CN113393029A - Method and equipment for predicting rail transit passenger flow - Google Patents

Method and equipment for predicting rail transit passenger flow Download PDF

Info

Publication number
CN113393029A
CN113393029A CN202110651976.5A CN202110651976A CN113393029A CN 113393029 A CN113393029 A CN 113393029A CN 202110651976 A CN202110651976 A CN 202110651976A CN 113393029 A CN113393029 A CN 113393029A
Authority
CN
China
Prior art keywords
time
historical
station
target
people
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.)
Granted
Application number
CN202110651976.5A
Other languages
Chinese (zh)
Other versions
CN113393029B (en
Inventor
王新义
陆飞洋
吴昊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Yixun Information Technology Co ltd
Original Assignee
Shanghai Yixun Information Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Yixun Information Technology Co ltd filed Critical Shanghai Yixun Information Technology Co ltd
Priority to CN202110651976.5A priority Critical patent/CN113393029B/en
Publication of CN113393029A publication Critical patent/CN113393029A/en
Application granted granted Critical
Publication of CN113393029B publication Critical patent/CN113393029B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • 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

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Quality & Reliability (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Probability & Statistics with Applications (AREA)
  • Game Theory and Decision Science (AREA)
  • Algebra (AREA)
  • Development Economics (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (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 a method and a device for predicting rail transit passenger flow, wherein the method comprises the following steps: determining a target traffic pattern according to the passenger traffic characteristic elements at the future target time; determining a first probability, average time consumption of people and average time consumption of departure from each station and destination stations on a rail transit line according to passenger flow data corresponding to a destination flow mode in a preset historical time; determining a second probability of entering from each station at the historical entering time and exiting from the target station at the historical target time according to the average time consumed by people, the average difference of the time consumed by people and the internal time consumed by the historical station between the historical entering time and the historical target time; and determining the number of people who exit from the target station at the future target time according to the first probability, the second probability and the historical number of people who enter the station from each station at each historical entering time, so that a passenger flow predicted value accurate to ultra-fine time granularity can be obtained, and the accuracy and efficiency of rail transit passenger flow prediction are improved.

Description

Method and equipment for predicting rail transit passenger flow
Technical Field
The application relates to the technical field of rail transit big data processing, in particular to a method and equipment for predicting rail transit passenger flow.
Background
With the rapid development of cities, the range of activities of people in urban life is larger and larger, and rail transit plays an increasingly important role in urban traffic. The passenger flow analysis and prediction model based on the big data of the riding behavior of the user is a very important ring of the operation management and the passenger service of the rail transit enterprise and a very important demonstration project in the intelligent rail transit development outline. Scientific data is provided for all relevant departments, resources and manpower can be effectively distributed, and the safety, comfort and economic benefit of the whole traffic system are improved. The method can provide effective data support and decision basis for relevant departments to process emergency events, especially can help rail transit operation units to adjust and match the transportation capacity of corresponding passengers when large-scale activities are organized and the passenger flow prediction can help rail transit operation units to ensure that the activities are carried out smoothly and influence on other residents can be reduced.
There are various methods for predicting the traffic flow of the track, and the conventional methods include a historical Average algorithm, an ARIMA (automated Integrated Moving Average model), an Integrated Moving Average Autoregressive model, and the like. Recently, due to the rise of machine learning, there are new algorithms using a Long Short-Term Memory mechanism LSTM (Long Short-Term Memory Network) and a variant DCRNN (Diffusion Convolutional Neural Network) thereof. For the traditional method, the prediction is given by mostly only considering the flow information of the traffic flow acquisition point as a reference, the precision is insufficient, the deviation is large, and in addition, the emergency is worse; for a novel machine learning algorithm, a large amount of computing resources and computing time are needed, and the prediction length of a general machine learning algorithm is short, such as 1 hour, so that the requirements of traffic prediction on rapidness, long time and high efficiency cannot be met.
Therefore, how to improve the accuracy and efficiency of rail transit passenger flow volume prediction is a technical problem to be solved at present.
Disclosure of Invention
The invention provides a method for predicting rail transit passenger flow, which is used for solving the technical problems of insufficient precision, large deviation and low efficiency in rail transit passenger flow prediction in the prior art, and comprises the following steps:
determining a target traffic pattern according to the passenger traffic characteristic elements at the future target time;
determining a first probability, average time consumption of people and average time consumption of departure from each station and destination stations on a rail transit line according to passenger flow data corresponding to the destination flow pattern in a preset historical time;
determining a second probability of entering from each station at the historical entering time and exiting from the target station at the historical target time according to the average time of people, the average difference of time consumption and the historical in-station time consumption between the historical entering time and the historical target time;
determining the number of people who come out of the target station at the future target time according to the first probability, the second probability and the historical number of people who come in from the stations at the historical arrival time;
the passenger flow characteristic elements are determined according to preset dimensional characteristics and preset time marks which influence passenger flow, the historical arrival time is the time of arrival from each station every day in the target flow mode in the preset historical time, and the historical target time is the time corresponding to the future target time every day in the preset historical time.
Preferably, the first probability, the average time consumption of people and the average time consumption of the arrival from each station and the departure from the target station on the rail transit line are determined according to the passenger flow data corresponding to the target flow pattern within the preset historical time, specifically:
determining the first probability according to the ratio of the number of inbound people who enter the station from each station to the number of outbound people who enter the station from each station and exit the station from the target station in the passenger flow data;
determining the average time consumed by the people according to the ratio of the in-station time consumed sum of the number of the people who go out to the station;
and determining the average difference of the consumed time according to the average consumed time of people and the in-station consumed time for entering from each station and exiting from the target station.
Preferably, the second probability is determined according to a first formula, which is specifically:
Figure BDA0003111918910000021
wherein the content of the first and second substances,
Figure BDA0003111918910000022
is the second probability, σ (Z)m，Zi) Mu (Z) for the time-consuming averagingm，Zi) The time is consumed for the people
Figure BDA0003111918910000023
Is the time consumption in the history station.
Preferably, the determining, according to the first probability, the second probability and the historical number of people entering the station from each of the historical entry times, the number of people exiting the target station at the future target time is specifically:
determining all historical number of people who enter the station at the historical entering time and exit the target station at the historical target time according to the first probability, the second probability and the historical number of people who enter the station;
and summing the historical number of people going out from each historical arrival time to the historical target time, and determining the number of people going out from the target station at the future target time according to the result of the summation.
Preferably, the historical number of people leaving the station is determined according to a formula two, wherein the formula two specifically comprises:
Figure BDA0003111918910000031
out (Z)i，tx，tw) M represents each of the sites, n is the number of sites,
Figure BDA0003111918910000032
for the number of said historical arrival people, P (Z)m，Zi) Is the first probability, ZiRepresenting said target site i, txRepresenting historical arrival times x, twRepresenting the historical target time w.
Correspondingly, the invention also provides a device for predicting rail transit passenger flow, which comprises:
the first determining module is used for determining a target traffic mode according to the passenger traffic characteristic elements at the future target time;
the second determining module is used for determining a first probability, average time consumption of people and average time consumption difference of entering from each station and exiting from the target station on the rail transit line according to the passenger flow data corresponding to the target flow mode in the preset historical time;
a third determining module, configured to determine, according to the average time spent by people, the average difference in time spent, and the historical intra-site time spent between the historical entry time and the historical target time, a second probability of entering from each of the sites at the historical entry time and exiting from the target site at the historical target time;
a fourth determining module, configured to determine, according to the first probability, the second probability, and historical number of people entering from each of the sites at each of the historical arrival times, the number of people exiting from the target site at the future target time;
the passenger flow characteristic elements are determined according to preset dimensional characteristics and preset time marks which influence passenger flow, the historical arrival time is the time of arrival from each station every day in the target flow mode in the preset historical time, and the historical target time is the time corresponding to the future target time every day in the preset historical time.
Preferably, the second determining module is specifically configured to:
determining the first probability according to the ratio of the number of inbound people who enter the station from each station to the number of outbound people who enter the station from each station and exit the station from the target station in the passenger flow data;
determining the average time consumed by the people according to the ratio of the in-station time consumed sum of the number of the people who go out to the station;
and determining the average difference of the consumed time according to the average consumed time of people and the in-station consumed time for entering from each station and exiting from the target station.
Preferably, the third determining module is specifically configured to:
determining the second probability according to a first formula, wherein the first formula specifically is as follows:
Figure BDA0003111918910000041
wherein the content of the first and second substances,
Figure BDA0003111918910000042
is the second probability, σ (Z)m，Zi) Mu (Z) for the time-consuming averagingm，Zi) It takes time for the person to do so,
Figure BDA0003111918910000045
is the time consumption in the history station.
Preferably, the fourth determining module is specifically configured to:
determining all historical number of people who enter the station at the historical entering time and exit the target station at the historical target time according to the first probability, the second probability and the historical number of people who enter the station;
and summing the historical number of people going out from each historical arrival time to the historical target time, and determining the number of people going out from the target station at the future target time according to the result of the summation.
Preferably, the fourth determining module is further specifically configured to:
determining the historical number of people leaving the station according to a formula II, wherein the formula II specifically comprises the following steps:
Figure BDA0003111918910000043
out (Z)i，tx，tw) M represents each of the sites, n is the number of sites,
Figure BDA0003111918910000044
for the number of said historical arrival people, P (Z)m，Zi) Is the first probability, ZiRepresenting said target site i, txRepresenting historical arrival times x, twRepresenting the historical target time w.
Compared with the prior art, the invention has the following beneficial effects:
by applying the technical scheme, a target traffic mode is determined according to the passenger traffic characteristic elements at the future target time; determining a first probability, average time consumption of people and average time consumption of departure from each station and destination stations on a rail transit line according to passenger flow data corresponding to the destination flow pattern in a preset historical time; determining a second probability of entering from each station at the historical entering time and exiting from the target station at the historical target time according to the average time of people, the average difference of time consumption and the historical in-station time consumption between the historical entering time and the historical target time; determining the number of people who come out of the target station at the future target time according to the first probability, the second probability and the historical number of people who come in from the stations at the historical arrival time; the passenger flow characteristic elements are determined according to preset dimensional characteristics and preset time marks which influence passenger flow, the historical arrival time is the time of arrival from each station every day in the target flow mode in the preset historical duration, the historical target time is the time corresponding to the future target time every day in the preset historical duration, and the outbound passenger flow is obtained in a mode of applying probability density to the arrival passenger flow and the flow mode on the basis of combining the passenger flow characteristic elements, so that a passenger flow predicted value accurate to ultra-fine time granularity can be obtained, the accuracy of rail transit passenger flow prediction is improved, less external data is relied on, and the prediction efficiency is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flow chart illustrating a method for predicting rail transit passenger flow according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating different traffic patterns in an embodiment of the invention;
FIG. 3 is a schematic diagram illustrating the probability of station ingress and egress in an embodiment of the invention;
FIG. 4 is a table showing the probability distribution of inbound and outbound traffic in an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating average elapsed time in an embodiment of the present invention;
FIG. 6 is a table showing the average elapsed time distribution according to an embodiment of the present invention;
FIG. 7 is a schematic diagram illustrating time-consuming averaging in an embodiment of the present invention;
FIG. 8 is a table showing the time-consuming averaging distribution according to an embodiment of the present invention;
FIG. 9 illustrates a time-consuming profile feature in an embodiment of the present invention;
fig. 10 is a schematic structural diagram illustrating a device for predicting rail transit traffic volume according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The embodiment of the application provides a method for predicting rail transit passenger flow, as shown in fig. 1, comprising the following steps:
and step S101, determining a target traffic pattern according to the passenger traffic characteristic elements at the future target time.
Rail transit is rail transit, and in the embodiment, the passenger flow volume of a target station at a future target time at a future preset date is predicted, the future target time is a time (such as 13:00) of a certain day in the future, the passenger flow volume of each station on a rail transit line is related to various factors, the passenger flow volume characteristic elements are determined according to preset dimensional characteristics and preset time marks, the preset dimensional characteristics comprise dates, weather, holidays and event days, the event days can comprise large conference holding days, large sports event holding days, large concert holding days, large exhibition holding days and the like, and the preset time marks comprise early peak, late peak and daily time.
Optionally, the passenger flow characteristic element is a characteristic element related to the destination station or a characteristic element related to the rail transit line. Specifically, when the distance of the rail transit line is long, and each station is far away, some passenger flow characteristic elements such as large-scale sports activities may only affect a part of stations, and the passenger flow characteristic elements are characteristic elements related to the target stations at the moment; when the distance of the rail-crossing line is short, the passenger flow characteristic element can influence the passenger flow of the whole rail-crossing line, and the passenger flow characteristic element is a characteristic element related to the rail-crossing line.
In this embodiment, a plurality of preset traffic patterns, such as a working day plus sunny day traffic pattern, a holiday plus rainy day traffic pattern, and the like, are determined in advance according to different combinations of the passenger traffic characteristic elements, a correspondence between different passenger traffic characteristic elements and the preset traffic patterns is established, and a target traffic pattern is determined from the plurality of preset traffic patterns according to the passenger traffic characteristic elements at a future target time and the correspondence.
Those skilled in the art can determine various preset flow modes according to actual needs.
And S102, determining a first probability, average time consumption of people and average time consumption of departure from each station and each station on the railway line according to the passenger flow data corresponding to the target flow mode in the preset historical time.
In this embodiment, historical passenger flow volume data within a preset historical duration may be obtained from a historical database, where the historical passenger flow volume data may include passenger flow volume data in multiple flow volume modes, the passenger flow volume data corresponding to the target flow volume mode is selected from the historical passenger flow volume data, and a first probability, a man-shared time consumption, and a time consumption difference of entering from each station and exiting from the target station on the rail transit line are determined according to the passenger flow volume data.
In order to accurately determine the first probability, the average time-consuming per capita and the average time-consuming per capita, in a preferred embodiment of the present application, the first probability, the average time-consuming per capita and the average time-consuming per capita of the train entering from each station and exiting from the target station on the rail transit line are determined according to the passenger traffic data corresponding to the target traffic pattern within a preset historical time, specifically:
determining the first probability according to the ratio of the number of inbound people who enter the station from each station to the number of outbound people who enter the station from each station and exit the station from the target station in the passenger flow data;
determining the average time consumed by the people according to the ratio of the in-station time consumed sum of the number of the people who go out to the station;
and determining the average difference of the consumed time according to the average consumed time of people and the in-station consumed time for entering from each station and exiting from the target station.
In the embodiment, the number of inbound people who enter the station from each station and the number of outbound people who enter the station from each station and exit the target station are determined according to passenger flow data, and a first probability can be determined according to the ratio of the number of inbound people and the number of outbound people; determining the station time consumption sum of the number of people who come out according to the passenger flow data, and determining the average time consumption of people according to the ratio of the station time consumption sum to the number of people who come out; the difference of the consumed time can be determined according to the average consumed time of people and the consumed time in the station which enters the station and leaves the station from the target station.
It should be noted that the above solution of the preferred embodiment is only a specific implementation solution proposed in the present application, and other ways of determining the first probability, the time consumption for everyone, and the time consumption difference according to the passenger flow data all belong to the protection scope of the present application.
In order to accurately determine the time-consuming average difference, in a preferred embodiment of the present application, the time-consuming average difference is determined according to a formula four, where the formula four specifically is:
Figure BDA0003111918910000071
wherein x isiAnd representing the time consumption in the station, N representing the number of the time consumption in the station, sigma being the average difference of the time consumption, and mu being the average time consumption of people.
Step S103, determining a second probability of entering from each station at the historical entering time and exiting from the target station at the historical target time according to the average time of people, the average difference of time consumption and the historical in-station time consumption between the historical entering time and the historical target time.
In this embodiment, the historical arrival time is the time of arrival from each station every day in the target flow mode within the preset historical duration, and the historical target time is the time corresponding to the future target time every day within the preset historical duration, for example, if the future target time is 13:00, the historical target time is also 13: 00. And entering the station from each historical arrival time and exiting the station from the corresponding historical target time, wherein the consumed time in the in-station period is the consumed time in the historical station. And determining a second probability of entering from each station at the historical entering time and exiting from the target station at the historical target time according to the average time consumption, the average difference of time consumption and the internal time consumption of the historical station.
In order to accurately determine the second probability, in a preferred embodiment of the present application, the second probability is determined according to a first formula, which specifically is:
Figure BDA0003111918910000081
wherein the content of the first and second substances,
Figure BDA0003111918910000082
is the second probability, σ (Z)m，Zi) Mu (Z) for the time-consuming averagingm，Zi) It takes time for the person to do so,
Figure BDA0003111918910000083
is the time consumption in the history station.
It should be noted that the above solution of the preferred embodiment is only a specific implementation solution proposed in the present application, and other ways of determining the second probability according to the average time consumption, the average difference time consumption and the time consumption in the historical station all belong to the protection scope of the present application.
And step S104, determining the number of people going out from the target station at the future target time according to the first probability, the second probability and the historical number of people going in from the stations at the historical arrival time.
In this embodiment, the historical number of people entering the station from each station at each historical arrival time is determined according to the passenger flow data, and the number of people leaving the station from the target station at a future target time can be determined according to the first probability, the second probability and the historical number of people entering the station.
In order to accurately determine the number of people who exit from the target station at the future target time, in a preferred embodiment of the present application, the number of people who enter from the target station at the future target time is determined according to the first probability, the second probability and the historical number of people who enter from the respective station at the respective historical entry time, specifically:
determining all historical number of people who enter the station at the historical entering time and exit the target station at the historical target time according to the first probability, the second probability and the historical number of people who enter the station;
and summing the historical number of people going out from each historical arrival time to the historical target time, and determining the number of people going out from the target station at the future target time according to the result of the summation.
In this embodiment, all the historical number of people who enter the station from each station at the historical entry time and exit the station from the target station at the historical target time are determined according to the first probability, the second probability and the historical number of people who enter the station, and after summing up the historical number of people who exit the station from each historical entry time to the historical target time, the result of the summation is taken as the number of people who exit the station from the target station at the future target time.
It should be noted that the above solution of the preferred embodiment is only one specific implementation solution proposed in the present application, and other ways of determining the number of people going out from the target site at the future target time according to the first probability, the second probability and the historical number of people coming in from the target site all belong to the protection scope of the present application.
In order to accurately determine the historical number of people who come out, in a preferred embodiment of the present application, the historical number of people who come out is determined according to a formula two, where the formula two is specifically:
Figure BDA0003111918910000091
out (Z)i，tx，tw) M represents each of the sites, n is the number of sites,
Figure BDA0003111918910000092
for the number of said historical arrival people, P (Z)m，Zi) Is the first probability, ZiRepresenting said target site i, txRepresenting historical arrival times x, twRepresenting the historical target time w.
It should be noted that the above solution of the preferred embodiment is only one specific implementation solution proposed in the present application, and other ways of determining the historical number of outbound persons according to the first probability, the second probability and the historical number of inbound persons all belong to the protection scope of the present application.
In order to accurately determine the number of people who will exit from the target site at the future target time, in a preferred embodiment of the present application, the number of people who will exit from the target site at the future target time is determined according to the formula three:
Figure BDA0003111918910000093
out (Z)i，tw) The number of people who are outbound from the destination site for the future destination time.
By applying the technical scheme, a target traffic mode is determined according to the passenger traffic characteristic elements at the future target time; determining a first probability, average time consumption of people and average time consumption of departure from each station and destination stations on a rail transit line according to passenger flow data corresponding to the destination flow pattern in a preset historical time; determining a second probability of entering from each station at the historical entering time and exiting from the target station at the historical target time according to the average time of people, the average difference of time consumption and the historical in-station time consumption between the historical entering time and the historical target time; determining the number of people who come out of the target station at the future target time according to the first probability, the second probability and the historical number of people who come in from the stations at the historical arrival time; the passenger flow characteristic elements are determined according to preset dimensional characteristics and preset time marks which influence passenger flow, the historical arrival time is the time of arrival from each station every day in the target flow mode in the preset historical duration, the historical target time is the time corresponding to the future target time every day in the preset historical duration, and the outbound passenger flow is obtained in a mode of applying probability density to the arrival passenger flow and the flow mode on the basis of combining the passenger flow characteristic elements, so that a passenger flow predicted value accurate to ultra-fine time granularity can be obtained, the accuracy of rail transit passenger flow prediction is improved, less external data is relied on, and the prediction efficiency is improved.
In order to further illustrate the technical idea of the present invention, the technical solution of the present invention will now be described with reference to specific application scenarios.
The embodiment of the application provides a method for predicting rail transit passenger flow, which comprises the following steps:
and firstly, acquiring passenger flow volume data in a preset historical time.
(II) flow rate mode
Based on passenger flow volume data in a preset historical time, dimensional characteristics such as date, weather, holidays and the like are added, and time (morning and evening peak, daily time) marks are introduced to construct different flow pattern groups, as shown in fig. 2.
(III) probability of station entrance/exit
P (Zi, Zj) is a slave station ZiThen station ZjThe probability of outbound is shown in fig. 3 as a diagram of the probability of inbound and outbound.
Slave station Z under different flow calculation modeiStation ZjProbability of outbound:
Figure BDA0003111918910000101
wherein:
NZifor access station Z within a preset historical time periodiThe number of inbound people;
M(Zi，Zj) For slave station Z within preset history durationiStation ZjThe number of people who are out of the station.
Fig. 4 shows an ingress and egress probability distribution table.
(IV) travel time consumption distribution
1. Average time consumption
μ(Zi，Zj): from station ZiTo station ZjThe average elapsed time of (2) is shown in fig. 5 as a graph of the average elapsed time.
Slave station Z under different flow calculation modeiStation ZjAverage time spent outbound:
Figure BDA0003111918910000111
wherein:
sum (time) from station ZiStation ZjThe sum of the time spent by all people outbound;
M(Zi，Zj): slave station Z in preset historical timeiStation ZjThe number of people who are out of the station.
Fig. 6 shows an average elapsed time distribution table.
2. Time-consuming mean difference
σ(Zi，Zj): from station ZiStation ZjThe average difference of the outbound time consumption distribution is shown in fig. 7 as a time consumption average difference diagram.
Figure BDA0003111918910000112
Wherein x isiAnd representing the time consumption in each station, N representing the number of the time consumption in each station, sigma representing the average difference of the time consumption, and mu representing the average time consumption of people.
FIG. 8 shows a time-consuming averaging distribution table.
3. Time-consuming profile characteristics
By analyzing the time-consuming distribution, it was found to substantially conform to the normal distribution, as shown in FIG. 9.
(V) outbound traffic prediction
The target traffic pattern is determined from the traffic characteristic element at "certain time w" in the future, and each parameter in the following steps is determined based on the traffic data corresponding to the target traffic pattern.
Figure BDA0003111918910000113
Wherein the content of the first and second substances,
Figure BDA0003111918910000114
for entering station m at time x and leaving station i at time wThe probability of the station, i.e. the second probability;
σ(Zm，Zi) The time consumption difference of station m entering and station i leaving is obtained;
μ(Zm，Zi) All people who enter the station m and exit the station i consume time;
Figure BDA0003111918910000115
is from time txTo time twThe time length is the time consumed in the historical station;
m is more than or equal to 1 and less than or equal to n, and n is the number of sites.
Figure BDA0003111918910000121
Out (Z)i，tx，tw) The historical number of people who arrive at the station from each station at the time x and leave the station i at the time w is the number of people who arrive at the station;
Figure BDA0003111918910000122
at a time txThe number of people who arrive at the station from the station m is the historical number of people who arrive at the station;
P(Zm，Zi) The probability of station m entering and station i leaving is the first probability;
Zirepresenting the target site i;
txrepresenting historical arrival times x, twRepresents the historical target time w, the time x is the time txTime w is instant tw。
Figure BDA0003111918910000123
Out (Z)i，tw) The number of people coming out from the station i for the future "certain moment w", i.e. the number of people coming out from the target station for the future target moment.
Corresponding to the method for predicting the rail transit passenger flow in the embodiment of the present application, an embodiment of the present application further provides a device for predicting the rail transit passenger flow, as shown in fig. 10, where the device includes:
a first determining module 201, configured to determine a target traffic pattern according to the passenger traffic characteristic element at a future target time;
a second determining module 202, configured to determine, according to passenger flow data corresponding to the target flow pattern within a preset historical time, a first probability, a per-person consumed time, and a consumed time difference of entering from each station and exiting from a target station on a rail transit line;
a third determining module 203, configured to determine a second probability of entering from each station at the historical entering time and exiting from the target station at the historical target time according to the average time of people, the average difference of time consumption, and the historical in-station elapsed time between the historical entering time and the historical target time;
a fourth determining module 204, configured to determine the number of people who get out from the target site at the future target time according to the first probability, the second probability, and the historical number of people who get in from the sites at the historical arrival times;
the passenger flow characteristic elements are determined according to preset dimensional characteristics and preset time marks which influence passenger flow, the historical arrival time is the time of arrival from each station every day in the target flow mode in the preset historical time, and the historical target time is the time corresponding to the future target time every day in the preset historical time.
In a specific application scenario of the present application, the second determining module 202 is specifically configured to:
determining the first probability according to the ratio of the number of inbound people who enter the station from each station to the number of outbound people who enter the station from each station and exit the station from the target station in the passenger flow data;
determining the average time consumed by the people according to the ratio of the in-station time consumed sum of the number of the people who go out to the station;
and determining the average difference of the consumed time according to the average consumed time of people and the in-station consumed time for entering from each station and exiting from the target station.
In a specific application scenario of the present application, the third determining module 203 is specifically configured to:
determining the second probability according to a first formula, wherein the first formula specifically is as follows:
Figure BDA0003111918910000131
wherein the content of the first and second substances,
Figure BDA0003111918910000132
is the second probability, σ (Z)m，Zi) Mu (Z) for the time-consuming averagingm，Zi) It takes time for the person to do so,
Figure BDA0003111918910000135
is the time consumption in the history station.
In a specific application scenario of the present application, the fourth determining module 204 is specifically configured to:
determining all historical number of people who enter the station at the historical entering time and exit the target station at the historical target time according to the first probability, the second probability and the historical number of people who enter the station;
and summing the historical number of people going out from each historical arrival time to the historical target time, and determining the number of people going out from the target station at the future target time according to the result of the summation.
In a specific application scenario of the present application, the fourth determining module 204 is further specifically configured to:
determining the historical number of people leaving the station according to a formula II, wherein the formula II specifically comprises the following steps:
Figure BDA0003111918910000133
out (Z)i，tx，tw) M represents each of the sites, n is the number of sites,
Figure BDA0003111918910000134
for the number of said historical arrival people, P (Z)m，Zi) Is the first probability, ZiRepresenting said target site i, txRepresenting historical arrival times x, twRepresenting the historical target time w.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not necessarily depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims (10)

1. A method for predicting rail transit passenger flow, the method comprising:
determining a target traffic pattern according to the passenger traffic characteristic elements at the future target time;
determining a first probability, average time consumption of people and average time consumption of departure from each station and destination stations on a rail transit line according to passenger flow data corresponding to the destination flow pattern in a preset historical time;
determining a second probability of entering from each station at the historical entering time and exiting from the target station at the historical target time according to the average time of people, the average difference of time consumption and the historical in-station time consumption between the historical entering time and the historical target time;
determining the number of people who come out of the target station at the future target time according to the first probability, the second probability and the historical number of people who come in from the stations at the historical arrival time;
the passenger flow characteristic elements are determined according to preset dimensional characteristics and preset time marks which influence passenger flow, the historical arrival time is the time of arrival from each station every day in the target flow mode in the preset historical time, and the historical target time is the time corresponding to the future target time every day in the preset historical time.
2. The method according to claim 1, wherein the first probability, the average time consumption for people and the average time consumption for departure from each station on the rail transit line are determined according to the passenger flow data corresponding to the target flow pattern within a preset historical time, specifically:
determining the first probability according to the ratio of the number of inbound people who enter the station from each station to the number of outbound people who enter the station from each station and exit the station from the target station in the passenger flow data;
determining the average time consumed by the people according to the ratio of the in-station time consumed sum of the number of the people who go out to the station;
and determining the average difference of the consumed time according to the average consumed time of people and the in-station consumed time for entering from each station and exiting from the target station.
3. The method of claim 2, wherein the second probability is determined according to formula one, in particular:
Figure FDA0003111918900000011
wherein the content of the first and second substances,
Figure FDA0003111918900000021
is the second probability, σ (Z)m，Zi) Mu (Z) for the time-consuming averagingm，Zi) It takes time for the person to do so,
Figure FDA0003111918900000022
is the time consumption in the history station.
4. The method of claim 3, wherein determining the number of people to pull from the target site at the future target time based on the first probability, the second probability, and the historical number of people to pull from the sites at the historical pull-in times comprises:
determining all historical number of people who enter the station at the historical entering time and exit the target station at the historical target time according to the first probability, the second probability and the historical number of people who enter the station;
and summing the historical number of people going out from each historical arrival time to the historical target time, and determining the number of people going out from the target station at the future target time according to the result of the summation.
5. The method of claim 4, wherein the historical number of outbound people is determined according to a formula two, the formula two being specifically:
Figure FDA0003111918900000023
out (Z)i，tx，tw) M represents each of the sites, n is the number of sites,
Figure FDA0003111918900000024
for the number of said historical arrival people, P (Z)m，Zi) Is the first probability, ZiRepresenting said target site i, txRepresenting historical arrival times x, twRepresenting the historical target time w.
6. An apparatus for predicting rail transit passenger flow, the apparatus comprising:
the first determining module is used for determining a target traffic mode according to the passenger traffic characteristic elements at the future target time;
the second determining module is used for determining a first probability, average time consumption of people and average time consumption difference of entering from each station and exiting from the target station on the rail transit line according to the passenger flow data corresponding to the target flow mode in the preset historical time;
a third determining module, configured to determine, according to the average time spent by people, the average difference in time spent, and the historical intra-site time spent between the historical entry time and the historical target time, a second probability of entering from each of the sites at the historical entry time and exiting from the target site at the historical target time;
a fourth determining module, configured to determine, according to the first probability, the second probability, and historical number of people entering from each of the sites at each of the historical arrival times, the number of people exiting from the target site at the future target time;
the passenger flow characteristic elements are determined according to preset dimensional characteristics and preset time marks which influence passenger flow, the historical arrival time is the time of arrival from each station every day in the target flow mode in the preset historical time, and the historical target time is the time corresponding to the future target time every day in the preset historical time.
7. The device of claim 6, wherein the second determining module is specifically configured to:
determining the first probability according to the ratio of the number of inbound people who enter the station from each station to the number of outbound people who enter the station from each station and exit the station from the target station in the passenger flow data;
determining the average time consumed by the people according to the ratio of the in-station time consumed sum of the number of the people who go out to the station;
and determining the average difference of the consumed time according to the average consumed time of people and the in-station consumed time for entering from each station and exiting from the target station.
8. The device of claim 7, wherein the third determining module is specifically configured to:
determining the second probability according to a first formula, wherein the first formula specifically is as follows:
Figure FDA0003111918900000031
wherein the content of the first and second substances,
Figure FDA0003111918900000032
is the second probability, σ (Z)m，Zi) Mu (Z) for the time-consuming averagingm，Zi) It takes time for the person to do so,
Figure FDA0003111918900000033
is the time consumption in the history station.
9. The device of claim 8, wherein the fourth determining module is specifically configured to:
determining all historical number of people who enter the station at the historical entering time and exit the target station at the historical target time according to the first probability, the second probability and the historical number of people who enter the station;
and summing the historical number of people going out from each historical arrival time to the historical target time, and determining the number of people going out from the target station at the future target time according to the result of the summation.
10. The device of claim 9, wherein the fourth determining module is further specifically configured to:
determining the historical number of people leaving the station according to a formula II, wherein the formula II specifically comprises the following steps:
Figure FDA0003111918900000041
out (Z)i，tx，tw) M represents each of the sites, n is the number of sites,
Figure FDA0003111918900000042
for the number of said historical arrival people, P (Z)m，Zi) Is the first probability, ZiRepresenting said target site i, txRepresenting historical arrival times x, twRepresenting the historical target time w.
CN202110651976.5A 2021-06-11 2021-06-11 Method and equipment for predicting rail transit passenger flow Active CN113393029B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110651976.5A CN113393029B (en) 2021-06-11 2021-06-11 Method and equipment for predicting rail transit passenger flow

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110651976.5A CN113393029B (en) 2021-06-11 2021-06-11 Method and equipment for predicting rail transit passenger flow

Publications (2)

Publication Number Publication Date
CN113393029A true CN113393029A (en) 2021-09-14
CN113393029B CN113393029B (en) 2023-11-17

Family

ID=77620407

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110651976.5A Active CN113393029B (en) 2021-06-11 2021-06-11 Method and equipment for predicting rail transit passenger flow

Country Status (1)

Country Link
CN (1) CN113393029B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117808184A (en) * 2024-02-29 2024-04-02 成都汇辙科技有限公司 Urban rail transit service management system driven by full production elements

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106485359A (en) * 2016-10-13 2017-03-08 东南大学 A kind of urban track traffic section passenger flow estimation method based on train schedule
WO2020019976A1 (en) * 2018-07-27 2020-01-30 阿里巴巴集团控股有限公司 Expected duration interval determining method and apparatus, and travel verification method and apparatus
CN111582605A (en) * 2020-05-21 2020-08-25 Oppo广东移动通信有限公司 Method and device for predicting destination site, electronic equipment and storage medium
CN112488388A (en) * 2020-11-30 2021-03-12 佳都新太科技股份有限公司 Outbound passenger flow prediction method and device based on probability distribution
CN112686417A (en) * 2019-10-18 2021-04-20 深圳先进技术研究院 Subway large passenger flow prediction method and system and electronic equipment

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106485359A (en) * 2016-10-13 2017-03-08 东南大学 A kind of urban track traffic section passenger flow estimation method based on train schedule
WO2020019976A1 (en) * 2018-07-27 2020-01-30 阿里巴巴集团控股有限公司 Expected duration interval determining method and apparatus, and travel verification method and apparatus
CN112686417A (en) * 2019-10-18 2021-04-20 深圳先进技术研究院 Subway large passenger flow prediction method and system and electronic equipment
CN111582605A (en) * 2020-05-21 2020-08-25 Oppo广东移动通信有限公司 Method and device for predicting destination site, electronic equipment and storage medium
CN112488388A (en) * 2020-11-30 2021-03-12 佳都新太科技股份有限公司 Outbound passenger flow prediction method and device based on probability distribution

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117808184A (en) * 2024-02-29 2024-04-02 成都汇辙科技有限公司 Urban rail transit service management system driven by full production elements
CN117808184B (en) * 2024-02-29 2024-05-10 成都汇辙科技有限公司 Urban rail transit service management system driven by full production elements

Also Published As

Publication number Publication date
CN113393029B (en) 2023-11-17

Similar Documents

Publication Publication Date Title
CN110047279B (en) Method for determining shared bicycle dispatching quantity based on order data
Toqué et al. Short & long term forecasting of multimodal transport passenger flows with machine learning methods
CN109034566B (en) A kind of intelligent dispatching method and device based on passenger flow above and below bus station
Wang et al. Passenger-demands-oriented train scheduling for an urban rail transit network
CN105575108B (en) A kind of intelligent public transportation dispatching method for running
CN103632212B (en) System and method for predicating time-varying user dynamic equilibrium network-evolved passenger flow
Chen et al. Uncertainty in urban mobility: Predicting waiting times for shared bicycles and parking lots
CN107358319A (en) Flow Prediction in Urban Mass Transit method, apparatus, storage medium and computer equipment
CN107564270A (en) A kind of intelligent public transportation dispatching method for running
CN113920769B (en) Intelligent scheduling list generation method based on public transportation multi-source data analysis
Huang et al. A novel bus-dispatching model based on passenger flow and arrival time prediction
CN109872535A (en) A kind of current prediction technique of wisdom traffic, device and server
Liu et al. Smart card data-centric replication of the multi-modal public transport system in Singapore
CN110444008B (en) Vehicle scheduling method and device
Jansen et al. Minimizing passenger transfer times in public transport timetables
CN114037158A (en) Passenger flow prediction method based on OD path and application method
CN110211379A (en) A kind of public transport method for optimizing scheduling based on machine learning
CN111882097A (en) Vehicle maintenance personnel scheduling method and system
CN115860594A (en) Simulation system and method applied to intelligent bus scheduling
CN113393029A (en) Method and equipment for predicting rail transit passenger flow
CN115759639A (en) Non-renting high-sending-period shared bicycle scheduling method based on small sample increment learning
CN109409563B (en) Method, system and storage medium for analyzing real-time number of people in public transport operation vehicle
Vlasov et al. Ground urban passenger transport in a period of digital transformation
Rusca et al. Algorithm for traffic allocation when are developed park and ride facilities
CN113345216B (en) Vehicle scheduling method for large-scale activities

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