CN106779429B - Track transfer station passenger flow congestion risk evaluation method based on AFC card swiping data - Google Patents
Track transfer station passenger flow congestion risk evaluation method based on AFC card swiping data Download PDFInfo
- Publication number
- CN106779429B CN106779429B CN201611211608.4A CN201611211608A CN106779429B CN 106779429 B CN106779429 B CN 106779429B CN 201611211608 A CN201611211608 A CN 201611211608A CN 106779429 B CN106779429 B CN 106779429B
- Authority
- CN
- China
- Prior art keywords
- passenger flow
- risk
- saturation
- evaluation
- station
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012546 transfer Methods 0.000 title claims abstract description 123
- 238000011156 evaluation Methods 0.000 title claims abstract description 62
- 208000027744 congestion Diseases 0.000 title claims description 60
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 238000012216 screening Methods 0.000 claims abstract description 7
- 238000013210 evaluation model Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000004308 accommodation Effects 0.000 claims description 6
- 230000010354 integration Effects 0.000 claims description 4
- 238000007689 inspection Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 abstract description 3
- 238000007418 data mining Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 abstract description 2
- 238000000605 extraction Methods 0.000 abstract description 2
- 230000008030 elimination Effects 0.000 abstract 1
- 238000003379 elimination reaction Methods 0.000 abstract 1
- 230000001174 ascending effect Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000012937 correction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005206 flow analysis Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an AFC card swiping data-based method for evaluating passenger flow congestion risks of a track transfer station, which comprises the following steps of: preprocessing AFC original transaction data; and evaluating passenger flow congestion of the track transfer station based on AFC card swiping data. Based on a database analysis and data mining technology, AFC original card swiping data is subjected to preprocessing such as extraction, elimination and screening, transfer passenger flow identification and the like. Acquiring the inbound and outbound traffic and the transfer passenger traffic of transfer stations at different time periods based on AFC card swiping data, and manually measuring physical parameters of each typical position in the transfer stations in the field; constructing a typical position passenger flow saturation degree evaluation index for comprehensively measuring the passenger flow demand and supply relation of each position; calibrating the model parameters by taking the index maximum value in the evaluation time period as an input value of the model; and calculating the weights of the saturation indexes of different typical positions in the model according to an entropy weight method, clustering the risk evaluation, and dividing five levels of risk levels of passenger flow congestion of the track transfer station.
Description
Technical Field
The invention relates to an AFC card swiping data-based method for evaluating passenger flow congestion risks of a track transfer station, and belongs to the field of public traffic data mining application and service evaluation.
Background
With the development of public transportation, rail line networks of various large cities are increasingly perfected, the passenger traffic pressure of rail transfer stations is gradually increased due to the large increase of rail passenger traffic volume, the continuous increase of the commuting distance of residents and the number of rail transfer times, and especially in the early and late peaks of working days and the period of major holidays, how to realize passenger flow monitoring inside the rail transfer stations and how to realize quantification and classification of passenger flow congestion risks in the rail transfer stations on a microscopic level become important problems influencing the improvement of the operation safety level of the rail transfer stations.
Because the rail transit in China develops rapidly, but the corresponding passenger flow monitoring means are relatively lagged behind, the conventional rail passenger flow monitoring method is not microcosmic, and only can evaluate the overall safety risk of a rail station, or the data acquisition mode is relatively complex, and the final risk evaluation can be realized by integrating the data of various field passenger flow acquisition equipment. The Chinese patent with application number 201410469673.1 discloses a dynamic safety risk evaluation method for a rail transit station. The method comprises the steps of firstly determining a set of dynamic evaluation index system, then calculating various dynamic index values according to data acquired by station equipment in real time, and finally dynamically evaluating the operation safety risk of the rail transit station based on a new method combining interval type two fuzzy number and TOPSIS. The method does not clarify the original data source and the processing process, and the key point is the calculation and evaluation method of the dynamic index value. Since the invention focuses on the comprehensive evaluation of the safety risk of the rail station, it is difficult to evaluate and monitor the risk of the typical position inside the rail station from a more microscopic perspective. Meanwhile, the method has weak pertinence to a large track transfer station with multi-line transfer and is insufficient in consideration of transfer passenger flow inside the track. The Chinese invention patent with the application number of 201410655770.X discloses an AFC passenger ticket data-based urban rail transit passenger flow analysis method. The invention provides a method for analyzing the space-time distribution characteristics of rail passenger flow by taking rail AFC system data as the basis and combining an urban rail transit network and an actual train operation diagram. However, the method is applicable to a track network, and cannot realize the calculation analysis of the passenger flow in the track transfer station.
With the increasing perfection of a large-city track network, the continuous expansion of user scale and the increase of transaction amount, the track AFC system data is used as a traffic data source with strong timeliness, uniform format and high accuracy, can record the passenger flow inlet and outlet amount of track stations in all weather, and basically realizes the centralized storage of the transaction data of the full-track network. By using the AFC data of the rail, the in-and-out and transfer passenger flow in the rail transfer station in the statistical period can be effectively identified and counted, the passenger flow congestion risk of the typical position in the rail transfer station can be accurately calculated by combining the geometric parameters of the main rail station, and further risk grading evaluation and early warning are realized. And the rail AFC data has the characteristics of sufficient data volume, continuously-increased scale and the like, so that the passenger flow congestion risk at the typical position inside a rail station with a plurality of transfer lines can be effectively identified through modeling and mining the AFC system data.
Disclosure of Invention
The invention aims to provide a method for calculating passenger flow congestion risks of a transfer station based on rail AFC card swiping data, which is used for acquiring dynamic passenger flow saturation states of typical positions in the rail transfer station, further realizing monitoring, evaluation and classification of the passenger flow congestion risks and providing support for improving the operation safety service level of the rail transfer station.
In order to achieve the above object, the present invention adopts the following technical solutions.
A rail transfer station passenger flow congestion risk evaluation method based on AFC card swiping data is characterized by comprising the following steps:
step 1.1, extracting main field content in AFC original data of a track;
the AFC original transaction data table has 42 fields, records a large amount of transaction information and simultaneously contains important passenger flow information. AFC transaction data mainly includes: user card number, time for entering and exiting station, line and station for entering and exiting station, etc. Extracting the main fields of the track AFC transaction data table comprises the following steps: a user CARD number (GRANT _ CARD _ CODE); inbound LINE number (ENTRY _ LINE _ NUM); an inbound STATION code (ENTRY _ STATION _ NUM); inbound TIME (ENTRY _ TIME); outbound LINE number (EXIT _ LINE _ NUM); an outbound STATION code (EXIT _ STATION _ NUM); transaction (outbound) TIME (DEAL _ TIME); transaction STATUS (DEAL _ STATUS).
TABLE 1 AFC transaction data sheet
Step 1.2, screening and removing original AFC card swiping data extracted by passenger flow facing a track transfer station;
the rules for eliminating error data and screening valid data are as follows:
1) removing transaction data of which the station entering time and the station exiting time are not on the same day;
2) eliminating the transaction data of which the outbound time is earlier than the inbound time;
3) eliminating records of the same inbound station and outbound station in the card swiping data records;
4) the field of 'DEAL _ STATUS namely transaction state' is screened out to be '2', and 2 represents the data of the record in the state of transaction completion;
step 1.3, judging track of rail transit trip station and identifying transfer passenger flow
The rail transit single trip only needs to swipe a card once when entering and leaving the station, and the transfer in the rail network is not needed to leave the station under the general condition. In order to monitor the passenger flow inside the rail transit, the transfer station of the traveler needs to be judged according to AFC card swiping data, and the transfer time needs to be estimated.
(1) Path and distance determination between any stations of rail transit
To estimate the transfer time, the distance from the departure station to the transfer station needs to be acquired. And searching the shortest path and distance between any stations OD of the track by using an A-x shortest path algorithm, and taking the path as a trip path of a trip person with track transfer. Establishing a RAIL _ PATH (shortest distance table) between any ODs of the RAIL transit, and establishing a new field TRACE in a transaction data table for storing the PATH between any stations of the RAIL transit and recording each station of the PATH between the ODs. For the case of O, D falling on different track routes, the path inferred from the shortest path algorithm may not match the actual path, but in this case it exists only when the two paths travel distances are close and is negligible.
TABLE 2 AFC card swiping data preprocessing result example for rail transit
(2) Rail transit transfer passenger flow identification
Based on the TRACE field, the internal TRANSFER station of the rail transit can be identified, and the TRANSFER TIME (TRANSFER _ TIME) can be calculated according to the number of stations (num (before TRANSFER)) before the TRANSFER station and the number of stations (num (after TRANSFER)) after the TRANSFER station in the travel track, the inbound TIME (ENTRY _ TIME) and the transaction (outbound) TIME (DEAL _ TIME). The transfer time calculation method is as follows:
a new field TRANSFER _ TIME is established in the transaction data table for storing the track traffic internal TRANSFER TIME.
when the passenger flow congestion risk of the track transfer station is evaluated, the complexity of a model and the difficulty degree of data acquisition need to be considered, so that evaluation indexes need to be screened. The main dimensionality of consideration of the passenger flow congestion risk evaluation index is traffic supply and demand, the traffic supply, the traffic demand evaluation index is selected from the incoming station, the outgoing station and the transfer passenger flow, the traffic capacity of each typical position in the track transfer station is selected as the traffic supply evaluation index, the typical position comprises three types of stairs, a channel and a gate, the channel comprises a transfer channel, and the gate comprises a security inspection machine. The inbound, outbound and TRANSFER passenger flow are summed up and counted based on inbound TIME ENTRY _ TIME, transaction/outbound TIME dead _ TIME and TRANSFER TIME TRANSFER _ TIME according to the preprocessed AFC card swiping data, and the minimum counting TIME dimension is set to be 15 minutes. The traffic capacity of each typical position in the transfer station is calculated based on a theoretical formula and measured physical parameters.
the track transfer station passenger flow congestion risk refers to the possibility that some loss occurs within a transfer station due to passenger flow congestion. In the passenger flow congestion risk evaluation, physical meanings represented by the respective indexes are different, and thus there is a difference in dimension. Since such a dimensionability is a main factor affecting the overall evaluation of the object, the evaluation index is uniformly converted into a quantitative value in the range of [0, 1] before the evaluation.
In order to realize the dimensionless evaluation index, the passenger flow saturation r of the typical position of the track transfer station is introduced, and the passenger flow saturation of the ith typical position is the passenger flow saturation r of the typical position ii:
The passenger flow saturation is based on the corrected value P of the total passenger flow passing through the typical position within the statistical timeiMeasured value C of the passing ability of the representative positioniThe ratio of the total passenger flow including the inbound, outbound and transfer passenger flows is calculated, and r is calculatediAnd (3) value taking and constraint:
thus, the indexes are integrated to have a value range of [0, 1]]Dimensionless index r ofiAnd evaluating the passenger flow congestion level of a typical position in the track transfer station within the statistical time based on the index comprehensively considering the passenger flow supply-demand relationship.
The essential of passenger flow congestion in the track transfer station is imbalance of supply and demand, namely the demand is larger than the supply capacity of a traffic system, and the passenger flow congestion risk level of the transfer station is determined as five levels by combining the characteristics of the passenger flow congestion risk:
first-order risk: low risk, capacity for passenger flow far greater than demand;
secondary risk: the risk is low, and the passenger flow accommodation capacity can well meet the requirement;
three-level risk: moderate risk, capacity of passenger flow to meet demand, and congestion in short time;
four-stage risk: the risk is high, the passenger flow accommodation capacity basically meets the requirement, and the congestion condition is obvious;
risk grade five: high risk, near saturation of passenger flow accommodation requirements, difficult capacity satisfaction, and severe congestion;
and determining the weight of each typical position based on the saturation data by using an entropy weight method, and respectively obtaining the saturation of three main regions of stairs, channels and gates and the overall saturation of the transfer station. And aiming at the saturation calculation of the whole large-class regions and transfer stations, the time granularity is small and is used as a minimum unit for dividing peak periods. And (3) taking the maximum saturation values of 4 and 15min in each hour as the saturation of the hour period, establishing a risk evaluation model:
wherein f (y) represents a risk ranking function; alpha represents a normalization coefficient, and 10/n is taken; w is aiEntropy weight coefficients representing typical locations i; r isiRepresenting the passenger flow saturation at a representative location i; n represents the total number of representative locations.
And classifying the passenger flow congestion risk, wherein the method adopts a mean clustering algorithm to classify the risk grade.
and determining objective weight according to the size of the saturation variability of each typical position by using an entropy weight method. If the information entropy of the saturation of a certain typical position is smaller, the variation degree of the saturation of the typical position is larger, the amount of the provided information is larger, the role played in the overall evaluation of the transfer site is larger, and the weight is larger. Conversely, the larger the information entropy of the typical position saturation, the smaller the degree of variation of the typical position saturation, the smaller the amount of information provided, the smaller the effect played by the comprehensive evaluation, and the smaller the weight thereof. The entropy weight method can reflect each index more objectively without any subjective information, and is an objective value assigning method in a complete sense.
Setting n typical positions in selected evaluation space dimension, namely typical position, transfer station and the like, and time dimension, namely 1 hour, 1 day and 1 month evaluation, wherein m sub-targets influencing the comprehensive evaluation value are used, and x is usedijThe data value of the ith target representing the saturation of the representative position j, the n representative positions m target values form a matrix R ═ xij)m×n
Its entropy weight is:
in the formula, gjA difference coefficient representing the saturation of the jth representative position; e.g. of the typejEntropy representing the saturation of the jth representative position; p is a radical ofijThe data value representing the ith sub-target is weighted, and is
Compared with the prior art, the invention has the following obvious advantages and beneficial effects:
(1) by relying on database analysis and data mining technology, AFC original card swiping data is subjected to preprocessing such as extraction, rejection and screening, track internal transfer identification and the like, so that the data quality is improved, and the data acquisition cost is reduced.
(2) The passenger flow congestion risk evaluation model based on the saturation of the typical position in the track transfer station is provided, the demand and the supply capacity of passenger flow in the transfer station are comprehensively considered, the microscopic risk evaluation detailed to the typical position in the transfer station can be realized, and the defect that the passenger flow congestion risk in the station is difficult to evaluate due to the fact that the prior relevant research only takes a track network as an object is overcome.
(3) According to the method, the weight of the saturation index of each typical position in the model is calibrated by using an entropy weight method, the influence degree of different areas on the passenger flow congestion risk of the track transfer station is distinguished, the passenger flow congestion risk of the track transfer station with different time dimensions and space dimensions is quantitatively evaluated, the evaluation method is easier to operate, the evaluation result is more practical, and effective data support can be provided for passenger flow congestion risk early warning and related management and control measures in the track transfer station.
Drawings
FIG. 1 is a flow chart of AFC raw swipe card data pre-processing;
FIG. 2 is a flow chart of a method for evaluating passenger flow congestion risk of a track transfer station based on AFC data;
FIG. 3 is a graph showing the variation of the saturation of the stairs at the entrance A;
FIG. 4 is a graph showing the variation of the saturation of stairs at the AB port;
fig. 5 is a graph of the change of the passenger flow congestion risk level within one week of the east-straightly gated track transfer station;
detailed description of the preferred embodiments
In this embodiment, a straight gate east track transfer station is selected as a calculation object, and passenger flow congestion risk levels of the transfer station in different periods from 3, 5 and 11 days in 2016 are calculated according to AFC card swiping data and physical parameters of various typical positions measured manually.
The embodiment comprises the following steps:
relevant track line (in the case of 2 # line, 13 # line and airport line) basic data and AFC original card swiping data which need to be calculated are imported into an Oracle database. The original AFC data is preprocessed by removing and screening according to the data preprocessing flow shown in FIG. 1.
the passenger flow congestion risk of the track transfer station is influenced by the passenger flow supply and demand relationship, so the evaluation indexes mainly comprise: the traffic capacity of the station is the traffic capacity of the station at each typical position (three categories including stairs, channels (including transfer channels) and gates (including security inspection machines)).
(1) Passenger flow volume
The inbound, outbound, and transfer passenger traffic may be summed and counted based on the pre-processed AFC transaction data, with a minimum time dimension of 15 minutes. The Dongtong gate transfer station is a junction station of the No. 2 line, the No. 13 line and the airport line, and respectively calculates the incoming and outgoing passenger flow of each line and the transfer passenger flow between any two lines.
Table 32016 year 3 month 5 day 2 line outbound passenger flow volume calculation result (part)
(2) Capability of passing through each typical location
And acquiring physical parameters of each typical position in the Dongtong gate transfer station through manual field measurement, and calculating the passing capacity of each typical position by combining a corresponding theoretical formula.
TABLE 4 staircase parameters
Stair | Number and direction of elevators | Stair total width (cm) | Guard bar width (cm) |
AB |
0 | 560 | 56 |
CD |
0 | 560 | 56 |
A-port stair | 1 (uplink) | 340 | 56 |
B- |
1 (uplink) | 320 | 8 |
|
0 | 516 | 0 |
|
1 (uplink) | 320 | 8 |
|
0 | 516 | 0 |
D-shaped stair | 1 (uplink) | 340 | 56 |
E-shaped stair | 1 (Up-run) 1 (Down-run) | 190 | 40 |
G-shaped stair | 1 (Down) | 364 | 58 |
H-shaped stair | 2 (Up-run) 2 (Down-run) | 780 | 44 |
No. 13 |
2 (Down) | 640 | 78 |
No. 13 |
2 (uplink) | 640 | 54 |
|
1 (Up-run) 1 (Down-run) | 0 | 0 |
|
0 | 290 | 48 |
TABLE 5 Gate parameters
TABLE 6 channel parameters
the total passenger flow of a typical position is divided by the passing capacity of the typical position, and the passenger flow saturation of the typical position in statistical time is calculated, so that the non-dimensionalization of the evaluation index is realized. The saturation of each typical position after integration will be used as the final evaluation index. Taking the passenger flow saturation of the A-port stair as an example:
rup stairs-a saturation level indicator at the staircase;
sq repairThe correction coefficient of the flow direction requirement of the q-th passenger flow passing through the stair is that 1 is taken down and 0.98 is taken up;
alpha-demand correction coefficient, taking 1.3 as the passenger flow passing through the A-port stair in consideration of using the subway facility as the underground passage street crossing facility.
θqThe proportion of the qth passenger flow direction in the total number of passenger flows corresponding to the line to which the q passenger flow direction belongs is assumed to be the same without the support of further survey data.
p is the number of inbound, outbound or transfer people per 15 minutes at the transfer site;
q is the number of types of traffic flowing through the stairway;
bi-width (m) of the ith elevator;
m is the number of ascending or descending elevators;
Celevator with a movable elevator carMaximum theoretical throughput capacity of elevator (people/h)M), 8100;
Bj-the total width (m) of the ascending or descending pedestrian stairs j;
bjthe width (m) between the j handrail and the wall of the walking stair is usually 0.24 m;
n is the number of the ascending or descending walking stairs;
Cstep ladderThe maximum theoretical passing capacity (people/h.m) of the walking stairs is 4200 for descending stairs, 3700 for ascending stairs and 3200 for bidirectional mixed-walking stairs.
Substituting 2016 passenger flow data of 3 months and 5 days (saturday) to 3 months and 11 days (friday) to calculate the saturation of the stairs at the A opening (see figure 3), the change of the saturation of the stairs at the A opening at the early peak of the working day (the maximum value is about 0.4) and the significant difference from the weekend (the maximum value is about 0.15) can be obviously seen.
The same algorithm is used for calculating the saturation of the stairs at the AB port platform (see figure 4, the maximum value is about 0.9), so that the obvious difference of the saturation of different positions in the passenger transfer station can be found, and the index can be used as a quantitative value for transversely comparing the congestion risk of passenger flow at each position.
calculating the maximum value max (r) of the saturation of each typical position in a statistical period by taking the saturation of each typical position as a base and taking 15 minutes as a basic time uniti) The statistical period may be hours, days, weeks, months, seasons, years, etc.
in the evaluation model, first, the weight w for each representative positioniAnd performing calibration, and taking the passenger flow saturation of each typical position every 15 minutes as a weight evaluation value. And calculating the weight of each typical position in the model by using an entropy weight method.
TABLE 7 typical location weights for stairs
TABLE 8 channel class exemplary position weights
TABLE 9 typical position weights for gate classes
For the east gate track transfer station, the typical total number of positions n is 28, so α is 10/28, i.e., 5/14. Therefore, the evaluation model of the passenger flow congestion risk of the track transfer station is as follows:
and f, (y) is the final comprehensive passenger flow congestion risk value of the east-straightaway transfer station.
And preliminarily setting and classifying the passenger flow congestion risks of the transfer stations into 4 classes by adopting a k-means clustering method based on the track transfer station risk value data taking hours as granularity. The final cluster centers are as follows:
TABLE 10 Risk value clustering results
Final cluster center
The boundary values for the various classes are: 1.4, 3.3, 5.9. The value of Sig is 0.000, which is less than 0.05, indicating that the difference between the classes is significant. The four classification criteria are:
table 11 transfer station passenger flow congestion risk classification (four levels)
Level of risk of congestion in passenger flow | Value of |
1 | <1.4 |
2 | 1.4-3.3 |
3 | 3.3-5.9 |
4 | ≥5.9 |
Since the data currently used for clustering is taken from 2016, 3, 5 to 11 days, and some special cases (such as holidays) of sudden increase of passenger flow do not exist, on the basis of four types of grading standards, the 5 th-level risk is added to describe the cases of unconventional risks such as holidays, case samples show that the maximum risk value of the track transfer station in a week during the non-holiday period is 7.2, and the 5 th-level boundary risk value is 7.2. Five classification criteria were obtained:
table 12 transfer station passenger flow congestion risk classification (five-level)
Risk rating | Value of | State description | |
1 | <1.4 | The capacity is far greater than the demand, and the surplus is great | |
2 | 1.4-3.3 | Better capacity to meet the demand and |
|
3 | 3.3-5.9 | Capacity is satisfied, and congestion occurs in a |
|
4 | 5.9-7.2 | The capacity basically meets the requirement, and the congestion condition is obvious | |
5 | 7.2-10 | The demand is close to saturation, the capacity is difficult to meet, and the congestion is serious |
According to five types of grading standards, the passenger flow congestion risk level of the east-straightaway transfer station from 5 th to 11 th 3 th month in 2016 is calculated, and the result is shown in fig. 5. From the figure, the regularity of the change in the risk level of congestion of passenger flow at transfer stations can be seen, with the low risk level on weekends (up to level 2), but there are also peak hours. The risk level of the working day peak period is obviously higher than that of weekends, the highest level is 4, and the double peak characteristic is obvious.
Claims (5)
1. A rail transfer station passenger flow congestion risk evaluation method based on AFC card swiping data is characterized by comprising the following steps:
step 1, preprocessing AFC card swiping transaction data;
step 1.1, extracting main field contents in AFC card swiping data;
step 1.2, screening and removing original AFC card swiping data extracted by passenger flow facing a track transfer station;
step 1.3, judging track of a rail transit trip station and identifying transfer passenger flow;
presume the travel station orbit among any orbit OD based on A shortest path algorithm, according to number of stations NUM _ before TRANSFER and number of stations NUM _ after TRANSFER station in the travel orbit, enter TIME ENTRY _ TIME and trade/TIME DEAL _ TIME of leaving, calculate TRANSFER TIME TRANSFER _ TIME;
step 2, selecting an evaluation index;
when the passenger flow congestion risk of the track transfer station is evaluated, the complexity of a model and the difficulty degree of data acquisition need to be considered, so that evaluation indexes need to be screened; the main dimensionality of consideration of the passenger flow congestion risk evaluation index is traffic supply and demand, the station entering, station exiting and transfer passenger flow are selected as traffic demand evaluation indexes, and the traffic capacity of each typical position in a track transfer station is selected as an evaluation index of traffic supply; the typical positions comprise three categories of stairs, channels and gates, wherein the channels comprise transfer channels, and the gates comprise security inspection machines;
step 3, carrying out nondimensionalization and index integration on the evaluation indexes;
the passenger flow congestion risk of the track transfer station refers to the possibility of certain loss caused by passenger flow congestion in the transfer station; in the passenger flow congestion risk evaluation, the physical meanings represented by the indexes are different, so that the indexes have dimensional difference; the different dimension is a main factor influencing the overall evaluation of the object, so that the evaluation indexes are uniformly converted into quantitative values in the range of [0, 1] before the evaluation;
in order to realize the dimensionless evaluation index, the passenger flow saturation r of the typical position of the track transfer station is introduced, and the passenger flow saturation of the ith typical position is the passenger flow saturation r of the typical position ii:
The saturation of passenger flow passes the typical value according to the statistical timeCorrection value P of total passenger flow of positioniMeasured value C of the passing ability of the representative positioniThe ratio of the total passenger flow including the inbound, outbound and transfer passenger flows is calculated, and r is calculatediAnd (3) value taking and constraint:
thus, the indexes are integrated to have a value range of [0, 1]]Dimensionless index r ofiEvaluating the passenger flow congestion level of a typical position in the track transfer station within the statistical time based on the index comprehensively considering the passenger flow supply and demand relationship;
step 4, establishing an evaluation model
The essential of passenger flow congestion in the track transfer station is imbalance of supply and demand, namely the demand is larger than the supply capacity of a traffic system, and the passenger flow congestion risk level of the transfer station is determined as five levels by combining the characteristics of the passenger flow congestion risk:
first-order risk: low risk, passenger flow capacity several times as much as the demand;
secondary risk: the risk is low, and the passenger flow accommodation capacity can meet the requirement;
three-level risk: moderate risk, capacity for passenger flow capacity to meet demand;
four-stage risk: the risk is high, the passenger flow accommodation capacity is consistent with the demand, and the congestion condition is obvious;
risk grade five: high risk, near saturation of passenger flow accommodation requirements, difficult capacity satisfaction, and severe congestion;
determining the weight of each typical position based on the saturation data by using an entropy weight method, and respectively obtaining the saturation of three main regions of stairs, channels and gates and the overall saturation of the transfer station; aiming at the overall saturation calculation of the large-class zone bits and the transfer stations, the time granularity is small and is used as a minimum unit for dividing peak time periods; and (3) taking the maximum saturation values of 4 and 15min in each hour as the saturation of the hour period, establishing a risk evaluation model:
wherein f (y) represents a risk ranking function; alpha represents a normalization coefficient, and 10/n is taken; w is aiEntropy weight coefficients representing typical locations i; r isiRepresenting the passenger flow saturation at a representative location i; n represents the total number of representative locations;
classifying the passenger flow congestion risk, and dividing the risk grade by adopting a mean clustering algorithm;
step 5, determining each typical position weight;
determining objective weight according to the magnitude of the saturation variability of each typical position by using an entropy weight method; if the information entropy of the saturation degree of a certain typical position is smaller, the variation degree of the saturation degree of the typical position is larger, the amount of the provided information is larger, the function which can be played in the overall evaluation of the transfer site is larger, and the weight of the transfer site is larger; conversely, the larger the information entropy of the typical position saturation, the smaller the variation degree of the typical position saturation, the smaller the amount of information provided, the smaller the role played in the comprehensive evaluation, and the smaller the weight thereof; the entropy weight method can reflect each index more objectively without any subjective information, and is an objective assignment method in a complete sense;
setting n typical positions in selected evaluation space dimension, namely typical position, transfer site and time dimension, namely 1 hour, 1 day and 1 month evaluation, setting m targets influencing comprehensive evaluation value, and using xijThe data value of the ith target representing the saturation of the representative position j, the n representative positions m target values form a matrix R ═ xij)m×n
Its entropy weight is:
in the formula, gjA difference coefficient representing the saturation of the jth representative position; e.g. of the typejEntropy representing the saturation of the jth representative position; p is a radical ofijThe data value representing the ith sub-target is weighted,ej∈[0,1];wjis not less than 0, j is not less than 1, 2, 3
2. The AFC swipe data-based track transfer station passenger flow congestion risk evaluation method according to claim 1, wherein the AFC original swipe data field extracted in step 1.1 includes: user card number, inbound route number, inbound station code, inbound time, outbound route number, outbound station code, transaction/outbound time, and transaction state.
3. The method for evaluating the passenger flow congestion risk of the track transfer station based on AFC card swiping data as claimed in claim 1, wherein the original AFC card swiping data screening and eliminating rules extracted by the steps 1 and 2 facing the passenger flow of the track transfer station are as follows:
1) removing transaction data of which the station entering time and the station exiting time are not on the same day;
2) eliminating the transaction data of which the outbound time is earlier than the inbound time;
3) eliminating records of the same inbound station and outbound station in the card swiping data records;
4) the data having the "DEAL _ STATUS transaction STATUS" field of "2" is screened out, and 2 represents that this record is in a transaction completed state.
4. The AFC card swiping data-based track transfer station passenger flow congestion risk evaluation method according to claim 1, wherein the index dimensionless method adopted in step 3 is equivalent to integrating original evaluation indexes into a comprehensive evaluation index, namely saturation, and comprehensively considering the relationship between passenger flow demand and supply at a typical position through saturation.
5. The AFC card swiping data based track transfer station passenger flow congestion risk evaluation method as claimed in claim 1, wherein the difference coefficient g isjThe larger, the more important the typical location; the normalization factor α takes 10 to bring the risk value range within 0-10.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611211608.4A CN106779429B (en) | 2016-12-25 | 2016-12-25 | Track transfer station passenger flow congestion risk evaluation method based on AFC card swiping data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611211608.4A CN106779429B (en) | 2016-12-25 | 2016-12-25 | Track transfer station passenger flow congestion risk evaluation method based on AFC card swiping data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106779429A CN106779429A (en) | 2017-05-31 |
CN106779429B true CN106779429B (en) | 2020-01-24 |
Family
ID=58920749
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611211608.4A Active CN106779429B (en) | 2016-12-25 | 2016-12-25 | Track transfer station passenger flow congestion risk evaluation method based on AFC card swiping data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106779429B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023118514A1 (en) * | 2021-12-24 | 2023-06-29 | Thales | System for monitoring the movement of passengers in a transport network, and associated method and computer program product |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107808235B (en) * | 2017-10-10 | 2020-06-02 | 东南大学 | AFC big data-based urban rail large passenger flow model construction method |
CN112749862A (en) * | 2019-10-31 | 2021-05-04 | 株洲中车时代电气股份有限公司 | Subway self-adaptive scheduling method based on passenger flow classification and electronic equipment |
CN110889092A (en) * | 2019-11-20 | 2020-03-17 | 北京市交通运行监测调度中心 | Short-time large-scale activity peripheral track station passenger flow volume prediction method based on track transaction data |
CN111063190B (en) * | 2019-12-10 | 2021-03-02 | 北京工业大学 | Pre-control method and system for oversaturated passenger flow of urban rail transit station platform |
CN111584091B (en) * | 2020-04-29 | 2023-10-24 | 北京交通大学 | Cross infection risk identification method and device for urban rail line-level close contact person |
CN112950943A (en) * | 2021-02-18 | 2021-06-11 | 重庆交通开投科技发展有限公司 | Transfer station calculation method based on multi-metadata |
CN113705623A (en) * | 2021-08-06 | 2021-11-26 | 深圳集智数字科技有限公司 | Rail transit station classification method and device |
CN115810271B (en) * | 2023-02-07 | 2023-04-28 | 安徽交欣科技股份有限公司 | Method for judging passenger flow corridor position based on card swiping data |
CN115952173A (en) * | 2023-03-13 | 2023-04-11 | 北京全路通信信号研究设计院集团有限公司 | Passenger flow data processing method and device, big data platform and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103955744A (en) * | 2014-04-23 | 2014-07-30 | 同济大学 | Method and device for parameter automatic calibration of rail transit passenger flow distributing model |
CN104376624A (en) * | 2014-07-22 | 2015-02-25 | 西南交通大学 | Urban rail transit passenger flow analysis method based on AFC (Automatic Fare Collection) passenger ticket data |
CN104765974A (en) * | 2015-04-24 | 2015-07-08 | 北京城建设计发展集团股份有限公司 | Urban rail transit passenger flow density index calculating method |
CN104809112A (en) * | 2014-01-23 | 2015-07-29 | 朱东霞 | Method for comprehensively evaluating urban public transportation development level based on multiple data |
-
2016
- 2016-12-25 CN CN201611211608.4A patent/CN106779429B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104809112A (en) * | 2014-01-23 | 2015-07-29 | 朱东霞 | Method for comprehensively evaluating urban public transportation development level based on multiple data |
CN103955744A (en) * | 2014-04-23 | 2014-07-30 | 同济大学 | Method and device for parameter automatic calibration of rail transit passenger flow distributing model |
CN104376624A (en) * | 2014-07-22 | 2015-02-25 | 西南交通大学 | Urban rail transit passenger flow analysis method based on AFC (Automatic Fare Collection) passenger ticket data |
CN104765974A (en) * | 2015-04-24 | 2015-07-08 | 北京城建设计发展集团股份有限公司 | Urban rail transit passenger flow density index calculating method |
Non-Patent Citations (3)
Title |
---|
《Experimental study on crowd flow passing through ticket gates in railway stations》;Kosuke Fujii 等;《ELSEVIER》;20141231;正文第630-635页 * |
《基于AFC数据的城轨站间客流量分布预测》;蔡昌俊等;《中国铁道科学》;20150131;第36卷(第1期);正文第126-132页 * |
《综合交通枢纽客流拥挤实时评价方法》;陈艳艳等;《公路交通科技》;20120731;第29卷;正文第75-80、120页 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023118514A1 (en) * | 2021-12-24 | 2023-06-29 | Thales | System for monitoring the movement of passengers in a transport network, and associated method and computer program product |
FR3131414A1 (en) * | 2021-12-24 | 2023-06-30 | Thales | SYSTEM FOR SUPERVISING THE MOVEMENT OF PASSENGERS IN A TRANSPORT NETWORK, METHOD AND PRODUCT, ASSOCIATED COMPUTER PROGRAM. |
Also Published As
Publication number | Publication date |
---|---|
CN106779429A (en) | 2017-05-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106779429B (en) | Track transfer station passenger flow congestion risk evaluation method based on AFC card swiping data | |
CN106651182B (en) | Entropy weight-based rail passenger flow congestion risk evaluation method | |
CN108961804B (en) | Bus route adjustment alternative set determination method based on multi-index classification intersection | |
CN105404850A (en) | Station passenger flow monitoring system, station passenger flow monitoring method, station passenger flow control system and station passenger flow control method | |
CN106845768A (en) | Bus hourage model building method based on survival analysis parameter distribution | |
CN105279572A (en) | City track traffic passenger flow density index calculating and releasing system | |
CN106651181B (en) | Method for evaluating bus passenger flow congestion risk under networked operation condition | |
CN111932429B (en) | OD-based rail transit station passenger flow structure similarity analysis method and device | |
CN114331234B (en) | Rail transit passenger flow prediction method and system based on passenger travel information | |
Li et al. | Dwell time estimation models for bus rapid transit stations | |
CN108805347B (en) | Passenger flow pool-based method for estimating passenger flow of associated area outside subway station | |
Bohari et al. | Improving the Quality of Public Transportation System: Application of simulation model for passenger movement | |
CN110889092A (en) | Short-time large-scale activity peripheral track station passenger flow volume prediction method based on track transaction data | |
CN115240431A (en) | Real-time online simulation system and method for traffic flow of highway toll station | |
CN109377050A (en) | One kind being based on risk analysis and AI artificial intelligence model risk analysis method | |
CN108665178B (en) | AFC-based passenger flow prediction method for stairs in subway station | |
Munizaga et al. | Fare evasion correction for smartcard-based origin-destination matrices | |
CN113643538B (en) | Bus passenger flow measuring and calculating method integrating IC card historical data and manual investigation data | |
CN114613128A (en) | Subway epidemic prevention monitoring management system based on data analysis | |
CN117035198A (en) | Urban rail transit station passenger flow state identification and control method | |
CN116415756B (en) | Urban virtual scene experience management system based on VR technology | |
CN115099685A (en) | Urban public transport service level evaluation method based on multi-source data | |
CN116187762A (en) | Method for identifying bottleneck risk of rail transit junction station based on analog | |
Zhuang et al. | Estimating escalator vs stairs choice behavior in the presence of entry railing: A field study | |
CN113780682B (en) | Rail transit system service fault prediction method based on fault markers |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20210428 Address after: 100080 319, 3rd floor, 29 Haidian West Street, Haidian District, Beijing Patentee after: Beijing Gewu botu Technology Co.,Ltd. Address before: 100124 Chaoyang District, Beijing Ping Park, No. 100 Patentee before: Beijing University of Technology |
|
TR01 | Transfer of patent right |