilities n f , living facility accessibility index D a , and living facility diversity index H in the time poverty population portrait index data set, and the corresponding data after processing are denoted as T Cn , D rwn , T wn , d wn , n fn , D an , and H n , and the calculation formula of the time poverty index TPI is:

[0036] TPI = x1T Cn + x2D rwn + x3T wn + x4d wn + x5n fn + x6D an + x7H n

[0037] wherein, x1, x2, …, x7 are the corresponding weight coefficients of each index, which are obtained by the analytic hierarchy process.

[0038] ​Further, the step S3 comprises: selecting Dijkstra algorithm for calculating the shortest path from the residence to the employment, setting the residence of the time poverty population as the start and end point, setting the work place of the time poverty population as the terminal point, constructing the road network model in the shortest path algorithm from the road network data in the spatio-temporal data integration platform, calculating the shortest path according to the Dijkstra algorithm, superimposing all the shortest paths into the same data file, and recording the content of the data file as the time poverty population commuting corridor.

[0039] Further, the step S4 comprises:

[0040] S41: obtaining the start and end time of the user commuting behavior based on the effective user stay data in the spatio-temporal data integration platform, and verifying the early and late peak time of the time poverty population commuting based on the commuting monitoring big data, and storing the early and late peak time of the time poverty population commuting to the target city spatio-temporal data integration platform;

[0041] S42: presetting the evaluation standard, constructing the constraint index system and storing it to the spatio-temporal data integration platform for evaluating the target city; the constraint index comprises: the average speed index of motor vehicles and the average speed index of regular buses, the average speed index of motor vehicles is that the average speed of motor vehicles on the expressway and the main road during the early and late peak period is not less than 30km / h and 20km / h respectively; the average speed index of regular buses is that the average speed of regular buses on the time poverty population commuting corridor during the early and late peak period on weekdays is not less than 15km / h;

[0042] S43: taking the static index and the dynamic index as the suggestive index and storing it to the spatio-temporal data integration platform under the time dimension of the early and late peak period of commuting; the static index refers to the index related to the time point, reflecting the commuting information of the target city at a certain time and having stability in time, including the coverage commuting proportion of the rail station and the average one-way commuting distance, the coverage commuting proportion of the rail station refers to the proportion of the number of residence and employment within the preset radius of the rail station to the total number of residence and employment in the whole city, the average one-way commuting distance refers to the average distance of the shortest path of one-way commuting of the time poverty population; the dynamic index refers to the index related to the time period, reflecting the commuting information of the target city in a certain time period and having timeliness, including the bus congestion degree and the business proportion of life facilities, the specific time period selected is the early and late commuting peak time of the target city obtained in step S41; wherein, the bus congestion degree refers to the proportion of the bus passenger flow on the time poverty population commuting corridor during the early and late peak period on weekdays to the rated passenger capacity; the business proportion of life facilities refers to the proportion of the facilities in the open state to the total number of life facilities in the time poverty population commuting corridor feature set during the early and late peak period on weekdays.

[0043] Further, the step S41 comprises:

[0044] S411: Extracting the sub-period commuting data set of each user in the time poverty population in each hour of a day based on the effective user stay point data and the user residence and employment data in the space-time data integration platform

[0045] {P1, P2, P3, …, P 24}, wherein P={user, residence plot number, work plot number, commuting behavior time under the time slice}, 24 is the time slice, and the slice length is one hour;

[0046] S412: Marking the time corresponding to the slice with the longest total commuting time of all time poverty populations under each time slice from zero to twelve as the morning peak, and marking the time corresponding to the slice with the longest total commuting time of all time poverty populations under each time slice from twelve to twenty-four as the evening peak based on the sub-period commuting data set;

[0047] S413: Extracting the average vehicle running speed v1 in the morning peak and the average vehicle running speed v2 in the evening peak marked in step S412 from the space-time data integration platform, and regarding the test as passed if v1 and v2 belong to the last three positions in the average vehicle running speed of each period.

[0048] Further, the step S5 comprises:

[0049] S51: Taking the administrative division vector boundary as the statistical unit, outputting the distribution area and quantity of the residence and work of the time poverty population in each statistical unit, and marking the corresponding time poverty population quantity; and calculating the mileage of the time poverty commuting corridor in each statistical unit, and marking the time poverty commuting corridor in the region;

[0050] S52: Calculating the constraint index of the target city in the space-time data integration platform, marking the constraint index item that fails the examination and the gap between the index item and the examination standard if the constraint index examination fails, ending the examination and outputting the result; and continuing step S53 if the constraint index examination passes.

[0051] S53: Calculating the evaluation of the target city suggestion index in the space-time data integration platform, and outputting the constraint index examination result and the suggestion index evaluation result of the target city, wherein the suggestion index evaluation result specifically includes the index name and the index suggestion value or the index requirement.

[0052] Based on the same inventive concept, the application also provides a commuting corridor identification and problem diagnosis system for the time poverty population, comprising:

[0053] A preprocessing module is configured to acquire commuting monitoring big data of a target city, location-based service (LBS) data, land use data, POI data, road network data and administrative division spatial data, pre-process the LBS data, screen effective users, calculate the residence and employment locations of the effective users, and assign a plot number to each residence and employment location, and then import the pre-processed data and the latitude and longitude of the residence and employment locations into a time-space data integration platform of the target city; the commuting monitoring big data includes average vehicle running speed of each road section in each time period and bus passenger flow and rated load flow of each road section in each time period;

[0054] A screening module is configured to construct an index system for the portrait of the time poverty population, calculate the index weight based on the analytic hierarchy process and construct a time poverty index formula, and calculate the time poverty index of the effective users in combination with the time-space data integration platform to screen the time poverty population;

[0055] An identification module is configured to perform spatial correlation between the time poverty population and the land use data through the time-space data integration platform, calculate and generate a shortest commuting path based on the shortest path algorithm, generate a buffer zone of the shortest commuting path, and obtain the time poverty commuting corridor features of the target city.

[0056] An index establishing module is configured to identify the commuting peak of the time poverty population of the target city and store the identified commuting peak time into the time-space data integration platform, construct a constraint index system and a suggestion index system as the evaluation and assessment standards for the diagnosis of the time poverty commuting corridor, and the suggestion index system includes static indexes and dynamic indexes.

[0057] An evaluation and test module is configured to perform constraint index evaluation and test in combination with the time poverty commuting corridor, output a result if the test fails, and continue to perform suggestion index evaluation if the test passes, and output the constraint index evaluation result and the suggestion index evaluation result on the time-space data integration platform.

[0058] Beneficial effects: compared with the prior art, the present application has the following remarkable advantages: 1, the present application takes the identification and problem diagnosis of the time-poor population commuting corridor as the observation dimension of the urban traffic status characteristic problem, compared with the existing urban traffic status problem identification method, the present method effectively reduces the number of observation samples, identifies the problem of the time-poor population commuting corridor with potential problems, improves the efficiency of urban traffic status problem identification, and can mark the commuting corridor with optimization potential and output the index evaluation result and reference value, facilitating the subsequent commuting corridor optimization work; 2, the present application proposes a time-poor population commuting corridor problem diagnosis method combining dynamic and static two dimensions after examination and evaluation, and divides the evaluation indexes into constraint indexes and recommended indexes, forming a rigid and flexible evaluation method, which can more flexibly cope with the actual situation; 3, the present application realizes dynamic index evaluation in the time dimension during the commuting peak period, can effectively identify the potential commuting problems under the condition of road traffic resource shortage, realizes the targeted physical examination under the extreme commuting condition, and greatly improves the accuracy of urban commuting physical examination; 4, the present application proposes a residence identification method based on the space-time anchoring of the stopping point, compared with the existing common method, the method proposed by the present application does not need land use data, but focuses on the stopping time and stopping stability of the user stopping point, reduces the identification steps of the residence, and significantly improves the identification accuracy; 5, the present application constructs a space-time data integration platform with data storage, data analysis, index interaction and result output, effectively solves the problems of disordered data information and difficult connection between data information processes in the present situation, and greatly reduces the data management difficulty. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 The method flowchart of the present application is shown in the figure;

[0060] Figure 2 The time-poor commuting corridor evaluation flowchart of the embodiment of the present application is shown in the figure;

[0061] Figure 3 The time-poor commuting corridor identification diagram of the embodiment of the present application is shown in the figure;

[0062] Figure 4 The track station surrounding coverage commuting proportion diagram of the embodiment of the present application is shown in the figure;

[0063] Figure 5 The conventional bus congestion degree diagram of the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings and the embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0065] The technical solutions of the present application are explained in detail by taking the identification and problem diagnosis of the commuting corridor of the time-poor population in a certain city as an example. As shown in FIG. 1, the identification and problem diagnosis method of the commuting corridor of the time-poor population in the embodiment comprises the following steps. Figure 1

[0066] S1: Obtain the commuting monitoring big data of the target city, the location-based service (LBS) data, the land use data, the POI data, the road network data and the administrative division spatial data, perform data preprocessing on the LBS data, filter the effective users, calculate the residence and the employment of the effective users, and assign a block number to each residence and employment, and then import the preprocessed data and the longitude and latitude of the residence and the employment into the spatio-temporal data integration platform of the target city; the commuting monitoring big data comprises the average vehicle running speed of each road section in each time period and the passenger flow and the rated load flow of each road section in each time period.

[0067] S2: Construct an index system of the time-poor population portrait, calculate the index weight based on the analytic hierarchy process, and construct a time-poor index formula; and calculate the time-poor index of the effective users in combination with the spatio-temporal data integration platform, and filter the time-poor population.

[0068] S3: Perform spatial correlation on the time-poor population and the land use data through the spatio-temporal data integration platform, calculate and generate the shortest commuting path based on the shortest path algorithm, generate the buffer zone of the shortest commuting path, and obtain the time-poor commuting corridor feature set of the target city.

[0069] S4: Identify the commuting peak of the time-poor population in the target city, store the identified commuting peak time into the spatio-temporal data integration platform, construct a constraint index system and a suggestion index system as the evaluation and assessment standards for the problem diagnosis of the time-poor commuting corridor, and the suggestion index system comprises static indexes and dynamic indexes.

[0070] S5: Test the constraint index evaluation in combination with the time-poor commuting corridor, if the test fails, output the result, if the test passes, continue to evaluate the suggestion index, and output the constraint index evaluation result and the suggestion index evaluation result on the spatio-temporal data integration platform.

[0071] Specifically, the detailed process of step S1 is as follows.

[0072] ​Step S11: Preprocessing of LBS data.

[0073] Step S111: Stay point identification. Cluster the trajectory points that have no movement within 10 minutes or move within 100 meters in the LBS data, identify the average center as a stay point, and generate information of the stay point (including user id, longitude and latitude, stay duration, stay start and end time). The average center refers to a point returned for each stay, which represents the average center in time and distance, and the generated elements will have a time type interval.

[0074] Step S112: Data cleaning and effective user screening. Summarize all identified stay points, count the stay days of each user ID, delete the user ID data with stay days less than half of the total days, mark the remaining user IDs as effective users, screen out the effective user IDs, and associate the original all stay point data according to the effective user IDs to obtain the effective user stay data.

[0075] Step S12: Identification of the residence of the target city user ID.

[0076] Step S121: Calculation of stay point spatio-temporal anchoring index. The spatio-temporal anchoring of the stay point represents the number of stay days, the stable and persistent stay duration, and the stable stay period of the user stay point within a period of time. The strength of the spatio-temporal anchoring of the stay point is represented by the spatio-temporal anchoring index M, and the larger the index M, the stronger the spatio-temporal anchoring of the user stay point. M is calculated based on the stay time entropy q, the stay duration T, and the stay days D. The spatio-temporal anchoring index M is proportional to the stay days and the stay duration, and inversely proportional to the stay time entropy q. The specific formula is:

[0077]

[0078] wherein D ij represents the total stay days of user i in the jth land unit, D i represents the total stay days of user i in all land units, T ij represents the total stay duration of user i in the jth land unit, T i represents the total stay duration of user i in all land units, the stay time entropy q ij represents the dispersion degree of user i in the daily stay time period of plot j, and the specific formula is:

[0079]

[0080] When |x n_ij -x n_ij+1 |≤12, |x n_ij -x n_ij+1 |>12, when |y n_ij -y n_ij+1 when |y when |y n_ij -y n_ij+1 when |y

[0081] wherein, denotes the Euclidean distance between point P n_ij and P n+1_ij , P n_ij (x n_ij , y n_ij ) represents the data discrete point of user i in the jth land unit on the nth day, x n_ij is the starting time of the longest daily stay period of user i in the jth land unit, y n_ij is the ending time of the longest daily stay period of user i in the jth land unit, x n_ij and y n_ij are rounded to the integer in the set [1, 24], and s_ij is the total number of days user i stays in the jth land unit.

[0082] Step S122: residence identification. The residence refers to a specific land unit that satisfies high spatio-temporal anchoring of the user on weekdays and non-weekdays. The specific identification method is as follows: calculate the spatio-temporal anchoring data set M alldays_i of all land units where user i has stayed in the observation data, take the maximum value in M alldays_i , and mark the corresponding land as the residence of the user.

[0083] Step S123: work place identification. The work place refers to a specific land unit that satisfies high spatio-temporal anchoring of the user on weekdays and is different from the residence. The specific identification method is as follows: calculate the spatio-temporal anchoring data set M workdays_i of all land units where user i has stayed on weekdays in the observation data, take the maximum value in M workdays_i , check whether the land marked by the maximum value is different from the residence, if different, mark the corresponding land as the work place of the user; if the same, mark the land corresponding to the second largest value in M workdays_i as the work place of the user.

[0084] Step S2: the detailed process is as follows:

[0085] Step S21: constructing the index system of the time poverty population portrait.

[0086] Basic index calculation. The basic index of the index system of the time poverty population portrait includes three dimensions of commuting cost dimension, working time dimension, and facility acquisition dimension, forming a time poverty population portrait index data set. The specific index information is shown in Table 1:

[0087] Table 1 Time poverty population portrait index system table

[0088]

[0089]

[0090] Among them, the buffer zone processing is to expand the user's residence and work block boundary outward according to the 15-minute life circle radius 800m determined in the Notice on the Issuance of the Complete Residential Community Construction Guide issued by the Ministry of Housing and Urban-Rural Development.

[0091] Step S22 Time poverty index calculation. The indicators of commuting time T C , commuting distance D rw , daily working time T w , weekly working days d w , number of life facilities n f , life facility accessibility index D a , life facility diversity index H in step S2-11 are normalized in the time poverty population portrait index data set, denoted as T Cn , D rwn , T wn , d wn , n fn , D an , H n , and the calculation formula of the time poverty index TPI is:

[0092] TPI = x1T Cn + x2D rwn + x3T wn + x4d wn + x5n fn + x6D an + x7H n

[0093] Among them, x1, x2, …, x7 are index corresponding weight coefficients, which are obtained by analytic hierarchy process. The specific method is: randomly select 100 residents in the study area to evaluate the importance of the index system of time poverty population portrait, construct a hierarchical evaluation structure model of time poverty index, then construct and calculate the judgment matrix, and calculate the weight coefficient after normalization.

[0094] Step S3 detailed process as follows:

[0095] Calculate the shortest path from residence to workplace. Dijkstra's algorithm is selected for calculating the shortest path. The study user's residence is set as the start and end point, and the study user's workplace is set as the final point. A road network model for the shortest path algorithm is constructed from the road network data in the spatiotemporal data integration platform described in step S1. The shortest path is calculated according to Dijkstra's algorithm, and all paths are superimposed to generate a commuting corridor for time-impoverished individuals. Figure 3 As shown.

[0096] The detailed process of step S4 is as follows:

[0097] Step S41: Identification of peak commuting times for time-impoverished individuals. Based on valid user dwell time data from the spatiotemporal data integration platform, obtain the start and end times of user commuting behavior, identify the morning and evening peak commuting times for time-impoverished individuals, and verify this based on big data from commuting monitoring. Store the morning and evening peak commuting times of time-impoverished individuals in the target city's spatiotemporal data integration platform.

[0098] Step S411: Extraction of Time-Segmented Commuting Dataset. Based on valid user dwell time data and user residence / workplace data from the spatiotemporal data integration platform, extract time-segmented commuting datasets {P1, P2, P3, ..., P} for each user in the time-poor population for each hour of the day. 24}, where P = {User ID, Residential Plot Number, Work Plot Number, Commuting Time in this Time Slice}, 24 is the time slice, and the slice length is one hour.

[0099] Step S412: Summarize commuting times by time slot. Based on the time-slot commuting dataset, summarize the total commuting time of all time-poor individuals under each time slice. Mark the slice with the longest total commuting time from midnight to 12:00 as the morning peak, and the slice with the longest total commuting time from 12:00 to 24:00 as the evening peak.

[0100] Step S413 Commute Peak Identification and Verification. Obtain the average vehicle speeds during different time periods along the commuting corridors for time-impoverished individuals from the spatiotemporal data integration platform. Extract the average vehicle speed v1 during the morning peak and v2 during the evening peak, as marked in step S4-12. If v1 and v2 are the last three digits of the average vehicle speed for each time period, the verification is considered successful.

[0101] Step S42: Assessment of binding indicators. The binding indicators of the target city are examined to verify whether they meet the requirements. The selected indicators are based on the "Basic Indicator System for Urban Health Examination (Trial)" and the "Commuting Monitoring Report of Major Chinese Cities," and the assessment results are stored in the target city's spatiotemporal data integration platform.

[0102] A binding indicator system was constructed based on the "Basic Indicator System for Urban Physical Examination (Trial Implementation)" and stored in a spatiotemporal data integration platform. The system assesses whether each indicator in the target city meets the thresholds in the standard, and cities that meet all assessment criteria are marked as meeting the binding indicator standards. The assessment indicators include: average speed of motor vehicles and average speed of regular public transport. Specifically, the average speed of motor vehicles during morning and evening peak hours on expressways and arterial roads is no less than 30 km / h and 20 km / h, respectively. The average speed of regular public transport is specifically no less than 15 km / h on the commuting corridors for impoverished populations during weekday morning and evening peak hours.

[0103] Step S43: Evaluation of Recommended Indicators. Based on both static and dynamic indicators, and within the time dimension of peak commuting hours, cities that meet the requirements for binding indicators are assessed for problems. The selected indicators are determined with reference to the "Evaluation Standards for Green Travel Creation Action" and the "Management Measures for National Public Transport City Construction Demonstration Projects," and are combined with the current situation of the target cities. The evaluation results are then stored in the target city's spatiotemporal data integration platform.

[0104] A suggested indicator system is constructed and stored in a spatiotemporal data integration platform. This suggested indicator system specifically includes static and dynamic indicators. Static indicators refer to those related to a specific point in time, reflecting commuting information of the target city at a given moment and exhibiting temporal stability. These include the commuting ratio around rail stations and the average one-way commuting distance. The commuting ratio around rail stations refers to the proportion of residences and workplaces within a preset radius of a rail station to the total number of residences and workplaces throughout the city. The preset distance is defined as the 15-minute living circle radius of 800m as specified in the "Notice on Issuing the Guidelines for the Construction of Complete Residential Communities" issued by the Ministry of Housing and Urban-Rural Development. Figure 4 As shown; the average one-way commuting distance refers to the average distance of the shortest one-way commuting path among the time-poor population; the dynamic indicators refer to indicators that are related to a specific time period, reflect the commuting information of the target city during a certain time period, and have timeliness, including the congestion level of regular public transportation and the operating ratio of living facilities. The specific time period selected is the morning and evening commuting peak hours of the target city obtained in step S41; wherein, the congestion level of regular public transportation refers to the proportion of passenger flow on regular public transportation to the rated passenger capacity on the commuting corridor of the time-poor population during the morning and evening peak hours on weekdays, such as Figure 5 As shown; the operating ratio of living facilities refers to the proportion of living facilities that are open during the morning and evening peak hours on weekdays to the total number of facilities in the poverty commuting corridor.

[0105] like Figure 2 As shown, the detailed process of step S5 is as follows:

[0106] S51: taking administrative division vector boundary as a statistical unit, outputting distribution areas and quantities of time poverty population living places and working places in each statistical unit, and marking corresponding time poverty population quantities; and calculating mileage of time poverty commuting corridors in each statistical unit, and marking time poverty commuting corridors in a region.

[0107] S52: calculating constraint indexes of a target city in a spatio-temporal data integration platform, if constraint index examination fails, marking constraint index items failing the examination and a gap between the index items and examination standards, and ending result output; if constraint index examination passes, continuing step S53;

[0108] S53: calculating a suggestion index evaluation of the target city in the spatio-temporal data integration platform, and outputting constraint index examination results and the suggestion index evaluation results of the target city, wherein the suggestion index evaluation results specifically include index names and index suggestion values or index requirements.

[0109] Based on the same inventive concept, the embodiment also provides a commuting corridor identification and problem diagnosis system for time poverty population, comprising:

[0110] A preprocessing module is configured to acquire commuting monitoring big data, location-based service (LBS) data, land use data, POI data, road network data and administrative division spatial data of a target city, perform data preprocessing on the LBS data, filter effective users, calculate living places and employment places of the effective users, and assign a land block number to each living place and employment place, and uniformly import the preprocessed data and the longitude and latitude of the living places and the employment places into a spatio-temporal data integration platform of the target city; the commuting monitoring big data includes average vehicle running speed of a road section in a time period, bus passenger flow and rated load flow of the road section in the time period.

[0111] A screening module is configured to construct an index system of a time poverty population portrait, calculate index weights based on an analytic hierarchy process and construct a time poverty index formula, and calculate a time poverty index of the effective users in combination with the spatio-temporal data integration platform, and screen time poverty population.

[0112] An identification module is configured to perform spatial correlation between the time poverty population and the land use data through the spatio-temporal data integration platform, calculate and generate a shortest commuting path based on a shortest path algorithm, generate a buffer zone of the shortest commuting path, and obtain a time poverty commuting corridor feature set of the target city.

[0113] The index establishing module is configured to identify a time poverty population commuting peak of a target city, and store the identified commuting peak time to the space-time data integration platform; a constraint index system and a suggestion index system are constructed as evaluation and assessment standards for diagnosing the time poverty commuting corridor problem; the suggestion index system includes static indexes and dynamic indexes;

[0114] The examination and verification module is configured to perform examination and verification of the constraint index examination in combination with the time poverty commuting corridor, output a result if the examination and verification fails, continue to perform the suggestion index evaluation if the examination and verification passes, and output the constraint index examination result and the suggestion index evaluation result on the space-time data integration platform.

Claims

1. A method for identifying and diagnosing problems of a commuting corridor for time-poor people, characterized in that, The application comprises the following steps: S1: obtaining commuting monitoring big data of a target city, location-based service (LBS) data, land use data, POI data, road network data and administrative division spatial data, pre-processing the LBS data, screening effective users, calculating the residence and employment of the effective users, assigning a land block number to each residence and employment, and importing the pre-processed data and the latitude and longitude of the residence and employment into a time-space data integration platform of the target city; the commuting monitoring big data comprises average vehicle running speed of a road section in a time period and bus passenger flow and rated load flow of a road section in a time period; The method for calculating the residence and employment of the effective users is as follows: The space-time anchoring property of the user staying point represents that the staying days of the user staying point in a period of time, the staying duration is stable and persistent, and the staying time period is stable. The space-time anchoring property index is set , The greater the space-time anchoring property is, the stronger the space-time anchoring property of the user staying point is. The user The space-time anchoring property index of the first land unit in the land use data The calculation formula is as follows: ， wherein, representing the user the total number of days spent by the user on the i-th land unit, representing the user the total number of days spent by the user on all land units, representing the user the total length of time spent by the user on the i-th land unit, representing the user the total length of time spent by the user on all land units, representing the user the total length of time spent by the user on all land units, the time spent entropy representing the user the degree of dispersion of the daily time spent by the user on the plot , which is specifically formulated as: ， When ; ; ; ; when ; ; ; ; wherein, the Euclidean distance between the point and , ( , ) represents the data discrete points of the user in the first plot unit on the first day, the start time of the longest period of stay of the user in the first plot unit per day, the end time of the longest period of stay of the user in the first plot unit per day, and the integer in the set after rounding off, the total number of days of stay of the user in the first plot unit. Identifying a residence, the method comprising: calculating a user spatial anchoring data set for all land units visited by the user , taking the maximum value and marking the corresponding land as the user's residence; The method for identifying the workplace is: calculating the user's location. Spatiotemporal anchoring dataset of all land use units visited during weekdays ,Pick The maximum value is used to check whether the application location marker is different from the residence location. If it is different from the residence location, the corresponding land use marker is marked as the user's work location; if they are the same, then... The land use marker corresponding to the second largest value is the user's work location; S2: constructing an index system of time poverty population portraits, calculating the index weight based on the analytic hierarchy process and constructing a time poverty index formula; and calculating the time poverty index of the effective users and screening the time poverty population in combination with the time-space data integration platform; S3: performing spatial correlation on the time poverty population and the land use data through the time-space data integration platform, calculating and generating a shortest commuting path based on the shortest path algorithm, generating a buffer zone of the shortest commuting path, and obtaining time poverty commuting corridor features of the target city; S4: identifying a commuting peak of the time poverty population in the target city, storing the identified commuting peak time into the time-space data integration platform, constructing a constraint index system and a suggestion index system as evaluation and assessment standards for diagnosing the time poverty commuting corridor, and the suggestion index system comprises static indexes and dynamic indexes; S5: testing the time poverty commuting corridor in combination with the constraint index, if the test fails, outputting a result, if the test passes, continuing to evaluate the suggestion index, and outputting the constraint index test result and the suggestion index evaluation result on the time-space data integration platform.

2. The method of claim 1, wherein the method is characterized by, The pre-processing method of the LBS data in step S1 is as follows: identifying a stay point, clustering trajectory points that do not move within a preset time or move within a preset range in the LBS data, identifying the average center of the clustering as a stay point, and generating stay point information, the stay point information comprising a user, latitude and longitude, stay duration and stay start and end time; the average center refers to a point returned for each stay, used to represent the average center in time and distance, and the stay point information has a time type interval; cleaning the data and screening effective users, specifically: aggregating all identified stay points, counting the stay days of each user, deleting user data with stay days less than half of the total days, marking the remaining users as effective users, and associating the original all stay point data with the effective users to obtain effective user stay data. 3.The method of claim 1, wherein, The time-space data integration platform in step S1 comprises basic spatial data and time-space trajectory data, the basic spatial data is obtained from a public website and comprises land use data, POI data, road network data and administrative division spatial data; and the user time-space trajectory data comprises effective user stay data, user residence and user work data and commuting monitoring big data obtained from the public website.

4. The method of claim 1, wherein the method is characterized by: The index system for constructing the time poverty population portrait in step S2 includes: S21: calculating the basic indexes of the index system for constructing the time poverty population portrait to form the time poverty population portrait index data set; the basic indexes include a commuting cost dimension, a working time length dimension, and a facility acquisition dimension; The commuting cost dimension includes commuting time length and commuting distance, the commuting time length is obtained from the effective user stay data, taking the average time length of the effective user commuting between the residence and the workplace, denoted as The commuting distance is obtained from the effective user residence data and land use data, taking the straight-line distance between the effective user's residence and the workplace, denoted as ; The working time length dimension includes single-day working time length and single-week working day number, the single-day working time length is obtained from the effective user stay data, taking the average length of stay of the effective user at the working place, denoted as The single-week working day number is obtained from the effective user stay data, taking the average number of days of the effective user appearing in the stay record of the working place per week and the stay time length being greater than the preset time length, denoted as ; The facility acquisition dimension includes the number of life facilities, the accessibility of life facilities, and the diversity of life facilities, the number of life facilities is acquired by format POI data, land data, and user residential and working land data; the life facility classification includes three categories of health management, commercial service, and sports fitness, an extended boundary is expanded by a preset distance outside the block boundary of the effective user residential land and working land, the total number of various life facilities in the extended boundary is counted, and is denoted as ; The living facility accessibility is obtained from format POI data, land use data, and user residential land data, and the specific formula is as follows: ， ， wherein, is the number of the class of living facilities in the residential or working area, is the number of the class of living facilities in the residential or working area, is the number of the class of living facilities in the residential or working area, is the total number of the class of living facilities that can be covered to the plot, is the average value of the straight-line distance from the geometric center of the residential or working area plot to the class of living facilities, is the living facilities accessibility index of the residential or working area of the user; The living facility diversity is obtained from format POI data, land use data, and user residential land data, and the specific formula is as follows: ， ， wherein, is the proportion of the total number of life facilities in the user's residence or workplace, is the total number of life facilities covered to the user's residence or workplace, is the life facility diversity index of the user's residence and workplace; S22: The aforementioned indicator is commuting time. Commuting distance Daily working hours Number of working days per week Number of living facilities Accessibility index of living facilities and living facilities diversity index The dataset of poverty population profile indicators based on the time of composition is normalized, and the corresponding data after normalization is denoted as . , , , , , and Time poverty index The calculation formula is: ， wherein, are weight coefficients corresponding to each index, which are obtained by the analytic hierarchy process. 5.The method for identifying and diagnosing a commuting corridor for a time-poor population according to claim 1, wherein, Step S3 includes selecting the Dijkstra algorithm for calculating the shortest path from the residential land to the employment land, setting the time poverty population residential land as the start and end points, setting the time poverty population working land as the terminal point, constructing a road network model in the shortest path algorithm from the road network data in the space-time data integration platform, calculating the shortest path according to the Dijkstra algorithm, superimposing all the shortest paths into the same data file, and recording the data file content as the time poverty population commuting corridor. 6.The method of identifying and diagnosing a commuting corridor for a time-poor population according to claim 1, wherein, Step S4 includes: S41: obtaining the user commuting behavior start and end time based on the effective user stay data in the space-time data integration platform, and obtaining the morning and evening peak time of the time poverty population commuting, and verifying based on the commuting monitoring big data, storing the morning and evening peak time of the time poverty population commuting to the target city space-time data integration platform; S42: presetting an evaluation standard, constructing a constraint index system and storing it to the space-time data integration platform for evaluating the target city; the constraint index includes a motor vehicle average speed index and a regular bus average speed index, the motor vehicle average speed index is that the average speed of the expressway and the main road during the morning and evening peak periods is not less than 30 km / h and 20 km / h respectively, and the regular bus running speed index is that the average speed of the regular bus on the time poverty population commuting corridor during the morning and evening peak periods on weekdays is not less than 15 km / h. S43: Taking the static indicators and dynamic indicators as the suggestive indicators and storing them to the spatiotemporal data integration platform under the time dimension of the morning and evening peak hours of the commute; the static indicators refer to the indicators related to the time point, reflecting the target city commute information at a time point and having stability in time, including the rail station surrounding coverage commute proportion and the one-way average commute distance, the rail station surrounding coverage commute proportion refers to the proportion of the number of the residential areas and employment areas within the preset radius of the rail station to the total number of the residential areas and employment areas in the whole city, the one-way average commute distance refers to the average distance of the shortest path of the one-way commute of the time poverty population; the dynamic indicators refer to the indicators related to the time period, reflecting the target city commute information in a time period and having timeliness, including the bus congestion degree and the life facility business proportion, the specific time period selected is the morning and evening peak time of the target city obtained in step S41; wherein, the bus congestion degree refers to the proportion of the bus passenger flow on the time poverty commute corridor to the rated passenger capacity during the morning and evening peak hours of the weekdays; the life facility business proportion refers to the proportion of the life facilities in the open state to the total number of the life facilities in the time poverty commute corridor characteristics set during the morning and evening peak hours of the weekdays.

7. The method of claim 6, wherein the method is characterized by, The step S41 comprises: S411: Based on the effective user dwell time data and user residence and employment location data in the spatiotemporal data integration platform, extract the time-segmented commuting dataset {P1, P2, P3, ..., P} for each user in the time-impoverished population for each hour of the day. 24 }, where P={user, residential plot number, workplace plot number, commuting time under time slice}, 24 is the time slice, and the slice length is one hour; S412: Based on the time period commute data, the total commute time of all the time poverty populations under each time slice is summarized, the time corresponding to the slice with the longest total commute time from zero to twelve o'clock is marked as the morning peak, and the time corresponding to the slice with the longest total commute time from twelve o'clock to twenty-four o'clock is marked as the evening peak; S413: Obtain the average running speed of vehicles in the time-poor population commuting corridor in the time period from the space-time data integration platform, and extract the average running speed of vehicles in the morning peak time marked in step S412 , the average running speed of vehicles in the evening peak period ; as , , if it belongs to the last three of the average running speed of vehicles in each period, it is considered to pass the test. 8.The method of claim 1, wherein, The step S5 comprises: S51: Taking the administrative division vector boundary as the statistical unit, the distribution area and the number of the time poverty population residential areas and work areas in each statistical unit are outputted, and the corresponding time poverty population number is marked; the mileage of the time poverty commute corridor in each statistical unit is counted, and the time poverty commute corridor in the region range is marked; S52: The target city constraint indicators are calculated in the spatiotemporal data integration platform, if the constraint indicator examination fails, the constraint indicator item that fails the examination and the gap between the indicator item and the examination standard are marked, the examination is ended and the result is outputted; if the constraint indicator examination passes, step S53 is continued; S53: The target city suggestive indicator evaluation is calculated in the spatiotemporal data integration platform, and the target city constraint indicator examination result and the suggestive indicator evaluation result are outputted, wherein the suggestive indicator evaluation result specifically includes the indicator name and the indicator suggestion value or the indicator requirement.

9. A commuting corridor identification and problem diagnosis system for time-poor people, characterized in that, Comprise: The preprocessing module is configured to acquire commuting monitoring big data of a target city, location-based service (LBS) data, land use data, POI data, road network data and administrative division spatial data, pre-process the LBS data, screen effective users, calculate the residence and employment locations of the effective users, assign a plot number to each residence and employment location, and import the pre-processed data and the latitude and longitude of the residence and employment locations into a time-space data integration platform of the target city; the commuting monitoring big data includes average vehicle running speed of each road section in each time period and bus passenger flow and rated load flow of each road section in each time period; The method for calculating the residence and employment locations of the effective users is as follows: The space-time anchoring property of the user staying point represents that the staying days of the user staying point in a period of time, the staying duration is stable and persistent, and the staying time period is stable. The space-time anchoring property index is set , The greater the space-time anchoring property of the user staying point is, the stronger the space-time anchoring property of the user staying point is; the user The space-time anchoring property index of the first land unit in the land use data The calculation formula is as follows: ， wherein, represents the user the total number of days of stay on the i-th land unit, represents the user the total number of days of stay on all land units, represents the user the total length of stay on the i-th land unit, represents the user the total length of stay on all land units, represents the user the total length of stay on all land units, stay time entropy represents the user in the plot the dispersion degree of the daily stay time period, and the specific formula is: ， When ; ; ; ; when ; ; ; ; wherein, representing points and Euclidean distance, ( , ) represent the data discrete points of the user in the first land unit on the first day, is the starting time of the longest period of stay of the user in the first land unit per day, is the ending time of the longest period of stay of the user in the first land unit per day, and the integer in the set after rounding, is the total number of days of stay of the user in the first land unit. Identifying a residence, the method comprising: calculating a user a spatio-temporal anchoring dataset for all land units visited by the user , taking the maximum value and marking the corresponding land unit as the user's residence; The method for identifying the work place is: calculating the user The all land unit spatio-temporal anchoring data set stayed on weekdays , take The maximum value, test whether it is different from the residential land mark, if different from the residential land, mark the corresponding land as the user work place; if the same, mark the second largest value in The corresponding land of the user work place; The screening module is configured to construct an index system for the portrait of the time poverty population, calculate the index weight based on the analytic hierarchy process and construct a time poverty index formula, calculate the time poverty index of the effective users in combination with the time-space data integration platform, and screen the time poverty population; The recognition module is configured to perform spatial correlation between the time poverty population and the land use data through the time-space data integration platform, calculate and generate a shortest commuting path based on the shortest path algorithm, generate a buffer zone of the shortest commuting path, and obtain a time poverty commuting corridor feature set of the target city; The index establishing module is configured to identify a commuting peak of the time poverty population of the target city, store the identified commuting peak time into the time-space data integration platform, construct a constraint index system and a suggestion index system as an evaluation and assessment standard for diagnosis of the time poverty commuting corridor, and the suggestion index system includes static indexes and dynamic indexes; The examination and verification module is configured to perform constraint index examination and verification in combination with the time poverty commuting corridor, output a result if the examination and verification fails, and continue to perform suggestion index evaluation if the examination and verification passes, and output constraint index examination results and suggestion index evaluation results on the time-space data integration platform.

Images