A method and system for evaluating urban service facilities based on commuting path prediction

Abstract

The application discloses a kind of urban service facility evaluation method and system based on commuting path prediction, it is related to the technical field of city planning, and the crowd flow feature database is constructed by city space vectorization data, LBS data and traffic information data, the commuting transfer point in crowd commuting path is obtained by identifying crowd commuting path, and then the layout matching degree of the urban service facility around commuting transfer point is evaluated, the matching degree of crowd flow feature and urban service facility is connected, and the urban service facility is evaluated from the dynamic angle of crowd actual use, to provide technical support for the function layout, planning design of service facility around urban traffic facility transfer hub.

Description

Technical Field

[0001] This invention relates to the field of urban planning technology, specifically to a method and system for evaluating urban service facilities based on commuter route prediction. Background Technology

[0002] With rapid urban development, residents' demand for public service facilities is constantly increasing. However, insufficient supply and uneven regional distribution of facilities have become the norm, making the rational allocation of urban public service facilities particularly important. On the other hand, the development of urban public transportation has greatly changed people's travel patterns. The population agglomeration brought about by the huge commuting flow has spurred new demands for public service facilities in different areas. Urban planning needs to pay attention to the changes in demand for regional public service facilities brought about by commuting, thereby optimizing the allocation of public service facilities along important commuting corridors to improve the efficiency of public service facility use and achieve precise and rational allocation of urban resources.

[0003] Currently, common methods for evaluating the matching degree of urban public service facilities fall into two categories. One is based on analysis of urban public service facility data and resident population data. However, this approach fails to consider changes in urban population distribution due to commuting, making it difficult to achieve a scientific allocation of public service facilities based on actual population needs. Furthermore, the relatively outdated population census data leads to analysis results that are not granular, lack specificity, and are inaccurate. The other method uses the "per thousand people index" to calculate the density of specific facilities per unit area to explore the fairness of public service facility supply. This method uses a top-down facility allocation evaluation standard, failing to consider regional differences in public service facility demand caused by changes in population flow during urban operations. Therefore, the calculation results cannot effectively serve the accurate understanding and regulatory decisions regarding changes in the supply and demand relationship of public service facility resources in daily urban operations. Summary of the Invention

[0004] To address the shortcomings mentioned in the background section, the present invention aims to provide a method and system for evaluating urban service facilities based on commuting route prediction.

[0005] The objective of this invention can be achieved through the following technical solution: a method for evaluating urban service facilities based on commuter route prediction, the method comprising the following steps:

[0006] The system receives city-related data, performs spatial correlation on the city-related data, and verifies and matches the geographic coordinate system to obtain city data; wherein, the city data includes: city spatial vectorized data, city LBS data, and city traffic information data;

[0007] User trajectory lines are constructed based on urban LBS data, and then spatially connected with urban land use data in urban spatial vectorization data to obtain a personal path dataset.

[0008] By using urban land use data and residence time, the employment and residence locations of urban residents can be identified. By connecting the average speed of the commuting trajectory between the employment and residence locations of urban residents with the travel mode speed range table, the commuting modes of users can be divided.

[0009] By associating users' commuting modes with city LBS data, spatial vectorized data of people's commuting paths is obtained, and this spatial vectorized data of people's commuting paths is input into the personal path dataset.

[0010] Commuting transfer points are obtained by spatially vectorizing the commuting paths of people in the personal path dataset based on users' commuting travel modes, and then spatially correlated with urban land use data to obtain a commuting transfer point dataset.

[0011] By combining the commuter transfer point dataset with urban business data within the urban spatial vectorization data, public service facilities are rated from two aspects: transportation facilities and commercial facilities, resulting in multiple transfer point service facility capability levels.

[0012] The results of the capacity assessment of service facilities at multiple transfer points were verified, and the final evaluation results of the matching degree of service facilities at each commuter transfer point were obtained.

[0013] Preferably, the urban spatial vectorized data includes existing urban building data, urban road data, urban land use data, and urban business data; the urban LBS data includes anonymized population flow data; and the urban traffic information data includes urban traffic facility layout data, urban road monitoring data, and urban subway card swiping data.

[0014] Preferably, the personal path dataset includes the average speed, start time, and stop time of each user trajectory segment, as well as the land use attributes of the start and end points of each user trajectory segment.

[0015] Preferably, the process of identifying the place of employment and place of residence of the urban population includes the following steps:

[0016] Through N ps Δ t M inpts Three parameter values ​​are used to identify motion-dwelling states, thereby identifying the user's set of dwell points. Dwell points are identified by adding time constraints and giving the cluster boundary flexibility, as shown in the following expression:

[0017] P = {P1, P2, P3, ..., P} n}

[0018] P n =min{p1,…,p m ,…,p t,…,p u}

[0019] The spatiotemporal selection criteria for trajectory dwell points are as follows:

[0020] 0≤d(p m ,p t ) <N ps

[0021] 0≤t(p m ,p t )<Δ t

[0022] u≥M inpts

[0023] Where: P is the cluster set of trajectory points; P n Let p1, p be the cluster set of the nth trajectory point; m ,p t ,p u P respectively n The first, m, t, and u-th trajectory points in Pn; u is the number of trajectory points in Pn; d(p m ,p t () represents the trajectory point p m ,p t Spatial distance; t(p) m ,p t () represents the trajectory point p m ,p t Time difference; N ps Cluster radius constraint; Δt is the clustering time constraint; M inpts Minimum trajectory point constraint;

[0024] The set of residence points is split by day. For the daily residence point data, the user's daytime residence point set and nighttime residence point set are divided according to time to determine the place of employment and residence. The daytime period from 10:00 to 17:00 is set as working time. The non-residential land locations where the employees mainly reside are the residence point set of the place of employment. The two time periods from 0:00 to 6:00 and from 22:00 to 24:00 are set as rest time. The residence points within the residential land during the period are the residence point set.

[0025] For both daytime and nighttime stay locations, the cumulative stay duration and number of days at each location are calculated. The location with the longest cumulative daytime stay exceeding 60% of the total observation days is considered the workplace; the location with the longest cumulative nighttime stay exceeding 80% of the total observation days is considered the residence. For each user, the straight-line distance between the workplace and the nighttime residence is calculated. If the distance is less than 1km, the user is marked as a home-based worker. Finally, the stay location records for home-based workers are filtered and deleted using the following formula:

[0026] t adi =Max(t) 1da ,t 2da ,t 3da ,t 4da ,…,t nda )

[0027]

[0028] Where D represents the total number of observation days, t adi This represents the period during which user a spends the longest time in non-residential area i during the working hours of day d, and their total stay time is t. adi W ai This refers to the location identification index for user A. A value of 1 indicates successful identification, and a value of 0 indicates failure. There can be at most one W. ai =1, Place of employment successfully identified;

[0029] q adi =Max(t) 1da ,t 2da ,t 3da ,t 4da ,…,t nda )

[0030]

[0031] Where D represents the total number of observation days, q adi This represents the period during which user a spends the longest total time at residential site i during the rest period on day d, and their total stay time is q. adi R ai This refers to the location identification index for user A. A value of 1 indicates successful identification, and a value of 0 indicates failure. There can be at most one R. ai =1, residence identification successful.

[0032] Preferably, the user's commuting mode includes slow-moving transportation, public transportation, and private rapid transportation. Slow-moving transportation includes walking, bicycles, and electric vehicles. Public transportation includes subways and buses. Private rapid transportation includes private cars and taxis.

[0033] Preferably, the latitude and longitude coordinates, timestamps, and user ID information of all subdivided road segments in the spatial vectorized data of the commuting path are input into the personal commuting path dataset.

[0034] Preferably, the identification process of the commuter transfer point is as follows:

[0035] Since walking plays an intermediate role in various commuting transfer modes, the commuting route is regarded as multiple "walking-X-walking" trajectories. The user's commuting mode is identified, and the transfer point is identified based on the change value of the user's commuting mode and walking speed.

[0036] The mathematical description is as follows:

[0037]

[0038]

[0039] i = 1, ..., m

[0040]

[0041] Wherein, t{(W n ), ε} are multi-class optimization functions, W n ε is the method vector of the training set of individual commuting routes. i is the boundary error relaxation factor, c is the penalty coefficient, m represents the six transfer modes to be identified, and yi is the transportation mode label. x is the intercept of the boundary function for sample n. n The final decision will be made regarding the mode of transfer;

[0042] The commuter transfer point dataset includes user ID, transfer method, transfer time, transfer point latitude and longitude, and transfer point land use attributes.

[0043] Preferably, the service facility capacity levels of the multiple transfer points are rated by calculating service facility matching evaluation factors. These service facility matching evaluation factors include transportation facility matching evaluation factors and commercial facility matching evaluation factors. The transportation facility matching evaluation factors include:

[0044] Number of sites N p Average transfer distance and average transfer time The number of stations N is obtained by counting the number of transit facility points (POIs) contained in the buffer zones of each commuter transfer point. p The number of stations is sequentially numbered 1, 2, 3, ..., p, and the transfer distance L between each commuter transfer point and each transportation facility point within the corresponding buffer zone is calculated. xAnd calculate the corresponding transfer time T. x The average transfer distance from each commuter transfer point to each transportation facility point within the buffer zone is obtained. and average transfer time The calculation formula is as follows:

[0045]

[0046]

[0047] In the formula, T x Let p be the transfer time from the commuter transfer point to the x-th transportation facility, p be the total number of transportation stations contained in the buffer zone of the commuter transfer point, and x be the transportation facilities studied in the buffer zone of the commuter transfer point, where x takes the values ​​1, 2, 3, ..., p. L represents the commuting time from the commuter transfer point to the xth transportation facility. x Let v be the transfer distance from this commuter transfer point to the xth transportation facility, and v be the travel speed of slow-moving transportation. Let x be the required waiting time at the xth transportation facility. Let x be a constant, when x represents an airport or a train station. Take 60 minutes, when x represents a long-distance bus station or a port terminal. Take 30 minutes, and when x represents a subway station or bus stop, Take 8 minutes; when x is a parking lot or a shared bicycle parking spot, Take 0 minutes;

[0048] The evaluation factors for the matching degree of commercial facilities include:

[0049] Business Format Diversity U k and average service facility size The calculation formula is:

[0050]

[0051]

[0052] In the formula, Service_q k The information entropy of the business types within the buffer zone k of this commuter transfer point is Service_q. k The larger the area, the richer the business formats (U) k The larger n is, the greater n is the total number of different types of commercial service facilities contained within the buffer zone k of the commuter transfer point. The percentage of Class P commercial service facilities. Let N be the average size of the service facilities at this commuter transfer point, and E be the total number of commercial service facilities. nLet n be the size of the nth commercial service facility, where n can be 1, 2, ..., N;

[0053] The evaluation factors for matching transportation facilities and commercial facilities are normalized. The normalized results are then weighted to obtain the service facility matching evaluation results for commuter transfer points. The service facility matching evaluation results for commuter transfer points are then normalized to obtain the service facility matching evaluation normalized results for commuter transfer points. The service facility matching evaluation normalized results for commuter transfer points are then classified into levels to obtain multiple service facility capability levels for transfer points.

[0054] Preferably, the verification process for the capacity levels of the service facilities at the multiple transfer points is as follows:

[0055] Calculate the crowd density at each commuter transfer point based on the number of individual identity IDs. p The capacity levels of service facilities at multiple transfer points are verified using the following calculation formula:

[0056]

[0057]

[0058] In the formula, O p Let M be the population density at the p-th commuter transfer point. p Let S be the number of people within the buffer zone of the p-th commuter transfer point, and S be the area of ​​a circular buffer zone with a radius of 1 km; W' pk The normalized result of the service facility matching degree evaluation for the Pth commuter transfer point. The average population density at commuter transfer points is γ, which is a check coefficient and is set to 0.25.

[0059] After verification, the normalized result W' of the service facility matching degree evaluation based on commuter transfer points pk The system is categorized into different levels, and the final output is the evaluation results of the matching degree of service facilities at each commuter transfer point.

[0060] Secondly, in order to achieve the above objectives, the present invention discloses an urban service facility evaluation system based on commuting route prediction, comprising:

[0061] Data preprocessing module: used to receive city-related data, perform spatial correlation on the city-related data and perform geographic coordinate system verification and matching to obtain city data; wherein, the city data includes: city spatial vectorized data, city LBS data and city traffic information data;

[0062] Dataset construction module: used to construct user trajectory lines based on urban LBS data, and spatially connect user trajectory lines with urban land use data in urban spatial vectorization data to obtain personal path dataset;

[0063] Trip segmentation module: It is used to identify the employment and residence of urban populations by using urban land use data and residence time. By connecting the average speed of the commuting trajectory between the employment and residence of urban populations with the travel mode speed range table, the user's commuting travel mode is segmented.

[0064] Data association module: used to associate users' commuting travel modes with city LBS data to obtain spatial vectorized data of people's commuting paths, and input the spatial vectorized data of people's commuting paths into the personal path dataset;

[0065] Commuter Transfer Module: This module is used to obtain commuter transfer points based on users' commuting modes by using spatial vectorized data of commuting paths within the personal path dataset, and spatially correlate commuter transfer points with urban land use data to obtain a commuter transfer point dataset.

[0066] The rating module combines the commuter transfer point dataset with urban business data within the urban spatial vectorization data to rate public service facilities from both transportation and commercial perspectives, resulting in multiple transfer point service facility capability levels.

[0067] The results output module is used to verify the capability levels of service facilities at multiple transfer points and ultimately obtain the matching degree evaluation results of service facilities at each commuter transfer point.

[0068] The beneficial effects of this invention are:

[0069] This invention constructs an urban population mobility feature database, and accurately identifies the residences and workplaces of urban populations based on urban LBS data and urban land use data, using the DBSCAN clustering algorithm and the spatiotemporal selection conditions of trajectory dwell points. By obtaining a training set of individual commuting routes through trajectory information, it greatly improves the efficiency and accuracy of population commuting route identification.

[0070] This invention uses the support vector machine method to identify multi-modal travel modes. Based on the changes in travel mode and travel speed, it identifies transfer points. Compared with traditional commuter transfer point identification, the results are more scientific and accurate.

[0071] This invention improves the efficiency of evaluating the service capacity of public service facilities at commuter transfer points by constructing a service capacity matching evaluation system and calculating the service facility matching degree of each commuter transfer point. Compared with traditional service capacity evaluation, it saves a lot of manpower and resources. Attached Figure Description

[0072] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a flowchart of a method according to an embodiment of the present invention;

[0074] Figure 2 This is a schematic diagram of the system structure of the present invention;

[0075] Figure 3 This is a diagram illustrating the process of identifying commuter routes in an embodiment of the present invention.

[0076] Figure 4 This is a diagram showing the commuter transfer point identification results according to an embodiment of the present invention;

[0077] Figure 5 This is a diagram showing the identification results of service facilities around commuter transfer points according to an embodiment of the present invention.

[0078] Figure 6 This is a graph showing the evaluation results of the matching degree of commuter transfer point service facilities in an embodiment of the present invention. Detailed Implementation

[0079] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0080] like Figure 1 As shown, an evaluation method for urban service facilities based on commuting route prediction is proposed, which includes the following steps:

[0081] The system receives city-related data, performs spatial correlation on the city-related data, and verifies and matches the geographic coordinate system to obtain city data; wherein, the city data includes: city spatial vectorized data, city LBS data, and city traffic information data;

[0082] In this embodiment, urban spatial vectorization data, urban LBS data, and urban traffic information data are imported into the geographic information platform database management system via a mobile data interface with 2TB of memory and USB 3.0. The urban spatial vectorization data includes urban existing building data, urban road data, urban land use data, and urban business data; the urban LBS data includes anonymized population flow data; and the urban traffic information data includes urban traffic facility layout data, urban road monitoring data, and urban subway card swiping data.

[0083] User trajectory lines are constructed based on urban LBS data, and then spatially connected with urban land use data in urban spatial vectorization data to obtain a personal path dataset.

[0084] The raw data for the personal path dataset includes urban LBS data and urban land use data. The urban LBS data includes the user's ID, timestamp, longitude, and latitude of their current location. The urban vector land use data contains land use number, land use code, and specific location information of land use boundaries. Both are based on the geographic coordinate system WGS1984. Using a DGX workstation equipped with ArcGIS Pro 3.0.2, the point-to-track segment tool was used to construct the user's trajectory line based on the urban LBS data points. The start and end points of the user's trajectory line were then spatially connected to the urban vector land use data. This resulted in the personal path dataset, which includes the average speed, start and stop times, and land use attributes of the start and end points for each trajectory segment.

[0085] By using urban land use data and residence time, the employment and residence locations of urban residents can be identified. By connecting the average speed of commuting trajectories between the employment and residence locations of urban residents with the speed range table of travel modes, the commuting modes of users can be divided.

[0086] Identification of place of employment and place of residence

[0087] When a user is in a resident state, the spatial spacing of LBS data is small and the density is high, therefore, through N ps Δ t M inpts Three parameter values ​​are used to identify motion-dwelling states, thereby identifying the user's set of dwell points. Based on the DBSCAN clustering algorithm, dwell points are identified by adding time constraints and giving the cluster boundaries flexibility. Its mathematical expression is as follows:

[0088] P = {P1, P2, P3, ..., P} n}

[0089] P n =min{p1,…,p m ,…,p t ,…,pu}

[0090] The spatiotemporal selection criteria for trajectory dwell points are as follows:

[0091] 0≤d(p m ,p t ) <N ps

[0092] 0≤t(p m ,p t )<Δ t

[0093] u≥M inpts

[0094] Where: P is the cluster set of trajectory points; P n Let p1, p be the cluster set of the nth trajectory point; m ,p t ,p u P respectively n The first, m, t, and u-th trajectory points in Pn; u is the number of trajectory points in Pn; d(p m ,p t () represents the trajectory point p m ,p t Spatial distance; t(p) m ,p t () represents the trajectory point p m ,p t Time difference; N ps Cluster radius constraint; Δt is the clustering time constraint; M inpts This is a minimum trajectory point constraint.

[0095] The location data is split by day. For each day's location data, the user's daytime location data and nighttime location data are divided to distinguish between workplace and residence. The period from 10:00 to 17:00 is defined as working hours; the non-residential locations where employees primarily reside constitute the workplace location data set. The periods from 0:00 to 6:00 and from 22:00 to 24:00 are defined as rest periods; locations within residential areas during these periods constitute the residential location data set.

[0096] For both daytime and nighttime stay locations, the cumulative stay duration and number of days at each location were calculated. The location with the longest cumulative daytime stay exceeding 60% of the total observation days was considered the workplace; the location with the longest cumulative nighttime stay exceeding 80% of the total observation days was considered the residence. For each user, the straight-line distance between their workplace and nighttime residence was calculated. If the distance was less than 1km, they were marked as a home-based worker, and their stay records were filtered and deleted.

[0097] t adi =Max(t) 1da ,t 2da ,t 3da ,t 4da ,…,t nda )

[0098]

[0099] Where D represents the total number of observation days, t adi This represents the period during which user a spends the longest time in non-residential area i during the working hours of day d, and their total stay time is t. adi W ai This refers to the location identification index for user A. A value of 1 indicates successful identification, and a value of 0 indicates failure. There can be at most one W. ai =1, meaning the workplace was successfully identified.

[0100] q adi =Max(t) 1da ,t 2da ,t 3da ,t 4da ,…,t nda )

[0101]

[0102] Where D represents the total number of observation days, q adi This represents the period during which user a spends the longest total time at residential site i during the rest period on day d, and their total stay time is q. adi R ai This refers to the location identification index for user A. A value of 1 indicates successful identification, and a value of 0 indicates failure. There can be at most one R. ai =1, meaning the place of residence has been successfully identified.

[0103] Classification of travel modes

[0104] Trajectory information between residence and workplace is selected from the generated personal route training set. Based on the average speed, start and stop times, and land use attributes of the origin and destination points of each trajectory segment, a travel mode speed comparison table is used for comparison. The mode of transportation for each trajectory is determined using fuzzy set theory, resulting in the personal commuting route training set. Slow-moving transportation includes walking, cycling, and electric vehicles; public transportation includes subways and buses; and private rapid transit includes private cars and taxis, as shown in Table 1.

[0105] Table 1: Comparison of Travel Mode Speeds

[0106] slow traffic <10km / h Public transport 10-25km / h Private express transportation >25km / h

[0107] The OD information and travel mode information of each trajectory in the personal commuting route training set were input into the Gaode Maps route planning platform for route planning simulation. Recommended route 1 was selected as the spatial vectorization data of the commuting route. The latitude and longitude, timestamps, and user ID information of all subdivided road segments in the spatial vectorization data of the commuting route were input into a DGX computer workstation equipped with ArcGIS Pro 3.0.2. Using a projection tool, the latitude and longitude information of the Gaode coordinate system was projected onto the geographic coordinate system WGS1984. The transformed route information was used as an extension of the personal commuting route dataset. The process of identifying commuting routes in a certain area of ​​a certain city is as follows. Figure 3 As shown.

[0108] By associating users' commuting modes with city LBS data, spatial vectorized data of people's commuting paths is obtained, and this spatial vectorized data of people's commuting paths is input into the personal path dataset.

[0109] Commuting transfer points are obtained by spatially vectorizing the commuting paths of people in the personal path dataset based on users' commuting travel modes, and then spatially correlated with urban land use data to obtain a commuting transfer point dataset.

[0110] It should be noted that the specific process in this embodiment is as follows:

[0111] Commuter transfer point identification and selection

[0112] Since walking serves as an intermediate link between various commuting modes, the commuting route is viewed as multiple "walk-X-walk" trajectories. Support Vector Machines are used to identify multi-modal travel, and transfer points are identified based on changes in travel mode and walking speed.

[0113] The mathematical description is as follows:

[0114]

[0115]

[0116] i = 1, ..., m

[0117]

[0118] Wherein, t{(W n ), ε} are multi-class optimization functions, W n ε is the method vector of the training set of individual commuting routes. i is the boundary error relaxation factor, c is the penalty coefficient, m represents the six transfer modes to be identified, and yi is the transportation mode label. x is the intercept of the boundary function for sample n.n The final decision will be made regarding the transfer method.

[0119] Table 2: Transfer Options

[0120] 1 walk slow traffic 2 walk Public transport 3 walk Private express transportation 4 slow traffic walk 5 Public transport walk 6 Private express transportation walk

[0121] Construction of commuter transfer point dataset

[0122] Based on the identification and selection of transfer methods and transfer points, the data is input into a DGX workstation running ArcGIS Pro 3.0.2. Spatial association is performed with the corresponding transportation facility land use data to construct a commuter transfer point dataset containing user ID, transfer method, transfer time, transfer point latitude and longitude, and transfer point land use attributes. The commuter transfer point identification results for a certain area in a certain city are shown below. Figure 4 As shown.

[0123] By combining the commuter transfer point dataset with urban business data within the urban spatial vectorization data, and rating public service facilities from both transportation and commercial aspects, multiple transfer point service facility capability levels are obtained.

[0124] Identification of transfer point service facilities

[0125] Using the acquired commuter transfer point dataset, a buffer zone with a radius of 1km was constructed for each commuter transfer point in a DGX workstation running ArcGIS version 10.8. Based on the acquired urban business data, the POI data of transportation facilities and POI data of commercial service facilities falling into each commuter transfer point buffer zone were statistically analyzed. The transportation facility POI data included subway stations, bus stops, parking lots, long-distance bus stations, airports, ports, train stations, and shared bicycle parking areas. The commercial service facility POI data included restaurants, shopping malls, supermarkets, home furnishing and building materials markets, flower, bird, fish and insect markets, general markets, convenience stores, hotels, home appliance and electronics stores, barbershops, and cinemas. The identification results of service facilities around commuter transfer points in a certain area of ​​a city are as follows: Figure 5 As shown.

[0126] Calculation of service facility matching evaluation factors

[0127] Based on the acquired POI data of transportation facilities and the commuter flow path identification model, the transportation facility matching degree evaluation factor value of each commuter transfer point is calculated, including the number of stations N. p Average transfer distance and average transfer time The number of stations N is obtained by counting the number of transit facility points (POIs) contained in the buffer zones of each commuter transfer point. pThe number of stations is sequentially numbered 1, 2, 3, ..., p, and the transfer distance L between each commuter transfer point and each transportation facility point within the corresponding buffer zone is calculated. x And calculate the corresponding transfer time T. x The average transfer distance from each commuter transfer point to each transportation facility point within the buffer zone is obtained. and average transfer time The calculation formula is as follows:

[0128]

[0129]

[0130] In the formula, T x Let p be the transfer time from the commuter transfer point to the x-th transportation facility, p be the total number of transportation stations contained in the buffer zone of the commuter transfer point, and x be the transportation facilities studied in the buffer zone of the commuter transfer point, where x takes the values ​​1, 2, 3, ..., p. L represents the commuting time from the commuter transfer point to the xth transportation facility. x Let v be the transfer distance from this commuter transfer point to the xth transportation facility, and v be the travel speed of slow-moving transportation. Let x be the required waiting time at the xth transportation facility. Let x be a constant, when x represents an airport or a train station. Take 60 minutes, when x represents a long-distance bus station or a port terminal. Take 30 minutes, and when x represents a subway station or bus stop, Take 8 minutes; when x is a parking lot or a shared bicycle parking spot, Take 0 minutes.

[0131] Wherein, the transfer distance L between each commuter transfer point and each transportation facility point within the corresponding buffer zone x The method involves taking the commuter transfer point as the "origin" and the xth transportation facility point as the "destination". Based on a crowd flow commuter path identification model, the commuter path between the origin and destination is obtained. The length and number of commuter paths are counted, and the average length of each path is taken to obtain the transfer distance L at the xth transportation facility point. x .

[0132] Based on the acquired POI data for commercial service facilities, the matching factor value of commercial facilities at each commuter transfer point is calculated, including the business format richness U. k and average service facility size The calculation formula is:

[0133]

[0134]

[0135] In the formula, Service_q k The information entropy of the business types within the buffer zone k of this commuter transfer point is Service_q. k The larger the area, the richer the business formats (U) k The larger n is, the greater n is the total number of different types of commercial service facilities contained within the buffer zone k of the commuter transfer point. The percentage of Class P commercial service facilities. Let N be the average size of the service facilities at this commuter transfer point, and E be the total number of commercial service facilities. n Let n be the size of the nth commercial service facility, where n takes the values ​​1, 2, ..., N. Table 3 shows the evaluation factors for the matching degree of service facilities at a commuter transfer point in a certain street of a certain city.

[0136] Table 3 - Evaluation Factors for Matching Degree of Commuter Transfer Point Services in a Certain Street of a Certain City

[0137]

[0138]

[0139] Service facility matching evaluation and rating

[0140] The obtained evaluation factor values ​​for the matching degree of transportation facilities at commuter transfer points (number of stations N) p Average transfer distance and average transfer time ) and the evaluation factor value of business facility matching degree (business format richness U) k and average service facility size Normalization is performed, and the normalization results are weighted to obtain the service facility matching degree evaluation result W of commuter transfer points. pk The calculation formula is:

[0141]

[0142] W pk =δ·D' p1 +∈D' p2 +∈D' p3 +δ·D' p4 +δ·D' p5

[0143] In the formula, i represents the category of the facility matching evaluation factor, and i takes the values ​​1, 2, 3, 4, and 5, which respectively represent the number of sites N. p Average transfer distance Average transfer time Business Format Diversity U kand average service facility size D' pi For the normalized result of the evaluation factor of the matching degree of the i-th facility at the p-th commuter transfer point, D pi Let D be the value of the facility matching evaluation factor for the p-th commuter transfer point i-th facility. imin Let D be the minimum value of the facility matching evaluation factor for the p-th commuter transfer point i-th facility. imax W represents the maximum value of the facility matching evaluation factor for the i-th commuter transfer point at the p-th commuter transfer point; pk Let δ and ∈ be the service facility matching degree evaluation results of the Pth commuter transfer point. Since the number of stations, the richness of business formats and the average scale of service facilities have a positive effect on the facility matching degree, δ is taken as 1, while the average transfer distance and the average transfer time have a negative effect on the facility matching degree, so ∈ is taken as -1.

[0144] Evaluation results of the matching degree of service facilities at commuter transfer points W pk After normalization, the service facility matching degree is divided into evaluation levels, and the calculation formula is as follows:

[0145]

[0146] In the formula, W' pk W is the normalized result of the service facility matching degree evaluation for the Pth commuter transfer point. pk W represents the service facility matching evaluation result for the Pth commuter transfer point. kmin W represents the minimum service facility matching degree for commuter transfer points. kmin This represents the maximum service facility matching degree for commuter transfer points.

[0147] The method for classifying the matching degree of commuter transfer point service facilities is as follows: when W' pk When the score is ≤0.25, the commuter transfer point is designated as a first-level service facility matching point; when it is 0.25... <W' pk When the score is ≤0.5, the commuter transfer point is designated as a secondary matching point for service facilities; when it is 0.5... <W' pk When W' is ≤0.75, the commuter transfer point is designated as a commuter transfer point with a service facility matching degree of level three. pk When the score is ≥0.75, the commuter transfer point is classified as a Level 4 service facility matching point. Table 4 shows the service facility matching level results for a commuter transfer point in a certain street of a certain city.

[0148] Table 4 - Matching Degree Results of Commuter Transfer Point Service Facilities in a Certain Street of a Certain City

[0149]

[0150] The results of the capacity assessment of service facilities at multiple transfer points were verified, and the final evaluation results of the matching degree of service facilities at each commuter transfer point were obtained.

[0151] Calculate the crowd density O at each commuter transfer point based on the number of obtained human identity IDs. p The service facility matching degree results are verified using the following calculation formula:

[0152]

[0153]

[0154] In the formula, O p Let M be the population density at the p-th commuter transfer point. p Let S be the number of people within the buffer zone of the p-th commuter transfer point, and S be the area of ​​a circular buffer zone with a radius of 1 km; W' pk The normalized result of the service facility matching degree evaluation for the Pth commuter transfer point. γ represents the average population density at commuter transfer points, and γ is the verification coefficient, which is set to 0.25.

[0155] After verification, the normalized result W' of the service facility matching degree evaluation based on commuter transfer points pk The matching degree of commuter transfer point service facilities is determined, and the evaluation results of the matching degree of service facilities at each commuter transfer point are finally output. The evaluation results of the matching degree of commuter transfer point service facilities in a certain area of ​​a certain city are as follows: Figure 6 As shown.

[0156] Interactive display of categorized matching degree evaluation

[0157] The obtained commuter transfer point service facility information table and service facility matching degree evaluation results are input into the geographic information platform and visualized by type and level. The visualization results are output through a 2K resolution, 27-inch LED display screen. The commuter transfer point service facility information table includes numerical information on five characteristic factors for each commuter transfer point: number of stations, average transfer distance, average transfer time, business type richness, and average service facility size. The output is a table in .xlsx format. The commuter transfer point service facility matching degree evaluation results include all commuter transfer point service facility matching degree evaluation results and the verified service facility matching degree evaluation results. Commuter points with the same service facility matching degree evaluation result are filled with a single color, and different evaluation degrees correspond to different colors. The commuter transfer point service facility matching degree evaluation results are visualized and output as images in .jpg format.

[0158] Evaluation results and output.

[0159] The obtained commuter transfer point service facility information table and service facility matching degree evaluation results are visualized and linked to the databases of planning and design and planning management departments. The commuter transfer point service facility information table and service facility matching degree evaluation map are printed out to guide the planning department in the planning, design and management of transportation hubs.

[0160] On the other hand, such as Figure 2 As shown in the figure, this invention also discloses an urban service facility evaluation system based on commuting route prediction, comprising:

[0161] Data preprocessing module: used to receive city-related data, perform spatial correlation on the city-related data and perform geographic coordinate system verification and matching to obtain city data; wherein, the city data includes: city spatial vectorized data, city LBS data and city traffic information data;

[0162] Dataset construction module: used to construct user trajectory lines based on urban LBS data, and spatially connect user trajectory lines with urban land use data in urban spatial vectorization data to obtain personal path dataset;

[0163] Trip segmentation module: It is used to identify the employment and residence of urban populations by using urban land use data and residence time. By connecting the average speed of the commuting trajectory between the employment and residence of urban populations with the travel mode speed range table, the user's commuting travel mode is segmented.

[0164] Data association module: used to associate users' commuting travel modes with city LBS data to obtain spatial vectorized data of people's commuting paths, and input the spatial vectorized data of people's commuting paths into the personal path dataset;

[0165] Commuter Transfer Module: This module is used to obtain commuter transfer points based on users' commuting modes by using spatial vectorized data of commuting paths within the personal path dataset, and spatially correlate commuter transfer points with urban land use data to obtain a commuter transfer point dataset.

[0166] The rating module combines the commuter transfer point dataset with urban business data within the urban spatial vectorization data to rate public service facilities from both transportation and commercial perspectives, resulting in multiple transfer point service facility capability levels.

[0167] The results output module is used to verify the capability levels of service facilities at multiple transfer points and ultimately obtain the matching degree evaluation results of service facilities at each commuter transfer point.

[0168] Based on the same inventive concept, this invention also provides a computer device, comprising: one or more processors, and a memory for storing one or more computer programs; the programs include program instructions, and the processor executes the program instructions stored in the memory. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, used to implement one or more instructions, specifically for loading and executing one or more instructions stored in a computer storage medium to implement the above-described method.

[0169] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, performs the above-described method. This storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0170] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0171] The foregoing has shown and described the basic principles, main features, and advantages of this disclosure. Those skilled in the art should understand that this disclosure is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of this disclosure. Various changes and modifications can be made to this disclosure without departing from its spirit and scope, and all such changes and modifications fall within the scope of this disclosure as claimed.

Claims

1. A method for evaluating urban service facilities based on commuter route prediction, characterized in that, The method includes the following steps: The system receives city-related data, performs spatial correlation on the city-related data, and verifies and matches the geographic coordinate system to obtain city data; wherein, the city data includes: city spatial vectorized data, city LBS data, and city traffic information data; User trajectory lines are constructed based on urban LBS data, and then spatially connected with urban land use data in urban spatial vectorization data to obtain a personal path dataset. By using urban land use data and residence time, the employment and residence locations of urban residents can be identified. By connecting the average speed of the commuting trajectory between the employment and residence locations of urban residents with the travel mode speed range table, the commuting modes of users can be divided. By associating users' commuting modes with city LBS data, spatial vectorized data of people's commuting paths is obtained, and this spatial vectorized data of people's commuting paths is input into the personal path dataset. Commuting transfer points are obtained by spatially vectorizing the commuting paths of people in the personal path dataset based on users' commuting travel modes, and then spatially correlated with urban land use data to obtain a commuting transfer point dataset. By combining the commuter transfer point dataset with urban business data within the urban spatial vectorization data, public service facilities are rated from two aspects: transportation facilities and commercial facilities, resulting in multiple transfer point service facility capability levels. The service facility capacity levels of multiple transfer points are rated by calculating service facility matching evaluation factors. These evaluation factors include transportation facility matching evaluation factors and commercial facility matching evaluation factors. The transportation facility matching evaluation factors include: Number of sites Average transfer distance and average transfer time The number of stations is determined by counting the number of Points of Interest (POIs) for transportation facilities contained in the buffer zones of each commuter transfer point. The number of stations was sequentially numbered 1, 2, 3, ..., p, and the transfer distances between each commuter transfer point and each transportation facility point within the corresponding buffer zone were calculated. And calculate the corresponding transfer time. The average transfer distance from each commuter transfer point to each transportation facility point within the buffer zone is obtained. and average transfer time The calculation formula is as follows: In the formula, the Let p be the transfer time from the commuter transfer point to the x-th transportation facility, p be the total number of transportation stations contained in the buffer zone of the commuter transfer point, and x be the transportation facilities studied in the buffer zone of the commuter transfer point, where x takes values ​​of 1, 2, 3, ..., p. Let x be the commuting time from the commuter transfer point to the xth transportation facility. Let v be the transfer distance from this commuter transfer point to the xth transportation facility, and v be the travel speed of slow-moving transportation. Let x be the required waiting time at the xth transportation facility. Let x be a constant, when x represents an airport or a train station. Take 60 minutes, when x represents a long-distance bus station or a port terminal. Take 30 minutes, and when x represents a subway station or bus stop, Take 8 minutes; when x is a parking lot or a shared bicycle parking spot, Take 0 minutes; The evaluation factors for the matching degree of commercial facilities include: Business format diversity and average service facility size The calculation formula is: In the formula, Let n be the information entropy of the business types within the buffer zone k of this commuter transfer point, and n be the total number of different types of commercial service facilities contained within the buffer zone k of this commuter transfer point. The percentage of Class P commercial service facilities. Let N be the average size of the service facilities at this commuter transfer point, and let N be the total number of commercial service facilities. Let n be the size of the nth commercial service facility, where n can be 1, 2, ..., N; The evaluation factors for matching the transportation facilities and the evaluation factors for matching the commercial facilities are normalized. The normalized results are then weighted to obtain the evaluation results for the matching of service facilities at commuter transfer points. The evaluation results for the matching of service facilities at commuter transfer points are then normalized to obtain the normalized evaluation results for the matching of service facilities at commuter transfer points. The normalized evaluation results for the matching of service facilities at commuter transfer points are then classified into levels to obtain multiple service facility capability levels for transfer points. The results of the capacity assessment of service facilities at multiple transfer points were verified, and the final evaluation results of the matching degree of service facilities at each commuter transfer point were obtained.

2. The method for evaluating urban service facilities based on commuting route prediction according to claim 1, characterized in that, The urban spatial vectorized data includes existing urban building data, urban road data, urban land use data, and urban business data; the urban LBS data includes anonymized population flow data; and the urban traffic information data includes urban traffic facility layout data, urban road monitoring data, and urban subway card swiping data.

3. The method for evaluating urban service facilities based on commuting route prediction according to claim 1, characterized in that, The personal path dataset includes the average speed, start time, and stop time of each user trajectory segment, as well as the land use attributes of the start and end points of each user trajectory segment.

4. The urban service facility evaluation method and system based on commuter route prediction according to claim 1, characterized in that, The process of identifying the place of employment and residence of the urban population includes the following steps: pass , , Three parameter values ​​are used to identify motion-dwelling states, thereby identifying the user's set of dwell points. Dwell points are identified by adding time constraints and giving the cluster boundary flexibility, as shown in the following expression: P= The spatiotemporal selection criteria for trajectory dwell points are as follows: 0 0 u Where: P is the cluster set of trajectory points; This is the cluster set of the nth trajectory point; They are respectively The first, m, t, and u-th trajectory points in Pn; u is the number of trajectory points in Pn; For trajectory points Spatial distance; For trajectory points Time difference; Δt is the clustering radius constraint; Δt is the clustering time constraint. Minimum trajectory point constraint; The set of residence points is split by day. For the daily residence point data, the user's daytime residence point set and nighttime residence point set are divided according to time to determine the place of employment and residence. The daytime period from 10:00 to 17:00 is set as working time. The non-residential land locations where the employees mainly reside are the residence point set of the place of employment. The two time periods from 0:00 to 6:00 and from 22:00 to 24:00 are set as rest time. The residence points within the residential land during the period are the residence point set. For both daytime and nighttime stay locations, the cumulative stay duration and number of days at each location are calculated. The location with the longest cumulative daytime stay exceeding 60% of the total observation days is considered the workplace; the location with the longest cumulative nighttime stay exceeding 80% of the total observation days is considered the residence. For each user, the straight-line distance between the workplace and the nighttime residence is calculated. If the distance is less than 1km, the user is marked as a home-based worker. Finally, the stay location records for home-based workers are filtered and deleted using the following formula: in, Represents the total number of observation days. This represents the period during which user a spends the longest total time in non-residential land area i during the working hours of day d. Their total stay is... , This refers to the location identification index for user A. A value of 1 indicates successful identification, and a value of 0 indicates failure. There can be at most one such index. =1, Employment location identification successful; in, Represents the total number of observation days. This represents the period during which user a spends the longest total time at residential site i during the rest period on day d. Their total stay is... , This refers to the location identification index for user A. A value of 1 indicates successful identification, and a value of 0 indicates failure. There can be at most one such index. =1, Residence identification successful.

5. The method for evaluating urban service facilities based on commuter route prediction according to claim 1, characterized in that, The user's commuting modes include slow-moving transportation, public transportation, and private express transportation. Slow-moving transportation includes walking, bicycles, and electric vehicles. Public transportation includes subways and buses. Private express transportation includes private cars and taxis.

6. The method for evaluating urban service facilities based on commuter route prediction according to claim 1, characterized in that, The latitude and longitude coordinates, timestamps, and user ID information of all subdivided road segments in the spatial vectorized data of the commuting routes of the population are input into the personal commuting route dataset.

7. The method for evaluating urban service facilities based on commuter route prediction according to claim 1, characterized in that, The process for identifying commuter transfer points is as follows: Since walking plays an intermediate role in various commuting transfer modes, the commuting route is regarded as multiple "walking-X-walking" trajectories. The user's commuting mode is identified, and the transfer point is identified based on the change value of the user's commuting mode and walking speed. The mathematical description is as follows: Min t = s.t. i=1,…,m ,…,k Among them, t For multi-class optimization functions, The method vectors for the training set of individual commuting routes. is the boundary error relaxation factor, c is the penalty coefficient, and m represents the six transfer modes to be identified. For transportation mode labels, The boundary function intercept for sample n, The final decision will be made regarding the mode of transfer; The commuter transfer point dataset includes user ID, transfer method, transfer time, transfer point latitude and longitude, and transfer point land use attributes.

8. The method for evaluating urban service facilities based on commuter route prediction according to claim 1, characterized in that, The verification process for the capacity levels of the service facilities at the multiple transfer points is as follows: Calculate the crowd density at each commuter transfer point based on the number of individual IDs. The capacity levels of service facilities at multiple transfer points are verified using the following calculation formula: In the formula, Let p be the population density at the p-th commuter transfer point. Let S be the number of people in the buffer zone of the p-th commuter transfer point, and let S be the area of ​​the circular buffer zone with a radius of 1 km. The normalized result of the service facility matching degree evaluation for the Pth commuter transfer point. The average population density at commuter transfer points. For verification coefficients, Take 0.25; Normalized results of service facility matching evaluation based on commuter transfer points after verification The system is categorized into different levels, and the final output is the evaluation results of the matching degree of service facilities at each commuter transfer point.

9. A system for evaluating urban service facilities based on commuter route prediction, characterized in that, include: Data preprocessing module: used to receive city-related data, perform spatial correlation on the city-related data and perform geographic coordinate system verification and matching to obtain city data; wherein, the city data includes: city spatial vectorized data, city LBS data and city traffic information data; Dataset construction module: used to construct user trajectory lines based on urban LBS data, and spatially connect user trajectory lines with urban land use data in urban spatial vectorization data to obtain personal path dataset; Trip segmentation module: It is used to identify the employment and residence of urban populations by using urban land use data and residence time. By connecting the average speed of the commuting trajectory between the employment and residence of urban populations with the travel mode speed range table, the user's commuting travel mode is segmented. Data association module: used to associate users' commuting travel modes with city LBS data to obtain spatial vectorized data of people's commuting paths, and input the spatial vectorized data of people's commuting paths into the personal path dataset; Commuter Transfer Module: This module is used to obtain commuter transfer points based on users' commuting modes by using spatial vectorized data of commuting paths within the personal path dataset, and spatially correlate commuter transfer points with urban land use data to obtain a commuter transfer point dataset. The rating module combines the commuter transfer point dataset with urban business data within the urban spatial vectorization data to rate public service facilities from both transportation and commercial perspectives, resulting in multiple transfer point service facility capability levels. The service facility capacity levels of multiple transfer points are rated by calculating service facility matching evaluation factors. These evaluation factors include transportation facility matching evaluation factors and commercial facility matching evaluation factors. The transportation facility matching evaluation factors include: Number of sites Average transfer distance and average transfer time The number of stations is determined by counting the number of Points of Interest (POIs) for transportation facilities contained in the buffer zones of each commuter transfer point. The number of stations was sequentially numbered 1, 2, 3, ..., p, and the transfer distances between each commuter transfer point and each transportation facility point within the corresponding buffer zone were calculated. And calculate the corresponding transfer time. The average transfer distance from each commuter transfer point to each transportation facility point within the buffer zone is obtained. and average transfer time The calculation formula is as follows: In the formula, the Let p be the transfer time from the current commuter transfer point to the x-th transportation facility, p be the total number of transportation stations contained in the buffer zone of the current commuter transfer point, and x be the transportation facilities studied in the buffer zone of the current commuter transfer point, where x takes values ​​of 1, 2, 3, ..., p. Let x be the commuting time from the commuter transfer point to the xth transportation facility. Let v be the transfer distance from this commuter transfer point to the xth transportation facility, and v be the travel speed of slow-moving transportation. Let x be the required waiting time at the xth transportation facility. Let x be a constant, when x represents an airport or a train station. Take 60 minutes, when x represents a long-distance bus station or a port terminal. Take 30 minutes, and when x represents a subway station or bus stop, Take 8 minutes; when x is a parking lot or a shared bicycle parking spot, Take 0 minutes; The evaluation factors for the matching degree of commercial facilities include: Business format diversity and average service facility size The calculation formula is: In the formula, Let n be the information entropy of the business types within the buffer zone k of this commuter transfer point, and n be the total number of different types of commercial service facilities contained within the buffer zone k of this commuter transfer point. The percentage of Class P commercial service facilities. Let N be the average size of service facilities at this commuter transfer point, and N be the total number of commercial service facilities. Let n be the size of the nth commercial service facility, where n can be 1, 2, ..., N; The evaluation factors for matching the transportation facilities and the evaluation factors for matching the commercial facilities are normalized. The normalized results are then weighted to obtain the evaluation results for the matching of service facilities at commuter transfer points. The evaluation results for the matching of service facilities at commuter transfer points are then normalized to obtain the normalized evaluation results for the matching of service facilities at commuter transfer points. The normalized evaluation results for the matching of service facilities at commuter transfer points are then classified into levels to obtain multiple service facility capability levels for transfer points. The results output module is used to verify the capability levels of service facilities at multiple transfer points and ultimately obtain the matching degree evaluation results of service facilities at each commuter transfer point.

Images