CN105550789A - Method for predicting bus taking passenger flow - Google Patents
Method for predicting bus taking passenger flow Download PDFInfo
- Publication number
- CN105550789A CN105550789A CN201610094427.1A CN201610094427A CN105550789A CN 105550789 A CN105550789 A CN 105550789A CN 201610094427 A CN201610094427 A CN 201610094427A CN 105550789 A CN105550789 A CN 105550789A
- Authority
- CN
- China
- Prior art keywords
- bus
- card
- passenger flow
- trip
- time
- 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 52
- 238000012546 transfer Methods 0.000 claims description 37
- 230000001680 brushing effect Effects 0.000 claims description 17
- 230000008901 benefit Effects 0.000 claims description 10
- 230000003203 everyday effect Effects 0.000 claims description 8
- 238000007726 management method Methods 0.000 description 9
- 238000011160 research Methods 0.000 description 5
- 238000013341 scale-up Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000002354 daily effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012843 least square support vector machine Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012360 testing method 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 invention discloses a method for predicting the bus taking passenger flow. The method comprises the following steps: 1, collecting card swiping data of IC cards of passengers, and recording positional information of a bus in real time; 2, obtaining names of stops which the bus passes through and time when the bus arrives at and leaves the stops; 3, obtaining the names of the stops where the passengers take the bus, the time when the passengers take the bus and bus information; 4, judging the stops where the passengers get off the bus; 5, calculating OD data of bus passenger flow and passenger flow in traffic zones according to OD data of daily outgoing of the passengers, and multiplying the OD data of the bus passenger flow with the outgoing coefficient, so that OD data of passenger flow of all bus routes are obtained, then obtaining the outgoing characteristics of all the passengers, and predicting the bus taking passenger flow in the future time according to the outgoing characteristics. By means of the method, the technical problem that the bus taking passenger flow is not easy to predict in real time is solved, and great guiding significance is achieved for making bus dispatching plans and optimizing urban public transportation systems and assisting in bus operation policy making.
Description
Technical field
The present invention relates to technical field of intelligent traffic.More particularly, the present invention relates to a kind of Forecasting Methodology of bus trip passenger flow.
Background technology
At present, the passenger OD data gathering public transit system are still one of difficult problem of restriction bus dispatching decision-making and planning.Customer information data are huge, and real-time update is quick, and passenger flow changing features is rapid, and these features make artificial collection have certain limitation.Therefore, the method for intelligentized Dynamic Acquisition OD data is introduced into this region gradually, plays a role.The Real-time Obtaining of dynamic bus passenger flow data, process and analysis are the important and difficult issues of research at present, and it needs to change in time and constantly upgrades, processes.Along with the development of infotech, IC Card Fee is brought into use in many big cities, it has not only increased substantially the labour productivity of Public Transport Trade, and provide easily by bus environment for citizen, and provide a kind of new method and means for the acquisition of dynamic bus passenger flow data and short-term prediction.
Existing public transport OD evaluation method mainly contains: go-outside for civilian by bus is investigated, and the terminal, object, time, distance, number of times etc. of namely bus being gone on a journey to passenger with car inquire investigation; Resident's bus IC card matching process, utilizes the Fare Collection System platform of bus IC card and existing computer database technology, by database classification process, is decomposed on public bus network by the Based on Bus IC Card Data within the scope of the whole city.According to the dispatching log on the same day and the membership function of the site location of public bus network and running time, obtain the information such as the Entrucking Point of each IC-card record and the travel direction of vehicle.The each IC-card of system database record go out line item in the recent period, according to the get-off stop going out line item in the recent period and infer every bar IC-card record, and the time getting off.
But the method for current public transport OD passenger flow estimation divides resident's full mode, namely after full mode trip prediction, obtain bus trip OD by model split, the result finally obtained is passenger flow distribution between traffic zone and inner.This shortcut estimation and forecast method is more complicated, and workload is large, low precision and do not have ageing, is difficult to the Real-Time Scheduling demand meeting bus operation.
Summary of the invention
An object of the present invention is to solve at least the problems referred to above and/or defect, and the advantage will illustrated at least is below provided.
A further object of the invention is to provide a kind of Forecasting Methodology of bus trip passenger flow, and the method that it utilizes IC-card and GPS to combine gathers Customer information.IC card system structure, determines the mode obtaining OD data from the payment information of IC-card.During passenger loading, hold IC-card and to swipe the card in vehicle-mounted POS, complete the payment of fare, in POS, leave transaction record; Be kept in the database of mass transit card company after the record derivation of swiping the card of POS.On the other hand, the GPS device that public transit vehicle carries, can record the positional information of vehicle in real time, and these data, after the process of public transport company, can reflect that public transit vehicle passes in and out the precise moments of a certain bus stop intuitively.If take out a part useful to the reckoning of bus passenger flow OD in IC-card record and public transport GPS traveling record, according to the corresponding relation of the corresponding relation of POS and public transit vehicle, charge time and time out of the station, just can infer the holder an IC-card be in when, any station point taken which car of which bar public bus network.If the get-off stop of holder can be extrapolated further, so just can according to the website of getting on or off the bus often opening IC-card, count the information such as number of getting on or off the bus of the passenger flow OD of bus routes, each website, information is used as following bus trip passenger flow of database prediction thus, and solves the technical matters of bus trip passenger flow not easily real-time estimate.
In order to realize, according to these objects of the present invention and other advantage, providing a kind of Forecasting Methodology of bus trip passenger flow, comprise the following steps:
Step 1): the brushing card data gathering passenger's IC-card, the positional information of real time record bus;
Step 2): the real-time position information of bus is matched in map of website database, obtains the site name of bus process and pass in and out the moment of this website;
Step 3): in the moment passing in and out each website according to charge time of IC-card and bus, obtain the site name of getting on the bus of passenger, moment and information of vehicles;
Step 4): the whole records of swiping the card transferring in described brushing card data same IC-card same day, and according to time sequence, according to swiping the card, moment and website of swiping the card judge the trip relation between recording of swiping the card for i-th time and the i-th+1 time, one by one until complete the contrast of this IC-card whole brushing card data on the same day; If the record of swiping the card of adjacent twice meets round-trip travel relation, then the website of swiping the card for the i-th+1 time is i-th get-off stop of riding; If the record of swiping the card of adjacent twice meets transfer trip relation, then the website of swiping the card for the i-th+1 time is i-th get-off stop of riding; The like judge i-th+2 and i-th+1 adjacently goes out to swipe the card whether meet the relation of round-trip travel or transfer trip between record for twice, until obtain swipe the card at every turn inception point and terminus, namely passenger's OD data of at every turn going on a journey;
Step 5): the OD data of going on a journey every day according to passenger calculate the OD data of passenger flow in bus passenger flow, traffic zone, and, the OD data of bus passenger flow are multiplied by out the OD data that row coefficient obtains all public bus network passenger flows, and then obtain the trip characteristics of all passengers, according to the bus trip passenger flow of this trip characteristics prediction future time.
Preferably, wherein, described step 1) in, adopt and gather brushing card data with car terminal reader, adopt the real-time position information with car gps system record bus, wherein, described terminal reader and gps system identity information interrelated.
Preferably, wherein, described terminal reader is POS machine, and it gathers the charge time of IC-card, the swipe the card amount of money and whether transfer benefit.
Preferably, wherein, the real-time position information of brushing card data and bus described in public traffic management platform wireless receiving, and the traffic route of the real-time position information of bus and this bus described and website are matched, obtain bus through the title of website and moment out of the station, described map of website database is connected with described public traffic management platform.
Preferably, wherein, the site information on the traffic route of bus and route is included in described map of website database.
Preferably, wherein, described step 4) in, if the record of swiping the card of adjacent twice does not meet transfer trip or round-trip travel relation simultaneously, judge the terminus of riding for i-th time according to the history travelling OD data of this IC-card.
Preferably, wherein, described step 4) in, it is A station that the record of swiping the card of adjacent twice meets the inception point of riding for i-th time, the inception point of riding for the i-th+1 time is B station, the inception point of riding for the i-th+2 times is A station, and A station and B stand on a public bus network, then the record of swiping the card of adjacent twice meets round-trip travel relation.
Preferably, wherein, described step 4) in, it is C station that the record of swiping the card of adjacent twice meets the inception point of riding for i-th time, the inception point of riding for the i-th+1 time is D station, and the inception point of riding for the i-th+2 times is E station, and C station and D stand on a public bus network, and the charge time interval of adjacent twice is not more than 90 minutes or adjacent middle second time of swiping the card for twice swipes the card and meet transfer benefit condition, then the record of swiping the card of adjacent twice meets transfer trip relation.
Preferably, wherein, described step 5) in, to the bus trip passenger flow of every day than ever, the bus trip passenger flow contrasting each time period on the same day simultaneously, what obtain the bus passenger flow of each time period goes out row coefficient.
Preferably, wherein, described step 5) in, use the bus trip passenger flow of least square method supporting vector machine prediction future time.
The present invention at least comprises following beneficial effect:
(1) IC card system, GPS and generalized information system combine by the present invention, the passenger flow data of IC-card collection and locator data are uploaded to management platform in real time, and merge mutually with GIS data, improve the accuracy of bus passenger flow data acquisition, real-time, science and comprehensive;
(2) the present invention adopts the OD based on IC-card and gps data to calculate algorithm, the travelling OD distribution of single card can be obtained, and on this basis can to the travelling OD data obtaining public bus network aspect and traffic zone aspect, thus to the plan of formulation bus dispatching, Optimizing City public transit system has great importance;
(3) the present invention also according to the passenger's trip characteristics obtained, can take into account regularity and the time variation feature of traffic behavior, the bus trip passenger flow of real-time accurately predicting future time, so that tackle the changes in demand of urban mass-transit system in time.
Part is embodied by explanation below by other advantage of the present invention, target and feature, part also will by research and practice of the present invention by those skilled in the art is understood.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the Forecasting Methodology of bus trip passenger flow in one embodiment of the present of invention;
Fig. 2 is the schematic diagram utilizing IC card system information in another example of the present invention and determine passenger loading website with the association between car GPS driving system information;
Fig. 3 is the Forecasting Methodology process flow diagram of bus trip passenger flow in another example of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described in further detail, can implement according to this with reference to instructions word to make those skilled in the art.
Fig. 1 ~ Fig. 3 shows according to a kind of way of realization of the present invention, and it comprises the following steps:
Step 1): the brushing card data gathering passenger's IC-card, the positional information of real time record bus simultaneously, in this example, with reference to Fig. 1, adopt and gather brushing card data with car terminal reader, described terminal reader is POS machine, adopt the real-time position information with car gps system record bus, wherein, described terminal reader and gps system identity information interrelated, namely the POS of this car and the gps system identity information of this car corresponding, POS is connected with public traffic management base station radio with gps system, to transmit bus passenger flow information in real time, during passenger loading, hold IC-card to swipe the card in vehicle-mounted POS, complete the payment of fare, transaction record is left in POS, this transaction record specifically comprises the charge time of IC-card, to swipe the card the amount of money and whether transfer benefit,
Step 2): the real-time position information of bus is matched in map of website database, obtain the site name of bus process and pass in and out the moment of this website, the real-time position information of the bus of gps system record comprises bus change in location in time, the site information on the traffic route of bus and route is included in described map of website database, the real-time position information of brushing card data and bus described in public traffic management platform wireless receiving, and the traffic route of the real-time position information of bus and this bus described and website are matched, just can obtain bus through the title of website and moment out of the station, wherein, described map of website database is connected with described public traffic management platform,
Step 3): the moment concrete out of the station that bus passes in and out a certain platform can specifically learn, passes in and out the moment of each website from public traffic management platform according to the charge time of IC-card and bus, can obtain the site name of getting on the bus of passenger, moment and information of vehicles;
Step 4): the whole records of swiping the card transferring in described brushing card data same IC-card same day, and according to time sequence, according to swiping the card, moment and website of swiping the card judge the trip relation between recording of swiping the card for i-th time and the i-th+1 time, one by one until complete the contrast of this IC-card whole brushing card data on the same day; If the record of swiping the card of adjacent twice meets round-trip travel relation, then the website of swiping the card for the i-th+1 time is i-th get-off stop of riding; If the record of swiping the card of adjacent twice meets transfer trip relation, then the website of swiping the card for the i-th+1 time is i-th get-off stop of riding; The like judge i-th+2 and i-th+1 adjacently goes out to swipe the card whether meet the relation of round-trip travel or transfer trip between record for twice, until obtain swipe the card at every turn inception point and terminus, namely passenger's OD data of at every turn going on a journey; If the record of swiping the card of adjacent twice does not meet transfer trip or round-trip travel relation simultaneously, the terminus of riding for i-th time is judged according to the history travelling OD data of this IC-card, if still cannot judge according to history travelling OD data, then this records by bus and does not have data cogency, not as evaluation prediction data;
Step 5): calculate bus passenger flow according to the OD data that passenger goes on a journey every day, the OD data of passenger flow in traffic zone, and, the OD data of bus passenger flow are multiplied by out the OD data that row coefficient obtains all public bus network passenger flows, and then obtain the trip characteristics of all passengers, according to the bus trip passenger flow of this trip characteristics prediction future time, public traffic management platform transfers the brushing card data of passenger's IC-card and the positional information of bus to analyze current bus trip passenger flow from base station, transfer simultaneously and analyze the brushing card data of passenger's IC-card and the positional information of bus of this period in the past, to make prediction promptly and accurately to the bus trip passenger flow of future time period, the bus trip passenger flow forecasting of this example is simply efficient, and the bus trip passenger flow of prediction future time period is more real-time accurately.
In another embodiment, described step 4) in, it is A station that the record of swiping the card of adjacent twice meets the inception point of riding for i-th time, the inception point of riding for the i-th+1 time is B station, the inception point of riding for the i-th+2 times is A station, and A station and B stand on a public bus network, then the record of swiping the card of adjacent twice meets round-trip travel relation;
It is C station that the record of swiping the card of adjacent twice meets the inception point of riding for i-th time, the inception point of riding for the i-th+1 time is D station, the inception point of riding for the i-th+2 times is E station, C station and D stand on a public bus network, and the charge time interval of adjacent twice is not more than 90 minutes or adjacent middle second time of swiping the card for twice swipes the card and meet transfer benefit condition, then the record of swiping the card of adjacent twice meets transfer trip relation.
In another example, described step 5) in, to the bus trip passenger flow of every day than ever, contrast the bus trip passenger flow of each time period on the same day simultaneously, what obtain the bus passenger flow of each time period goes out row coefficient, the scale-up factor obtaining bus trip every day passenger flow is analyzed from the data of history bus trip passenger flow, obtain the scale-up factor of each period bus passenger flow in some day simultaneously, described go out row coefficient and this two scale-up factors be directly proportional, according to these two scale-up factors, that can learn concrete some day in a certain moment goes out row coefficient, row coefficient is gone out according to upper period bus trip passenger flow and this, i.e. measurable subsequent period bus trip passenger flow.
In another embodiment, the bus trip passenger flow of least square support vector machines prediction future time is used.
Concrete, during passenger loading, hold IC-card and to swipe the card in vehicle-mounted POS, complete the payment of fare, in POS, leave transaction record; The record of swiping the card of POS is led in database.On the other hand, the GPS device that public transit vehicle carries, can record the positional information of vehicle in real time, and these data, after the process of public traffic management platform, can reflect that public transit vehicle passes in and out the precise moments of a certain bus stop intuitively.Take out a part useful to the reckoning of bus passenger flow OD in IC-card record and public transport GPS traveling record, according to the corresponding relation of the corresponding relation of POS and public transit vehicle, charge time and time out of the station, just can infer the holder an IC-card be in when, any station point taken which car of which bar public bus network.If the get-off stop of holder can be extrapolated further, so just according to the website of getting on or off the bus often opening IC-card, the information such as number of getting on or off the bus of the passenger flow OD of bus routes, each website can be counted.
OD based on IC-card and gps data calculates algorithm, is divided into website reckoning of getting on the bus, get-off stop calculates, bus trip passenger flow OD calculates and bus trip passenger flow OD expands sample four major parts.The reckoning of website of getting on or off the bus is carried out in units of circuit and IC-card, namely each the record of swiping the card of IC-card on a public bus network to be calculated, whole records temporally after series arrangement of swiping the card that IC-card same day is all extracted in each reckoning, the itemize ground record adjacent with its time is compared, is determined whether round-trip travel or change to and go on a journey, thus calculates the get-off stop of swiping the card each time.After website of getting on or off the bus has calculated, namely obtain the travelling OD data of at every turn swiping the card; Successively carry out collection meter to more macroscopic aspect on this basis, the travelling OD data of public bus network aspect, traffic zone aspect can be obtained.By calculating that the bus passenger flow OD data of gained are pressed coefficient and expanded sample method, extending to bulk sample from the sample that can calculate, just can reflect the trip characteristics of calculated circuit all passengers.
With reference to Fig. 1, the geographic information data obtaining IC-card data from mass transit card manager, obtain public transport GPS traveling record from bus operation side, obtain bus stop from geographical information technology (GIS) provider, above indirect data, jointly will become the basis of bus passenger flow OD reckoning, distribute with the travelling OD obtaining single card, public bus network OD distributes and the OD of traffic zone distributes.
With reference to Fig. 2, the reckoning of website of getting on the bus is the matching relationship based on charge time and time out of the station.The scene of swiping the card according to getting on the bus can be known, the moment that passenger swipes the card, and should drop between vehicle pull-in and the moment of departures.Travel record from the GPS of this vehicle and search record out of the station, comparison charge time and moment out of the station, determining swipes the card at every turn is at for which website, i.e. the website of getting on the bus of passenger.The reckoning rate of this projectional technique and order of accuarcy, depend on that whether the corresponding relation of order of accuarcy that GPS travels record and judge site location, the order of accuarcy of the record of time out of the station, time error, POS and the bus licence plate between gps data and POS data is correct etc., namely calculate that result is very large by the impact of the quality of data.
Taken out by whole records of swiping the card on same IC-card same day, according to time sequence, from same day record the earliest, comparison i-th record and the relation between record for the i-th+1 time, judge whether it exists round or transfer relation one by one; Come and go if meet or the corresponding data characteristics of transfer trip, then calculate get-off stop by corresponding rule; Do not meet or calculates and successfully then continue to compare the i-th+1 time record and the i-th+2 times records, so circulate, all to swipe the card the comparison of recording until complete this IC-card.
The daily commuter of passenger is the pattern of round-trip travel mostly, i.e. this trip mode of having a try, and passenger, from starting point, finally gets back to starting point again, by the hypothesis of " two site models ", the feature of this trip can be described.The trip of first time on the same day, passenger gets on the bus from A station, leaves record of swiping the card, and get-off stop is unknown; Second time trip, passenger gets on the bus from B station, leaves record of swiping the card.Then can infer according to the hypothesis of " two site models ": the get-off stop of passenger's first time trip is B station, the get-off stop of second time trip is A station; After twice trip is implemented, passenger completes whole day and travels frequently, and gets back to starting point.Such commuting tools, often opening IC-card should have on the same day record of swiping the card taking same circuit bus at least 2 times, and the website difference of getting on the bus of riding for twice, the up-downlink direction of riding for twice are contrary.
In daily trip, especially apart from longer trip, the transfer of passenger between different circuit, different mode of transportation is frequent generation.Due to the restriction of data sample and research difficulty, this research only Public Transport Transfer of consideration on same platform, the transfer of adjacent bus station of not considering to need to proceed on foot.The data sample that this research institute gets is a circuit public transport and hands over the data of website with its part circuit that can change to and rail, therefore only considers the trip of once changing in data area.In transfer trip, IC-card holder is at C station ride circuit X, and leave record of swiping the card, get-off stop is unknown; Next record of swiping the card of this IC-card is got on the bus at D station, taken circuit Y (or D station for rail friendship website, passenger is entered the station at this station), and according to transfer website correspondence table, circuit X and circuit Y really can in the transfer of D station, so just can think holder's ride circuit X time, get-off stop is D.Logic like this, can infer and in twice trip of transfer relation, front get-off stop of once riding.After once go on a journey swipe the card and deduct fees, probably meet the condition of transfer benefit, namely the value of " TJRLTXFG " field is 1, or the time interval of swiping the card for twice is not more than 90 minutes, and these two features can as strengthed condition or the test basis judging transfer trip.
More general trip mode possible existingly comes and goes, and also has transfer, or the trip mode of both combinations, as the round-trip travel of once changing to.By transfer manner, public transport-public transport may be there is, public transport-rail is handed over, rail friendship-public transport three kinds of situations; The once trip after the transfer of get-off stop cannot be extrapolated by the logic of transfer trip, get-off stop can be extrapolated according to the feature of round-trip travel.Therefore, in reckoning process, need round and transfer two kinds of patterns to combine to carry out data processing.
In city bus is built, the movement capacity of public transit system, vehicle type selection, vehicle fleet size, station stop board scale and construction investment etc. all need to determine according to bus passenger flow size, therefore, passenger flow estimation directly will determine form and the cost of public bus network to a great extent.The prediction of city bus passenger flow mainly comprises the circuit volume of the flow of passengers in Public transport network planning stage, average riding distance and website passenger collector-distributor volume, by using statistical Theories and methods, carrying out statistical study to relevant information, drawing rational prediction.Four steps can be divided into passenger flow estimation.The first step analyze each region, city population distribution and population character; Second step analyzes the trip requirements of population; 3rd step analysis has the demand of crowd for bus trip of trip requirements; 4th step utilizes Principle of Statistics to set up public transport network model, carries out passenger flow index prediction.First the passenger flow estimation of Public transport network planning will be predicted the total volume of the flow of passengers of circuit, by carrying out statistical study to each side information in city, to the present situation in city and public transport planning region starting point-terminal passenger flow analysing, add up and confirm the far-seeing plan of public transport.For the prediction of average riding distance, to public bus network, survey carried out to the resident of zones of different and gathers, carrying out statistical study on this basis, obtain the line length, vehicle type selection, station scale etc. that need to build.For the prediction of website passenger collector-distributor volume, refer to the two-way statistic of to get on or off the bus person-time of public bus network, the factor such as to be connected with the regional conditions of website, communication function, public transit system directly related.
The traffic passenger flow forecasting of short-term is roughly divided into linear model approach, nonlinear model method and mixed method three types.Linear prediction model method is comparatively simple, can predicting traffic flow overall trend, but only when predicted time is longer, effect is better, and when the time is shorter, error is very large, is difficult to the time variation correctly reflecting urban transportation.Nonlinear model not only considers multi-dimensional nature and the temporal correlation of traffic flow, also contemplates the empiric risk in forecasting process and confidence risk, has higher precision of prediction and adaptive ability, be widely used.But when traffic flow is in plateau, its precision of prediction step-down on the contrary, occurs that the possibility of the phenomenon of whole departure actual value becomes large.Visible, linear processes method all can not accurately predicting and embody the feature of short-term traffic passenger flow completely, propose mixed method thus, utilize the short-term bus passenger flow Forecasting Methodology based on multinuclear least square method supporting vector machine, take into account regularity and the time variation feature of traffic behavior.
From long-term angle, bus passenger flow presents seven days periodic laws as a circulation change, and the passenger flow of that day corresponding to each cycle before every day in the cycle has higher correlativity.From short-term angle, the traffic passenger flow of the size of public transport short-term passenger flow and this day operating time interior front certain hour section is based on stronger non-linear dependencies.Thus, model consider some day public transport flow sometime and the historical data of this day, before this week the data of n days and this few days ago s observe the period, that is:
Wherein: F represents traffic flow; D represents the date; T represents the observation period; M represents the observation data in front m week; N represents the front n days in this week; S represents front s the observation period of this day.
Introduce least square method supporting vector machine, its basic thought utilizes kernel function that data set is mapped to high-dimensional feature space, makes the nonlinear fitting problem in the input space transfer linear fit problem in high-dimensional feature space to.According to the Fitting Calculation of formula above, the bus trip passenger flow of future time period can be drawn.
Here the module number illustrated and treatment scale are used to simplify explanation of the present invention.The application of the Forecasting Methodology of bus trip passenger flow of the present invention, modifications and variations be will be readily apparent to persons skilled in the art.
As mentioned above, the present invention simply efficiently can obtain the OD data of bus passenger flow according to the bus passenger flow data of real-time accurate acquisition, and then obtain the trip characteristics of all passengers, and according to the bus trip passenger flow of the more real-time accurately predicting future time period of this trip characteristics, to the plan of formulation bus dispatching, Optimizing City public transit system and auxiliary bus operation decision-making have great importance.
Although embodiment of the present invention are open as above, it is not restricted to listed in instructions and embodiment utilization.It can be applied to various applicable the field of the invention completely.For those skilled in the art, can easily realize other amendment.Therefore do not deviating under the universal that claim and equivalency range limit, the present invention is not limited to specific details and illustrates here and the legend described.
Claims (10)
1. a Forecasting Methodology for bus trip passenger flow, is characterized in that, comprises the steps:
Step 1): the brushing card data gathering passenger's IC-card, the positional information of real time record bus;
Step 2): the real-time position information of bus is matched in map of website database, obtains the site name of bus process and pass in and out the moment of this website;
Step 3): in the moment passing in and out each website according to charge time of IC-card and bus, obtain the site name of getting on the bus of passenger, moment and information of vehicles;
Step 4): the whole records of swiping the card transferring in described brushing card data same IC-card same day, and according to time sequence, according to swiping the card, moment and website of swiping the card judge the trip relation between recording of swiping the card for i-th time and the i-th+1 time, one by one until complete the contrast of this IC-card whole brushing card data on the same day; If the record of swiping the card of adjacent twice meets round-trip travel relation, then the website of swiping the card for the i-th+1 time is i-th get-off stop of riding; If the record of swiping the card of adjacent twice meets transfer trip relation, then the website of swiping the card for the i-th+1 time is i-th get-off stop of riding; The like judge i-th+2 and i-th+1 adjacently goes out to swipe the card whether meet the relation of round-trip travel or transfer trip between record for twice, until obtain swipe the card at every turn inception point and terminus, namely passenger's OD data of at every turn going on a journey;
Step 5): the OD data of going on a journey every day according to passenger calculate the OD data of passenger flow in bus passenger flow, traffic zone, and the OD data of bus passenger flow are multiplied by out the OD data that row coefficient obtains all public bus network passenger flows, and then obtain the trip characteristics of all passengers, according to the bus trip passenger flow of this trip characteristics prediction future time.
2. the Forecasting Methodology of bus trip passenger flow as claimed in claim 1, it is characterized in that, described step 1) in, adopt and gather brushing card data with car terminal reader, adopt the real-time position information with car gps system record bus, wherein, described terminal reader and gps system identity information interrelated.
3. the Forecasting Methodology of bus trip passenger flow as claimed in claim 2, it is characterized in that, described terminal reader is POS machine, and it gathers the charge time of IC-card, the swipe the card amount of money and whether transfer benefit.
4. the Forecasting Methodology of bus trip passenger flow as claimed in claim 1, it is characterized in that, the real-time position information of brushing card data and bus described in public traffic management platform wireless receiving, and the traffic route of the real-time position information of bus and this bus described and website are matched, obtain bus through the title of website and moment out of the station, described map of website database is connected with described public traffic management platform.
5. the Forecasting Methodology of bus trip passenger flow as claimed in claim 1, is characterized in that, includes the site information on the traffic route of bus and route in described map of website database.
6. the Forecasting Methodology of bus trip passenger flow as claimed in claim 1, it is characterized in that, described step 4) in, if the record of swiping the card of adjacent twice does not meet transfer trip or round-trip travel relation simultaneously, judge the terminus of riding for i-th time according to the history travelling OD data of this IC-card.
7. the Forecasting Methodology of bus trip passenger flow as claimed in claim 1, it is characterized in that, described step 4) in, it is A station that the record of swiping the card of adjacent twice meets the inception point of riding for i-th time, the inception point of riding for the i-th+1 time is B station, the inception point of riding for the i-th+2 times is A station, and A station and B stand on a public bus network, then the record of swiping the card of adjacent twice meets round-trip travel relation.
8. the Forecasting Methodology of bus trip passenger flow as claimed in claim 1, it is characterized in that, described step 4) in, it is C station that the record of swiping the card of adjacent twice meets the inception point of riding for i-th time, the inception point of riding for the i-th+1 time is D station, the inception point of riding for the i-th+2 times is E station, C station and D stand on a public bus network, and the charge time interval of adjacent twice is not more than 90 minutes or adjacent middle second time of swiping the card for twice swipes the card and meet transfer benefit condition, then the record of swiping the card of adjacent twice meets transfer trip relation.
9. the Forecasting Methodology of bus trip passenger flow as claimed in claim 1, it is characterized in that, described step 5) in, to the bus trip passenger flow of every day than ever, contrast the bus trip passenger flow of each time period on the same day, what obtain the bus passenger flow of each time period goes out row coefficient simultaneously.
10. the Forecasting Methodology of bus trip passenger flow as claimed in claim 1, is characterized in that, described step 5) in, use the bus trip passenger flow of least square method supporting vector machine prediction future time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610094427.1A CN105550789A (en) | 2016-02-19 | 2016-02-19 | Method for predicting bus taking passenger flow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610094427.1A CN105550789A (en) | 2016-02-19 | 2016-02-19 | Method for predicting bus taking passenger flow |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105550789A true CN105550789A (en) | 2016-05-04 |
Family
ID=55829970
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610094427.1A Pending CN105550789A (en) | 2016-02-19 | 2016-02-19 | Method for predicting bus taking passenger flow |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105550789A (en) |
Cited By (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105868861A (en) * | 2016-04-08 | 2016-08-17 | 青岛海信网络科技股份有限公司 | Bus passenger flow evolution analysis method based on time-space data fusion |
| CN106055607A (en) * | 2016-05-25 | 2016-10-26 | 百度在线网络技术（北京）有限公司 | User visiting prediction model establishment and user visiting prediction method and apparatus |
| CN106295868A (en) * | 2016-08-03 | 2017-01-04 | 重庆市索美智能交通通讯服务有限公司 | Traffic trip data processing method and device |
| CN106874432A (en) * | 2017-01-24 | 2017-06-20 | 华南理工大学 | A kind of public transport passenger trip space-time track extraction method |
| CN106960261A (en) * | 2017-03-27 | 2017-07-18 | 重庆邮电大学 | A kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data |
| CN107358319A (en) * | 2017-06-29 | 2017-11-17 | 深圳北斗应用技术研究院有限公司 | Flow Prediction in Urban Mass Transit method, apparatus, storage medium and computer equipment |
| CN107483741A (en) * | 2017-08-30 | 2017-12-15 | 珠海市魅族科技有限公司 | A kind of based reminding method and device, terminal and readable storage medium storing program for executing |
| CN107862865A (en) * | 2017-10-27 | 2018-03-30 | 沈阳世纪高通科技有限公司 | The Forecasting Methodology and device of a kind of trip information |
| CN108171970A (en) * | 2017-12-11 | 2018-06-15 | 北京奇虎科技有限公司 | A kind of method and device for obtaining passenger on public transport track |
| CN108242146A (en) * | 2016-12-27 | 2018-07-03 | 北京亿阳信通科技有限公司 | Based on mass transit card data analysis passenger ride website it is timely between method and system |
| CN108959596A (en) * | 2018-07-12 | 2018-12-07 | 成都智达万应科技有限公司 | A kind of public transport ladder fare-pricing estimate method |
| CN109035770A (en) * | 2018-07-31 | 2018-12-18 | 上海世脉信息科技有限公司 | The real-time analyzing and predicting method of public transport passenger capacity under a kind of big data environment |
| CN109308543A (en) * | 2018-08-20 | 2019-02-05 | 华南理工大学 | The short-term passenger flow forecasting of subway based on LS-SVM and real-time big data |
| CN109558985A (en) * | 2018-12-10 | 2019-04-02 | 南通科技职业学院 | A kind of bus passenger flow amount prediction technique based on BP neural network |
| CN110084442A (en) * | 2019-05-16 | 2019-08-02 | 重庆大学 | A kind of method of joint public transport and the progress passenger flow OD calculating of rail traffic brushing card data |
| CN110188921A (en) * | 2019-04-29 | 2019-08-30 | 阿里巴巴集团控股有限公司 | Vehicle scheduling strategy determines method and apparatus |
| CN110929982A (en) * | 2019-10-15 | 2020-03-27 | 江苏纬信工程咨询有限公司 | Bus line network passenger flow detection method and system based on fusion data |
| CN111046937A (en) * | 2019-12-05 | 2020-04-21 | 南京智慧交通信息有限公司 | Two-segment passenger crowd trip purpose analysis method fusing public transportation data and POI data |
| CN111275241A (en) * | 2020-01-08 | 2020-06-12 | 天津大学 | Bus passenger getting-off station inference method based on machine learning decision tree |
| CN111667121A (en) * | 2020-06-15 | 2020-09-15 | 常州市规划设计院 | Method for predicting initial passenger flow of rail transit line based on mobile phone signaling data |
| CN111932925A (en) * | 2020-07-09 | 2020-11-13 | 中咨数据有限公司 | Method, device and system for determining travel passenger flow of public transport station |
| CN112488419A (en) * | 2020-12-15 | 2021-03-12 | 平安国际智慧城市科技股份有限公司 | Passenger flow distribution prediction method, device, equipment and storage medium based on OD analysis |
| US10956844B2 (en) | 2019-04-29 | 2021-03-23 | Advanced New Technologies Co., Ltd. | Method and apparatus for determining vehicle scheduling strategy |
| CN113537714A (en) * | 2021-06-11 | 2021-10-22 | 华设设计集团股份有限公司 | Old people bus preference scheme evaluation method and system based on multi-source big data fusion analysis |
| CN113537682A (en) * | 2020-04-21 | 2021-10-22 | 上海大众祥云运输管理有限公司 | Real-time operation monitoring method and device for commuter vehicle |
| CN113658433A (en) * | 2021-08-18 | 2021-11-16 | 苏州工业园区测绘地理信息有限公司 | Method for extracting passenger flow characteristics based on bus card swiping and code scanning data |
| CN116187585A (en) * | 2023-04-19 | 2023-05-30 | 杭州数知梦科技有限公司 | Method, device and application for predicting BRT bus route of passenger |
| CN116994435A (en) * | 2023-08-09 | 2023-11-03 | 广州地铁设计研究院股份有限公司 | Bus OD acquisition method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103730008A (en) * | 2014-01-15 | 2014-04-16 | 汪涛 | Bus congestion degree analysis method based on real-time data of bus GPS (Global Position System) and IC (Integrated Circuit) cards |
| CN105185105A (en) * | 2015-06-26 | 2015-12-23 | 东南大学 | Bus transfer identification method based on vehicle GPS and bus IC card data |
-
2016
- 2016-02-19 CN CN201610094427.1A patent/CN105550789A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103730008A (en) * | 2014-01-15 | 2014-04-16 | 汪涛 | Bus congestion degree analysis method based on real-time data of bus GPS (Global Position System) and IC (Integrated Circuit) cards |
| CN105185105A (en) * | 2015-06-26 | 2015-12-23 | 东南大学 | Bus transfer identification method based on vehicle GPS and bus IC card data |
Cited By (40)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105868861A (en) * | 2016-04-08 | 2016-08-17 | 青岛海信网络科技股份有限公司 | Bus passenger flow evolution analysis method based on time-space data fusion |
| CN106055607A (en) * | 2016-05-25 | 2016-10-26 | 百度在线网络技术（北京）有限公司 | User visiting prediction model establishment and user visiting prediction method and apparatus |
| CN106055607B (en) * | 2016-05-25 | 2020-05-19 | 百度在线网络技术（北京）有限公司 | User visit prediction model establishment method and device and user visit prediction method and device |
| CN106295868A (en) * | 2016-08-03 | 2017-01-04 | 重庆市索美智能交通通讯服务有限公司 | Traffic trip data processing method and device |
| CN108242146A (en) * | 2016-12-27 | 2018-07-03 | 北京亿阳信通科技有限公司 | Based on mass transit card data analysis passenger ride website it is timely between method and system |
| CN108242146B (en) * | 2016-12-27 | 2020-10-16 | 北京亿阳信通科技有限公司 | Method and system for analyzing passenger bus station and time based on bus card data |
| CN106874432A (en) * | 2017-01-24 | 2017-06-20 | 华南理工大学 | A kind of public transport passenger trip space-time track extraction method |
| CN106874432B (en) * | 2017-01-24 | 2019-08-20 | 华南理工大学 | A kind of public transport passenger trip space-time trajectory extracting method |
| CN106960261A (en) * | 2017-03-27 | 2017-07-18 | 重庆邮电大学 | A kind of track traffic for passenger flow Forecasting Methodology based on track IC-card and mobile phone signaling data |
| CN107358319A (en) * | 2017-06-29 | 2017-11-17 | 深圳北斗应用技术研究院有限公司 | Flow Prediction in Urban Mass Transit method, apparatus, storage medium and computer equipment |
| CN107483741A (en) * | 2017-08-30 | 2017-12-15 | 珠海市魅族科技有限公司 | A kind of based reminding method and device, terminal and readable storage medium storing program for executing |
| CN107862865A (en) * | 2017-10-27 | 2018-03-30 | 沈阳世纪高通科技有限公司 | The Forecasting Methodology and device of a kind of trip information |
| CN107862865B (en) * | 2017-10-27 | 2020-01-07 | 沈阳世纪高通科技有限公司 | Travel information prediction method and device |
| CN108171970A (en) * | 2017-12-11 | 2018-06-15 | 北京奇虎科技有限公司 | A kind of method and device for obtaining passenger on public transport track |
| CN108959596A (en) * | 2018-07-12 | 2018-12-07 | 成都智达万应科技有限公司 | A kind of public transport ladder fare-pricing estimate method |
| CN108959596B (en) * | 2018-07-12 | 2020-07-24 | 成都智达万应科技有限公司 | Bus step fare prediction method |
| CN109035770A (en) * | 2018-07-31 | 2018-12-18 | 上海世脉信息科技有限公司 | The real-time analyzing and predicting method of public transport passenger capacity under a kind of big data environment |
| CN109035770B (en) * | 2018-07-31 | 2022-01-04 | 上海世脉信息科技有限公司 | Real-time analysis and prediction method for bus passenger capacity in big data environment |
| CN109308543A (en) * | 2018-08-20 | 2019-02-05 | 华南理工大学 | The short-term passenger flow forecasting of subway based on LS-SVM and real-time big data |
| CN109308543B (en) * | 2018-08-20 | 2021-10-22 | 华南理工大学 | Subway short-term passenger flow prediction method based on LS-SVM and real-time big data |
| CN109558985A (en) * | 2018-12-10 | 2019-04-02 | 南通科技职业学院 | A kind of bus passenger flow amount prediction technique based on BP neural network |
| US10956844B2 (en) | 2019-04-29 | 2021-03-23 | Advanced New Technologies Co., Ltd. | Method and apparatus for determining vehicle scheduling strategy |
| CN110188921A (en) * | 2019-04-29 | 2019-08-30 | 阿里巴巴集团控股有限公司 | Vehicle scheduling strategy determines method and apparatus |
| CN110084442A (en) * | 2019-05-16 | 2019-08-02 | 重庆大学 | A kind of method of joint public transport and the progress passenger flow OD calculating of rail traffic brushing card data |
| CN110929982A (en) * | 2019-10-15 | 2020-03-27 | 江苏纬信工程咨询有限公司 | Bus line network passenger flow detection method and system based on fusion data |
| CN110929982B (en) * | 2019-10-15 | 2023-11-24 | 江苏纬信工程咨询有限公司 | Bus network passenger flow detection method and system based on fusion data |
| CN111046937A (en) * | 2019-12-05 | 2020-04-21 | 南京智慧交通信息有限公司 | Two-segment passenger crowd trip purpose analysis method fusing public transportation data and POI data |
| CN111275241A (en) * | 2020-01-08 | 2020-06-12 | 天津大学 | Bus passenger getting-off station inference method based on machine learning decision tree |
| CN113537682A (en) * | 2020-04-21 | 2021-10-22 | 上海大众祥云运输管理有限公司 | Real-time operation monitoring method and device for commuter vehicle |
| CN113537682B (en) * | 2020-04-21 | 2023-05-09 | 上海大众祥云运输管理有限公司 | Real-time operation monitoring method and device for commuter vehicle |
| CN111667121A (en) * | 2020-06-15 | 2020-09-15 | 常州市规划设计院 | Method for predicting initial passenger flow of rail transit line based on mobile phone signaling data |
| CN111932925A (en) * | 2020-07-09 | 2020-11-13 | 中咨数据有限公司 | Method, device and system for determining travel passenger flow of public transport station |
| CN112488419A (en) * | 2020-12-15 | 2021-03-12 | 平安国际智慧城市科技股份有限公司 | Passenger flow distribution prediction method, device, equipment and storage medium based on OD analysis |
| CN113537714A (en) * | 2021-06-11 | 2021-10-22 | 华设设计集团股份有限公司 | Old people bus preference scheme evaluation method and system based on multi-source big data fusion analysis |
| CN113537714B (en) * | 2021-06-11 | 2023-11-24 | 华设设计集团股份有限公司 | Old people public transportation preference scheme assessment method and system based on multi-source big data fusion analysis |
| CN113658433A (en) * | 2021-08-18 | 2021-11-16 | 苏州工业园区测绘地理信息有限公司 | Method for extracting passenger flow characteristics based on bus card swiping and code scanning data |
| CN116187585A (en) * | 2023-04-19 | 2023-05-30 | 杭州数知梦科技有限公司 | Method, device and application for predicting BRT bus route of passenger |
| CN116187585B (en) * | 2023-04-19 | 2023-07-14 | 杭州数知梦科技有限公司 | Method, device and application for predicting BRT bus route of passenger |
| CN116994435A (en) * | 2023-08-09 | 2023-11-03 | 广州地铁设计研究院股份有限公司 | Bus OD acquisition method and device |
| CN116994435B (en) * | 2023-08-09 | 2024-05-03 | 广州地铁设计研究院股份有限公司 | Bus OD acquisition method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105550789A (en) | Method for predicting bus taking passenger flow | |
| Zou et al. | Detecting home location and trip purposes for cardholders by mining smart card transaction data in Beijing subway | |
| Lin et al. | A review of travel-time prediction in transport and logistics | |
| CN102157075B (en) | Method for predicting bus arrivals | |
| CN101615340A (en) | Real-time information processing method in the bus dynamic dispatching | |
| CN105185115A (en) | Vehicle forecasting method and forecasting system | |
| CN102324128A (en) | Method for predicting OD (Origin-Destination) passenger flow among bus stations on basis of IC (Integrated Circuit)-card record and device | |
| CN104700646A (en) | Online GPS data based abnormal taxi track real-time detection method | |
| Hadavi et al. | Monitoring urban-freight transport based on GPS trajectories of heavy-goods vehicles | |
| Gong et al. | Hybrid dynamic prediction model of bus arrival time based on weighted of historical and real-time GPS data | |
| CN105046350A (en) | AFC data-based public transport passenger flow OD real-time estimation method | |
| CN104169990A (en) | Method for providing parking information on free parking spaces | |
| CN106448132A (en) | Conventional public traffic service index real-time evaluation system and method | |
| CN105489053A (en) | Establishment method for parallel parking system based on ACP method | |
| Čelan et al. | Bus-arrival time prediction using bus network data model and time periods | |
| CN105469602A (en) | Method for predicting bus passenger waiting time range based on IC card data | |
| Ram et al. | SMARTBUS: A web application for smart urban mobility and transportation | |
| Liyanage et al. | AI-based neural network models for bus passenger demand forecasting using smart card data | |
| CN116629738B (en) | Logistics path optimization method, related method, device, equipment and medium | |
| Montini et al. | Searching for parking in GPS data | |
| Gómez et al. | Evaluation of trade-offs between two data sources for the accurate estimation of origin–destination matrices | |
| CN106295868A (en) | Traffic trip data processing method and device | |
| JP2016030473A (en) | Traffic analysis system, traffic analysis program and traffic analysis method | |
| Sandim et al. | Using GPS-based AVL data to calculate and predict traffic network performance metrics: A systematic review | |
| Tu et al. | Analysis and forecast of passenger flow based on public transportation IC card and GPS data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160504 |