CN106960261A - A kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data - Google Patents
A kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data Download PDFInfo
- Publication number
- CN106960261A CN106960261A CN201710191811.8A CN201710191811A CN106960261A CN 106960261 A CN106960261 A CN 106960261A CN 201710191811 A CN201710191811 A CN 201710191811A CN 106960261 A CN106960261 A CN 106960261A
- Authority
- CN
- China
- Prior art keywords
- passenger flow
- passenger
- varchar
- flow
- circuit
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 230000011664 signaling Effects 0.000 title claims abstract description 11
- 238000012546 transfer Methods 0.000 claims abstract description 28
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 230000008859 change Effects 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000002123 temporal effect Effects 0.000 claims description 6
- 230000007306 turnover Effects 0.000 claims description 6
- 238000013459 approach Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims 1
- 230000008520 organization Effects 0.000 abstract description 3
- 238000010276 construction Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000000205 computational method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000004321 preservation Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
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"
-
- G06Q50/40—
Abstract
The present invention relates to a kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data, this method comprises the following steps：S1：Acquisition trajectory traffic passenger's IC-card data message, OD matchings are carried out to preprocessed data；S2：Commuting identification, inputs multiple workaday data, is grouped by user, judges the commuting rule of each user in multiple working days：When the user is identical, enter the station website and outbound website number of days reaches the 50% of total number of days, you can it is that the user is commuter to judge the trip；S3：Track traffic for passenger flow is counted, statistical indicator includes：Website passenger flow, circuit passenger flow, section passenger flow, transfer passenger flow, passenger flow OD, commute website passenger flow, and commute circuit passenger flow, and commute website OD passenger flows, and commute circuit OD passenger flows；S4：The analysis of the normality volume of the flow of passengers is carried out to analyze with the festivals or holidays volume of the flow of passengers；S5：Carry out normality passenger flow forecast and festivals or holidays passenger flow forecast.This method is capable of the volume of the flow of passengers of more precisely predicted orbit traffic, and rationally establishment conveyance equilibrium scheme, passenger organization scheme etc. are played an important role.
Description
Technical field
The invention belongs to data analysis technique field, particularly track traffic operation field, it is related to a kind of based on track IC
The track traffic for passenger flow Forecasting Methodology of card and mobile phone signaling data.
Background technology
Into 21 century, continuous with Chinese Urbanization level, urban population density will be increased rapidly, resulting
Traffic jam issue also can be increasingly severeer, and this is by as one of restriction economy, society, the key factor of cultural development.City
Track traffic as public transport important component, by as solve urban traffic blocking effective mode of transportation.Due to
Track traffic have energy-conservation, save ground, freight volume it is big, round-the-clock, on schedule, it is pollution-free and safe the features such as, it can be deduced that such knot
By：The development security compliance continuable development principle of urban track traffic.Therefore following a very long time would is that city rail
The booming gold period of road traffic.According to data, the city to China's operation track traffic of the end of the year 2016 reaches 26,
Run 3748.67 kilometers of total kilometrage.Wherein the track traffic of Shanghai Operating cities (including city underground and light rail, similarly hereinafter) overall length reaches
To 617 kilometers, it has also become operation rail line most long city in the world at present.554 kilometers, Guangzhou fortune are runed in Beijing
308 kilometers have been sought, world's operation track traffic most long city ranks are all entered.Now, track traffic has become many
Indispensable in city, the topmost vehicles.According to long term planning, to the year two thousand twenty National urban track traffic total kilometrage
6100 kilometers are up to, the wherein total kilometrage of Beijing's urban track traffic will be more than 1000 kilometers, and public transport will turn into
80% or so civic is mainly gone on a journey the vehicles, wherein the people for having 50% or so will be commonly using urban track traffic as going out
Row instrument.
Urban track traffic is as Public Welfare facility, and the development model for affecting city's spatial structure is advised with global
Draw, the integrated system that Construction of Urban Rail Traffic is cycle length, involved a wide range of knowledge.Wherein, Flow Prediction in Urban Mass Transit conduct
The investment decision basis of the system, can not only weigh construction project financial cost, and be after prediction construction project puts into effect
The key index of economic benefit.Passenger flow estimation process is also the process for blending the advanced communication technology and information technology.
With the cumulative year after year of the volume of the flow of passengers, although urban mass transit network constantly expands and perfect, but is still difficult to
Realize the equilibrium of supply and demand.On weekdays during morning, evening peak period and great festivals or holidays, passenger flow hypersaturated state turns into city rail
Normality phenomenon in traffic operation.Science is carried out to track traffic for passenger flow and accurately predicted so that conveyance equilibrium is rationally worked out
Scheme, the early warning of passenger organization scheme and accident and dispersal plan are possibly realized, so as to sufficiently utilize city
In existing track traffic resource.
Passenger flow estimation determines all designs and the selection of key element in City Rail Transit System, such as：The traffic capacity
Size, the mode of rail layout, the selection of train model, signal control and construction costs etc., income after even runing and
Efficiency of service.The accuracy of passenger flow estimation result plays very important in scientific and efficient City Rail Transit System is built
Effect.
The content of the invention
In view of this, it is an object of the invention to provide a kind of track traffic based on track IC-card and mobile phone signaling data
Passenger flow forecasting, this method first does OD matchings to preprocessed data；Commuting identification is carried out to the OD data matched；To track
Traffic passenger flow is counted；Passenger flow statisticses result is analyzed；Visitor is carried out finally according to passenger flow statisticses result and interpretation of result
Stream prediction.This method provides reference for track traffic relevant Decision person to the rational deployment of track traffic, helps to reach passenger flow
Amount and the equilibrium of supply and demand of train.
To reach above-mentioned purpose, the present invention provides following technical scheme：
A kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data, this method includes following step
Suddenly：
S1：Acquisition trajectory traffic passenger's IC-card data message, initial data turns into preprocessed data after cutting is arranged,
OD matchings are carried out to preprocessed data；
S2：Commuting identification, inputs multiple workaday data, is grouped by user, judges each in multiple working days
The commuting rule of user：When the user is identical, enter the station website and outbound website number of days reaches 50% (configurable) of total number of days, i.e.,
Can determine whether the trip is that the user is commuter；
S3：Track traffic for passenger flow is counted, statistical indicator includes：Website passenger flow, circuit passenger flow, section passenger flow is changed
Passenger flows, passenger flow OD, commute website passenger flow, and commute circuit passenger flow, and commute website OD passenger flows, and commute circuit OD passenger flows；
S4：Passenger flow analysing：It is divided into the analysis of the normality volume of the flow of passengers to analyze with the festivals or holidays volume of the flow of passengers；
S5：Passenger flow estimation：It is divided into normality passenger flow forecast and festivals or holidays passenger flow forecast.
Further, in step sl, OD matchings are carried out to the preprocessed data, mainly based on time series, to every
Individual user temporally carries out priority sequence respectively, and each pair OD of user is determined with reference to turnover station identifications；It is right in the matching process
Abnormal data is handled, and is specifically included：
1) data of only enter the station record or only outbound record are rejected；
2) turnover station identifications with time series on the contrary, the temporal information that enters the station of the last time in outbound record is modified
Matching, is rejected if not correcting；
3) repeatedly enter the station once it is outbound or once enter the station it is repeatedly outbound, by it is out of the station matched by time of closest approach (
Need to match time immediate two data together)；
4) the double double outbound feelings that enter the station after the sequence of situation about passing in and out, i.e. unique user elapsed time
Condition；Last time in the outbound record temporal information that enters the station is modified matching, and the four of the situation are rejected if not correcting
Data；
5) matched for the data that the data before and after daily 0 point of morning will be combined one day after, if still matched not
On then reject；
6) same after matching stand into the OD stood out is rejected；The data that the match is successful must be exported as intermediate result.
Further, in step s3, specifically include：
1) website passenger flow statisticses：Count each website volume of the flow of passengers out of the station (by time statistics out of the station)：
It is by half an hour statistics export structure：Site name varchar；Time DATE (between at the beginning of half an hour)；Enter
Standee flows number；Outbound passenger flow number；
Daily statistics export structure is：Site name varchar；Time DATE (my god)；The passenger flow that enters the station number；Go out standee
Flow number；
2) circuit passenger flow：Each bar track circuit volume of the flow of passengers is counted, is counted according to user path, if certain of user
Secondary trip route have passed through Line 1, No. 2 lines, No. 3 lines, then Line 1 passenger flow+1, No. 2 line passenger flows+1, No. 3 line passenger flows+1；
It is by half an hour statistics export structure：Circuit number varchar；Time DATE (between at the beginning of half an hour)；Line
Road passenger flow number；
Daily statistics export structure is：Circuit number varchar；Time DATE (my god)；Circuit passenger flow number；
3) section passenger flow：Divide the directional statistics rail network section volume of the flow of passengers (flow between adjacent sites)；Computational methods, such as
Guanyinqiao-Hua Xin street section flows are calculated it is necessary to find out the people of all process Guanyinqiao-Hua Xin streets section in final path,
Then sum again.
It is by half an hour statistics export structure：Site name A varchar；Site name B varchar；Time DATE
(between at the beginning of half an hour)；Circuit passenger flow number；
Daily statistics export structure is：Site name A varchar；Site name B varchar；Time DATE (my god) line
Road passenger flow number；
4) transfer passenger flow：Statistics transfer website and circuit transfer passenger flow amount, notice that transfer website has 1 or multiple numberings
Before (being typically now 2 numberings), retrieval is gone when the numbering of appearance transfer website in path according to path, and transfer website is adjacent
The corresponding website of latter two site number on a circuit, that is, is not designated as transfer passenger flow；
It is by half an hour statistics export structure：Change to site name varchar；Change to site number varchar；Circuit A
varchar；Circuit B varchar；Time DATE (between at the beginning of half an hour)；Transfer passenger flow number；
Daily statistics export structure is：Change to site name varchar；Change to site number varchar；Circuit A
varchar；Circuit B varchar；Time DATE (my god)；Transfer passenger flow number；
5) passenger flow OD：
1st, each website OD passenger flows are counted
It is by half an hour statistics export structure：O varchar (site name)；D varchar (site name)；Time
DATE (between at the beginning of half an hour)；OD passenger flows number；
Daily statistics export structure is：O varchar (site name)；D varchar (site name)；Time DATE
(my god)；OD passenger flows number；
2nd, each circuit OD passenger flows are counted
It is by half an hour statistics export structure：Oline varchar；Dline varchar；Time DATE be (half an hour
Time started)；OD passenger flows number
Daily statistics export structure is：Oline varchar；Dline varchar；Time DATE (my god)；OD passenger flows
number；
6) commute passenger flow：
1st, commute website passenger flow
It is by half an hour statistics export structure：Site name varchar；Time DATE (between at the beginning of half an hour)；
The passenger flow that enters the station number；Outbound passenger flow number；
Daily statistics export structure is：Site name varchar；Time DATE (my god)；The passenger flow that enters the station number；Go out
Standee flows number；
2nd, commute circuit passenger flow
It is by half an hour statistics export structure：Circuit number varchar；Time DATE (between at the beginning of half an hour)；
Circuit passenger flow number；
Daily statistics export structure is：Circuit number varchar；Time DATE (my god)；Circuit passenger flow number；
3rd, commute website OD passenger flows
It is by half an hour statistics export structure：O varchar (site name)；D varchar (site name)；Time
DATE (between at the beginning of half an hour)；OD passenger flows number；
Daily statistics export structure is：O varchar (site name)；D varchar (site name)；Time DATE
(my god)；OD passenger flows number；
4th, commute circuit OD passenger flows
It is by half an hour statistics export structure：Oline varchar；Dline varchar；Time DATE be (half an hour
Time started)；OD passenger flows number.
Daily statistics export structure is：Oline varchar；Dline varchar；Time DATE (my god)；OD passenger flows
number。
Further, in step s 4, specifically include：
1) in order to preferably analyze the variation tendency of ordinary day metro passenger flow, entering the station day for 20 complete 140 days weeks is chosen
The volume of the flow of passengers analyzes changing rule, and the change fluctuation of normality passenger flow is regular, i.e., be fluctuating change above and below the cycle with week；In order to
The changing rule of normality passenger flow is more accurately analyzed, have chosen made a concrete analysis of with week for object on this basis；
2) festivals or holidays volume of the flow of passengers analysis includes：Spring Festival guest flow statistics is analyzed；International Labour Day guest flow statistics is analyzed；The Mid-autumn Festival
Analyzed with National Day guest flow statistics；Christmas Day and the analysis of New Year's Day guest flow statistics.
Further, in step s 5, specifically include：
1) normality passenger flow forecast：City Rail Transit System is the subjective initiative system that a someone participates in, according to right
The statistical analysis of normality passenger flow, the feature such as urban track traffic normality passenger flow has lack of uniformity, non-linear and time variation；For
Traditional this nonlinear system of linear model approximate fits, can simplify amount of calculation, but it is far from each other to predict the outcome.It is based on
The characteristics of this considers that BP neural network risk mapping ability is extremely strong, model prediction accuracy is higher, from BP neural network pair
Track traffic normality passenger flow is predicted；
2) festivals or holidays passenger flow forecast：The essence of nonparametric Regression Model is：Model is not by priori but by history
What data were determined, " nonparametric " does not represent printenv, but parametric form and number are variable.In the model regression function not by
The influence of the distribution relation of variable, and take full advantage of the possible state information of all preservations, when there is major event to occur or
Regularly, the accuracy of the forecast model is higher for unstable state.Therefore the festivals or holidays volume of the flow of passengers is carried out from nonparametric Regression Model
Prediction.
The beneficial effects of the present invention are：The present invention considers influence of the festivals or holidays to track traffic for passenger flow amount, establishes
Normality passenger flow forecast model and festivals or holidays passenger flow forecast model, are capable of the volume of the flow of passengers of more precisely predicted orbit traffic,
Rationally establishment conveyance equilibrium scheme, passenger organization scheme etc. are played an important role.
Brief description of the drawings
In order that the purpose of the present invention, technical scheme and beneficial effect are clearer, the present invention provides drawings described below and carried out
Explanation：
Fig. 1 is OD matching algorithm flow charts；
Fig. 2 is passenger flow statisticses content block diagram；
Fig. 3 is BP neural network algorithm flow block diagram
Fig. 4 is nonparametric Regression Model block diagram；
Fig. 5 is the schematic flow sheet of the method for the invention.
Embodiment
Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described in detail.
Fig. 5 is the schematic flow sheet of the method for the invention, as illustrated, the method that the present invention is provided includes following step
Suddenly：
S1：Acquisition trajectory traffic passenger's IC-card data message, initial data turns into preprocessed data after cutting is arranged,
OD matchings are carried out to preprocessed data；
S2：Commuting identification, inputs multiple workaday data, is grouped by user, judges each in multiple working days
The commuting rule of user：When the user is identical, enter the station website and outbound website number of days reaches 50% (configurable) of total number of days, i.e.,
Can determine whether the trip is that the user is commuter；
S3：Track traffic for passenger flow is counted, statistical indicator includes：Website passenger flow, circuit passenger flow, section passenger flow is changed
Passenger flows, passenger flow OD, commute website passenger flow, and commute circuit passenger flow, and commute website OD passenger flows, and commute circuit OD passenger flows；
S4：Passenger flow analysing：It is divided into the analysis of the normality volume of the flow of passengers to analyze with the festivals or holidays volume of the flow of passengers；
S5：Passenger flow estimation：It is divided into normality passenger flow forecast and festivals or holidays passenger flow forecast.
Specifically：
In step sl, OD matchings are carried out to the preprocessed data, mainly based on time series, to each user point
Do not carry out priority sequence temporally, each pair OD of user is determined with reference to turnover station identifications；In the matching process to abnormal data
Handled, as shown in figure 1, specifically including：
1) data of only enter the station record or only outbound record are rejected；
2) turnover station identifications with time series on the contrary, the temporal information that enters the station of the last time in outbound record is modified
Matching, is rejected if not correcting；
3) repeatedly enter the station once it is outbound or once enter the station it is repeatedly outbound, by it is out of the station matched by time of closest approach (
Need to match time immediate two data together)；
4) the double double outbound feelings that enter the station after the sequence of situation about passing in and out, i.e. unique user elapsed time
Condition；Last time in the outbound record temporal information that enters the station is modified matching, and the four of the situation are rejected if not correcting
Data；
5) matched for the data that the data before and after daily 0 point of morning will be combined one day after, if still matched not
On then reject；
6) same after matching stand into the OD stood out is rejected；The data that the match is successful must be exported as intermediate result.
Fig. 2 is passenger flow statisticses content block diagram, in step s3, is specifically included：
1) website passenger flow statisticses：Count each website volume of the flow of passengers out of the station (by time statistics out of the station)：
It is by half an hour statistics export structure：Site name varchar；Time DATE (between at the beginning of half an hour)；Enter
Standee flows number；Outbound passenger flow number；
Daily statistics export structure is：Site name varchar；Time DATE (my god)；The passenger flow that enters the station number；Go out standee
Flow number；
2) circuit passenger flow：Each bar track circuit volume of the flow of passengers is counted, is counted according to user path, if certain of user
Secondary trip route have passed through Line 1, No. 2 lines, No. 3 lines, then Line 1 passenger flow+1, No. 2 line passenger flows+1, No. 3 line passenger flows+1；
It is by half an hour statistics export structure：Circuit number varchar；Time DATE (between at the beginning of half an hour)；Line
Road passenger flow number；
Daily statistics export structure is：Circuit number varchar；Time DATE (my god)；Circuit passenger flow number；
3) section passenger flow：Divide the directional statistics rail network section volume of the flow of passengers (flow between adjacent sites)；Computational methods, such as
Guanyinqiao-Hua Xin street section flows are calculated it is necessary to find out the people of all process Guanyinqiao-Hua Xin streets section in final path,
Then sum again.
It is by half an hour statistics export structure：Site name A varchar；Site name B varchar；Time DATE
(between at the beginning of half an hour)；Circuit passenger flow number；
Daily statistics export structure is：Site name A varchar；Site name B varchar；Time DATE (my god) line
Road passenger flow number；
4) transfer passenger flow：Statistics transfer website and circuit transfer passenger flow amount, notice that transfer website has 1 or multiple numberings
Before (being typically now 2 numberings), retrieval is gone when the numbering of appearance transfer website in path according to path, and transfer website is adjacent
The corresponding website of latter two site number on a circuit, that is, is not designated as transfer passenger flow；
It is by half an hour statistics export structure：Change to site name varchar；Change to site number varchar；Circuit A
varchar；Circuit B varchar；Time DATE (between at the beginning of half an hour)；Transfer passenger flow number；
Daily statistics export structure is：Change to site name varchar；Change to site number varchar；Circuit A
varchar；Circuit B varchar；Time DATE (my god)；Transfer passenger flow number；
5) passenger flow OD：
1st, each website OD passenger flows are counted
It is by half an hour statistics export structure：O varchar (site name)；D varchar (site name)；Time
DATE (between at the beginning of half an hour)；OD passenger flows number；
Daily statistics export structure is：O varchar (site name)；D varchar (site name)；Time DATE
(my god)；OD passenger flows number；
2nd, each circuit OD passenger flows are counted
It is by half an hour statistics export structure：Oline varchar；Dline varchar；Time DATE be (half an hour
Time started)；OD passenger flows number
Daily statistics export structure is：Oline varchar；Dline varchar；Time DATE (my god)；OD passenger flows
number；
6) commute passenger flow：
1st, commute website passenger flow
It is by half an hour statistics export structure：Site name varchar；Time DATE (between at the beginning of half an hour)；
The passenger flow that enters the station number；Outbound passenger flow number；
Daily statistics export structure is：Site name varchar；Time DATE (my god)；The passenger flow that enters the station number；Go out
Standee flows number；
2nd, commute circuit passenger flow
It is by half an hour statistics export structure：Circuit number varchar；Time DATE (between at the beginning of half an hour)；
Circuit passenger flow number；
Daily statistics export structure is：Circuit number varchar；Time DATE (my god)；Circuit passenger flow number；
3rd, commute website OD passenger flows
It is by half an hour statistics export structure：O varchar (site name)；D varchar (site name)；Time
DATE (between at the beginning of half an hour)；OD passenger flows number；
Daily statistics export structure is：O varchar (site name)；D varchar (site name)；Time DATE
(my god)；OD passenger flows number；
4th, commute circuit OD passenger flows
It is by half an hour statistics export structure：Oline varchar；Dline varchar；Time DATE be (half an hour
Time started)；OD passenger flows number.
Daily statistics export structure is：Oline varchar；Dline varchar；Time DATE (my god)；OD passenger flows
number。
In step s 4, specifically include：
1) in order to preferably analyze the variation tendency of ordinary day metro passenger flow, entering the station day for 20 complete 140 days weeks is chosen
The volume of the flow of passengers analyzes changing rule, and the change fluctuation of normality passenger flow is regular, i.e., be fluctuating change above and below the cycle with week；In order to
The changing rule of normality passenger flow is more accurately analyzed, have chosen made a concrete analysis of with week for object on this basis；
2) festivals or holidays volume of the flow of passengers analysis includes：Spring Festival guest flow statistics is analyzed；International Labour Day guest flow statistics is analyzed；The Mid-autumn Festival
Analyzed with National Day guest flow statistics；Christmas Day and the analysis of New Year's Day guest flow statistics.
In step s 5, specifically include：
1) normality passenger flow forecast：City Rail Transit System is the subjective initiative system that a someone participates in, according to right
The statistical analysis of normality passenger flow, the feature such as urban track traffic normality passenger flow has lack of uniformity, non-linear and time variation；For
Traditional this nonlinear system of linear model approximate fits, can simplify amount of calculation, but it is far from each other to predict the outcome.It is based on
The characteristics of this considers that BP neural network risk mapping ability is extremely strong, model prediction accuracy is higher, from BP neural network pair
Track traffic normality passenger flow is predicted, as shown in Figure 3；
2) festivals or holidays passenger flow forecast：The essence of nonparametric Regression Model is：Model is not by priori but by history
What data were determined, " nonparametric " does not represent printenv, but parametric form and number are variable.In the model regression function not by
The influence of the distribution relation of variable, and take full advantage of the possible state information of all preservations, when there is major event to occur or
Regularly, the accuracy of the forecast model is higher for unstable state.Therefore the festivals or holidays volume of the flow of passengers is carried out from nonparametric Regression Model
Prediction, as shown in Figure 4.
Finally illustrate, preferred embodiment above is merely illustrative of the technical solution of the present invention and unrestricted, although logical
Cross above preferred embodiment the present invention is described in detail, it is to be understood by those skilled in the art that can be
Various changes are made to it in form and in details, without departing from claims of the present invention limited range.
Claims (5)
1. a kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data, it is characterised in that：This method
Comprise the following steps：
S1：Acquisition trajectory traffic passenger's IC-card data message, initial data turns into preprocessed data after cutting is arranged, to pre-
Processing data carries out OD matchings；
S2：Commuting identification, inputs multiple workaday data, is grouped by user, judges each user in multiple working days
Commuting rule：When the user is identical, enter the station website and outbound website number of days reaches the 50% of total number of days, you can judge the trip
It is commuter for the user；
S3：Track traffic for passenger flow is counted, statistical indicator includes：Website passenger flow, circuit passenger flow, section passenger flow, transfer visitor
Stream, passenger flow OD, commute website passenger flow, and commute circuit passenger flow, and commute website OD passenger flows, and commute circuit OD passenger flows；
S4：Passenger flow analysing：It is divided into the analysis of the normality volume of the flow of passengers to analyze with the festivals or holidays volume of the flow of passengers；
S5：Passenger flow estimation：It is divided into normality passenger flow forecast and festivals or holidays passenger flow forecast.
2. a kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data as claimed in claim 1,
It is characterized in that：In step sl, OD matchings are carried out to the preprocessed data, mainly based on time series, used each
Family temporally carries out priority sequence respectively, and each pair OD of user is determined with reference to turnover station identifications；In the matching process to exception
Data are handled, and are specifically included：
1) data of only enter the station record or only outbound record are rejected；
2) turnover station identifications and time series is on the contrary, the temporal information that enters the station of the last time in outbound record is modified
Match somebody with somebody, rejected if not correcting；
3) repeatedly enter the station once it is outbound or once enter the station repeatedly it is outbound, matched out of the station by time of closest approach；
4) the double double outbound situation that enters the station after the sequence of situation about passing in and out, i.e. unique user elapsed time；
Last time in the outbound record temporal information that enters the station is modified matching, and four numbers of the situation are rejected if not correcting
According to；
5) to combine data one day after for the data before and after daily 0 point of morning match, if still unmatching
Reject；
6) same after matching stand into the OD stood out is rejected；The data that the match is successful must be exported as intermediate result.
3. a kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data as claimed in claim 1,
It is characterized in that：In step s3, specifically include：
1) website passenger flow statisticses：Count each website volume of the flow of passengers out of the station：
It is by half an hour statistics export structure：Site name varchar；Time DATE (between at the beginning of half an hour)；Enter standee
Flow number；Outbound passenger flow number；
Daily statistics export structure is：Site name varchar；Time DATE (my god)；The passenger flow that enters the station number；Outbound passenger flow
number；
2) circuit passenger flow：Each bar track circuit volume of the flow of passengers is counted, is counted according to user path, if certain of a user goes out
Walking along the street footpath have passed through Line 1, No. 2 lines, No. 3 lines, then Line 1 passenger flow+1, No. 2 line passenger flows+1, No. 3 line passenger flows+1；
It is by half an hour statistics export structure：Circuit number varchar；Time DATE (between at the beginning of half an hour)；Circuit visitor
Flow number；
Daily statistics export structure is：Circuit number varchar；Time DATE (my god)；Circuit passenger flow number；
3) section passenger flow：Divide the directional statistics rail network section volume of the flow of passengers (flow between adjacent sites)；
It is by half an hour statistics export structure：Site name A varchar；Site name B varchar；(half is small by time DATE
When at the beginning of between)；Circuit passenger flow number；
Daily statistics export structure is：Site name A varchar；Site name B varchar；Time DATE (my god) circuit visitor
Flow number；
4) transfer passenger flow：Statistics transfer website and circuit transfer passenger flow amount, go retrieval to work as in path and transfer stop occur according to path
The numbering of point, and the corresponding website of former and later two adjacent site numbers of transfer website is designated as transfer visitor not on a circuit, that is,
Stream；
It is by half an hour statistics export structure：Change to site name varchar；Change to site number varchar；Circuit A
varchar；Circuit B varchar；Time DATE (between at the beginning of half an hour)；Transfer passenger flow number；
Daily statistics export structure is：Change to site name varchar；Change to site number varchar；Circuit A varchar；
Circuit B varchar；Time DATE (my god)；Transfer passenger flow number；
5) passenger flow OD：
1st, each website OD passenger flows are counted
It is by half an hour statistics export structure：O varchar (site name)；D varchar (site name)；Time DATE
(between at the beginning of half an hour)；OD passenger flows number；
Daily statistics export structure is：O varchar (site name)；D varchar (site name)；Time DATE (my god)；
OD passenger flows number；
2nd, each circuit OD passenger flows are counted
It is by half an hour statistics export structure：Oline varchar；Dline varchar；Time DATE (the beginnings of half an hour
Time)；OD passenger flows number
Daily statistics export structure is：Oline varchar；Dline varchar；Time DATE (my god)；OD passenger flows number；
6) commute passenger flow：
1st, commute website passenger flow
It is by half an hour statistics export structure：Site name varchar；Time DATE (between at the beginning of half an hour)；Enter standee
Flow number；Outbound passenger flow number；
Daily statistics export structure is：Site name varchar；Time DATE (my god)；The passenger flow that enters the station number；Outbound passenger flow
number；
2nd, commute circuit passenger flow
It is by half an hour statistics export structure：Circuit number varchar；Time DATE (between at the beginning of half an hour)；Circuit visitor
Flow number；
Daily statistics export structure is：Circuit number varchar；Time DATE (my god)；Circuit passenger flow number；
3rd, commute website OD passenger flows
It is by half an hour statistics export structure：O varchar (site name)；D varchar (site name)；Time DATE
(between at the beginning of half an hour)；OD passenger flows number；
Daily statistics export structure is：O varchar (site name)；D varchar (site name)；Time DATE (my god)；
OD passenger flows number；
4th, commute circuit OD passenger flows
It is by half an hour statistics export structure：Oline varchar；Dline varchar；Time DATE (the beginnings of half an hour
Time)；OD passenger flows number.
Daily statistics export structure is：Oline varchar；Dline varchar；Time DATE (my god)；OD passenger flows number.
4. a kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data as claimed in claim 1,
It is characterized in that：In step s 4, specifically include：
1) in order to preferably analyze the variation tendency of ordinary day metro passenger flow, the passenger flows that enter the station day in 20 complete 140 days weeks are chosen
Amount analysis changing rule, normality passenger flow change fluctuation is regular, i.e., be fluctuating change above and below the cycle with week；In order to more accurate
The changing rule of true analysis normality passenger flow, have chosen made a concrete analysis of with week for object on this basis；
2) festivals or holidays volume of the flow of passengers analysis includes：Spring Festival guest flow statistics is analyzed；International Labour Day guest flow statistics is analyzed；The Mid-autumn Festival and state
Celebrating section guest flow statistics analysis；Christmas Day and the analysis of New Year's Day guest flow statistics.
5. a kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data as claimed in claim 1,
It is characterized in that：In step s 5, specifically include：
1) normality passenger flow forecast：Track traffic normality passenger flow is predicted from BP neural network；
2) festivals or holidays passenger flow forecast：The festivals or holidays volume of the flow of passengers is predicted from nonparametric Regression Model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710191811.8A CN106960261A (en) | 2017-03-27 | 2017-03-27 | A kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710191811.8A CN106960261A (en) | 2017-03-27 | 2017-03-27 | A kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106960261A true CN106960261A (en) | 2017-07-18 |
Family
ID=59471067
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710191811.8A Pending CN106960261A (en) | 2017-03-27 | 2017-03-27 | A kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106960261A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647832A (en) * | 2018-05-18 | 2018-10-12 | 辽宁工业大学 | A kind of subway circulation interval time control algolithm based on neural network |
| CN109583656A (en) * | 2018-12-06 | 2019-04-05 | 重庆邮电大学 | Passenger Flow in Urban Rail Transit prediction technique based on A-LSTM |
| CN109842848A (en) * | 2017-09-22 | 2019-06-04 | 江苏智谋科技有限公司 | A kind of region flow of the people predicting platform based on mobile phone signaling |
| CN109905845A (en) * | 2018-12-10 | 2019-06-18 | 华南理工大学 | A kind of bus passenger flow OD acquisition methods based on mobile phone signaling |
| CN110246332A (en) * | 2019-06-05 | 2019-09-17 | 北京交通大学 | The real-time passenger flow method for monitoring of rail traffic and system based on multisource data fusion |
| CN110570004A (en) * | 2018-06-05 | 2019-12-13 | 上海申通地铁集团有限公司 | subway passenger flow prediction method and system |
| CN110913345A (en) * | 2019-11-15 | 2020-03-24 | 东南大学 | Section passenger flow calculation method based on mobile phone signaling data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015171286A1 (en) * | 2014-05-07 | 2015-11-12 | Exxonmobil Upstream Research Company | Method of generating an optimized ship schedule to deliver liquefied natural gas |
| CN105550789A (en) * | 2016-02-19 | 2016-05-04 | 上海果路交通科技有限公司 | Method for predicting bus taking passenger flow |
| 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 |
-
2017
- 2017-03-27 CN CN201710191811.8A patent/CN106960261A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015171286A1 (en) * | 2014-05-07 | 2015-11-12 | Exxonmobil Upstream Research Company | Method of generating an optimized ship schedule to deliver liquefied natural gas |
| CN105550789A (en) * | 2016-02-19 | 2016-05-04 | 上海果路交通科技有限公司 | Method for predicting bus taking passenger flow |
| 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 |
Non-Patent Citations (1)
| Title |
|---|
| 何九冉: "城市轨道交通客流统计特征分析及组合预测方法实证研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109842848A (en) * | 2017-09-22 | 2019-06-04 | 江苏智谋科技有限公司 | A kind of region flow of the people predicting platform based on mobile phone signaling |
| CN108647832A (en) * | 2018-05-18 | 2018-10-12 | 辽宁工业大学 | A kind of subway circulation interval time control algolithm based on neural network |
| CN108647832B (en) * | 2018-05-18 | 2020-08-04 | 辽宁工业大学 | Subway operation interval time control algorithm based on neural network |
| CN110570004A (en) * | 2018-06-05 | 2019-12-13 | 上海申通地铁集团有限公司 | subway passenger flow prediction method and system |
| CN109583656A (en) * | 2018-12-06 | 2019-04-05 | 重庆邮电大学 | Passenger Flow in Urban Rail Transit prediction technique based on A-LSTM |
| CN109583656B (en) * | 2018-12-06 | 2022-05-10 | 重庆邮电大学 | Urban rail transit passenger flow prediction method based on A-LSTM |
| CN109905845A (en) * | 2018-12-10 | 2019-06-18 | 华南理工大学 | A kind of bus passenger flow OD acquisition methods based on mobile phone signaling |
| CN110246332A (en) * | 2019-06-05 | 2019-09-17 | 北京交通大学 | The real-time passenger flow method for monitoring of rail traffic and system based on multisource data fusion |
| CN110913345A (en) * | 2019-11-15 | 2020-03-24 | 东南大学 | Section passenger flow calculation method based on mobile phone signaling data |
| CN110913345B (en) * | 2019-11-15 | 2021-01-05 | 东南大学 | Section passenger flow calculation method based on mobile phone signaling data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106960261A (en) | A kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data | |
| CN106971534B (en) | Commuter characteristic analysis method based on number plate data | |
| CN104318324B (en) | Shuttle Bus website and route planning method based on taxi GPS records | |
| CN107766969B (en) | Large station fast line layout method based on subway service capacity bottleneck section identification | |
| CN105405293B (en) | A kind of road travel time short term prediction method and system | |
| CN106779408A (en) | The appraisal procedure and device of public transit system service quality | |
| CN107134142A (en) | A kind of urban road method for predicting based on multisource data fusion | |
| KR101385057B1 (en) | Prediction of urban congestion using ITS based data | |
| CN106448132A (en) | Conventional public traffic service index real-time evaluation system and method | |
| Ma et al. | Modeling bus travel time reliability with supply and demand data from automatic vehicle location and smart card systems | |
| CN105279572A (en) | City track traffic passenger flow density index calculating and releasing system | |
| Wilson et al. | The potential impact of automated data collection systems on urban public transport planning. | |
| CN106327871A (en) | Highway congestion forecasting method based on historical data and reservation data | |
| Gong et al. | Hybrid dynamic prediction model of bus arrival time based on weighted of historical and real-time GPS data | |
| CN101236642A (en) | Classified passenger flow monitoring system and its method based on ticketing data | |
| CN106951549A (en) | A kind of passenger's traffic path recognition methods based on track IC-card and mobile phone signaling data | |
| Chengula et al. | Assessment of the effectiveness of Dar Es Salaam bus rapid transit (DBRT) system in Tanzania | |
| Zhang et al. | A data-driven analysis for operational vehicle performance of public transport network | |
| CN107274000A (en) | Urban track traffic section passenger flow forecasting under a kind of accident | |
| Zhang et al. | Analysis of spatial-temporal characteristics of operations in public transport networks based on multisource data | |
| CN102117510B (en) | Unknown transportation mean passenger source predictive method in big activities | |
| Yang et al. | Determination of optimal toll levels and toll locations of alternative congestion pricing schemes | |
| Song et al. | Public transportation service evaluations utilizing seoul transportation card data | |
| Nasiboglu et al. | Origin-destination matrix generation using smart card data: Case study for Izmir | |
| Ova et al. | Evaluation of transit signal priority strategies for small-medium cities |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170718 |
|
| RJ01 | Rejection of invention patent application after publication |