CN104183119B  Based on the antiarithmetic for realtime traffic flow distribution forecasting method pushed away of section OD  Google Patents
Based on the antiarithmetic for realtime traffic flow distribution forecasting method pushed away of section OD Download PDFInfo
 Publication number
 CN104183119B CN104183119B CN201410410008.5A CN201410410008A CN104183119B CN 104183119 B CN104183119 B CN 104183119B CN 201410410008 A CN201410410008 A CN 201410410008A CN 104183119 B CN104183119 B CN 104183119B
 Authority
 CN
 China
 Prior art keywords
 traffic
 time
 volume
 real
 distribution
 Prior art date
Links
 239000011159 matrix materials Substances 0.000 claims description 26
 230000004087 circulation Effects 0.000 claims description 15
 239000003102 growth factors Substances 0.000 claims description 15
 230000035915 Distribution volume Effects 0.000 claims description 12
 238000000034 methods Methods 0.000 claims description 12
 238000004364 calculation methods Methods 0.000 claims description 8
 230000000875 corresponding Effects 0.000 claims description 8
 230000004907 flux Effects 0.000 claims description 6
 230000001808 coupling Effects 0.000 claims description 5
 238000010168 coupling process Methods 0.000 claims description 5
 238000005859 coupling reactions Methods 0.000 claims description 5
 238000000605 extraction Methods 0.000 claims description 4
 238000004220 aggregation Methods 0.000 claims description 3
 230000002776 aggregation Effects 0.000 claims description 3
 238000005096 rolling process Methods 0.000 claims description 3
 238000004642 transportation engineering Methods 0.000 claims description 3
 239000000284 extracts Substances 0.000 claims description 2
 238000005516 engineering processes Methods 0.000 abstract description 2
 230000000694 effects Effects 0.000 description 4
 238000004458 analytical methods Methods 0.000 description 2
 238000010586 diagrams Methods 0.000 description 2
 238000009499 grossing Methods 0.000 description 2
 280000648038 Digital City companies 0.000 description 1
 280000105711 Smart City companies 0.000 description 1
 230000003044 adaptive Effects 0.000 description 1
 238000007796 conventional methods Methods 0.000 description 1
 238000007667 floating Methods 0.000 description 1
 230000014759 maintenance of location Effects 0.000 description 1
 238000004393 prognosis Methods 0.000 description 1
 230000003068 static Effects 0.000 description 1
 238000006467 substitution reactions Methods 0.000 description 1
 238000000714 time series forecasting Methods 0.000 description 1
Abstract
Description
Technical field
The present invention relates to technical field of intelligent traffic, especially relate to generation and the traffic attraction of a kind of road grid traffic Realtime predicting method.
Background technology
Build and extensively applying and real time information system in management in smart city along with intelligent transportation system Growing, traffic administration gradually forms intelligent, Statistical, the feature of realtime.Vehicle supervision department Obtain substantial amounts of real time traffic data with research institution, and be applied to the field of traffic control such as realtime road. Substantial amounts of traffic information data builds historical data base, provides important data for traffic administration person and research worker Support.In order to carry out effective traffic administration and provide Traffic Information accurately for traveler, need base Trip generation forecast and traffic attraction prediction accurately is carried out in Realtime Traffic Information.Carry out in shortterm based on real time traffic data Traffic trip and traffic attraction prediction contribute to trip distribution modeling, this is by the important of traffic administration and planning Part.
Passing through traffic study, according to personnel's structure of each traffic zone traffic trip and the prediction of traffic attraction at present more The factors such as onetenth, motor vehicles retention calculate travel amount and the traffic attraction of traffic zone.First, the friendship of large sample amount Logical investigation needs substantial amounts of time and fund input；Secondly, divide the scope of traffic zone and can make according to difference Become the inaccurate of prediction, and the prediction of traffic zone cannot carry out Proper Match with traffic network, it was predicted that result Disagree with practical situation；3rd, based on static investigation and prediction be difficult to reflect realtime traffic stream mode and Carry out accurate shorttime traffic flow forecast.
Summary of the invention
In order to overcome abovementioned the deficiencies in the prior art, the invention provides a kind of based on section OD counter push away realtime Traffic flow distribution forecasting method, is traffic trip and the suction dividing the realtime OD distribution carried out based on traffic zone The Forecasting Methodology of the amount of drawing, for improving the accuracy of trip forecast of distribution.
To achieve these goals, the technical solution adopted in the present invention is:
The present invention be to " traffic zone based on section divide, the volume of traffic based on road section traffic volume community the most pre Survey, the distribution of OD Prediction of Traffic Flow in Realtime, link flow " carry out the integration of system, build based on section Traffic zone division methods and the Forecasting Methodology of realtime traffic flow distribution, it is achieved to travelling OD between hand in real time The prediction of flux, and further according to shortest path distribution method, by OD between realtime estimate traffic Amount is assigned in each section of road network, obtains realtime estimate traffic flow distribution situation.
A kind of based on the antiarithmetic for realtime traffic flow distribution forecasting method pushed away of section OD, comprise the following steps:
1.1, every section in transportation network is considered as traffic zone, to transport information (Floating Car GPS gathered Data) carry out OD extraction and section coupling；
1.2, according to prediction time, this is preto choose a few sky in the Floating Car OD traffic data storehouse of history The OD data of each half an hour before and after the survey moment, and the calculating that is averaged, based on result of calculation OD matrix, substitutes in OD matrix；
1.3, according to prediction time, the OD traffic data in the first two moment prediction time on the same day is chosen, right Its Time Series Method once moved, obtains prediction time each traffic zone (section) O, D Traffic volume forecast value, and as in the aggregation of forecasts value substitution OD matrix of each O, D；
1.4, push away travelling OD matrix is counter according to balanced growth Yfactor method Y, obtain OD under predicted time The volume of traffic between to；
1.5, according to the algorithm of shortest path, the prediction OD volume of traffic is carried out traffic flow distribution, obtains in real time Road traffic delay prediction.
Compared with prior art, the invention has the beneficial effects as follows: section is considered as traffic zone, mesh can be solved The problem that go on a journey in front traffic zone, the attraction volume of traffic is bad with road network, meanwhile, can reduce traffic zone The cost of investigation, obtains the volume of traffic gone on a journey between traffic zone, attract by real time traffic data, solves real Time traffic assignation forecasting problem.
Accompanying drawing explanation
Fig. 1 is the flow chart of the open method of the present invention.
Fig. 2 is to use the traffic zone division methods of the present invention to interrupt the road network of Panyu District of Guangzhou City to carry with OD Take schematic diagram.
Fig. 3 is that 01 distributes method flow chart.
Fig. 4 is that increment distributes method flow chart schematic diagram.
Fig. 5 is straight average method flow chart.
Detailed description of the invention
The present invention is further described below in conjunction with the accompanying drawings.
The techniqueflow of the present invention as shown in Figure 1, including division, the data base of traffic zone based on section Structure, realtime OD traffic volume forecast, arithmetic for realtime traffic flow distribution four steps.
1.1, traffic zone based on section divides:
The unit of traditional traffic fourphase model is traffic zone, the present invention section is replaced traffic zone as The unit of fourphase model, the section to whole region, according to the principle having crossing just to interrupt by the road in region Net interrupts the section into more junior unit.Traffic zone is instead based on section, antiwith the OD that this serves below Push away and traffic assignation.Specific algorithm flow process is as follows:
Choose the basic road network of research, interrupt according to crossing and section road axis angle of turning is less than The principle that the corner of 150 ° is interrupted, is broken into the section of less yardstick by Regional Road Network, the section after interrupting Traffic zone will be replaced, it is numbered and adds up the minimum single of OD travel amount as fourphase model Unit.Traditional OD matrix is all with traffic analysis community as analytic unit, so hands over carrying out Path selection Need the traffic flow of each traffic zone to match on the road network of correspondence when of throughflow distribution, this process meeting Introduce error.The most directly use section as the analytic unit of OD matrix, can avoid at this Systematic error is introduced during one.
1.2, the structure of data base:
The structure of data base be based on gps data is carried out realtime OD extraction, section coupling and based on section OD collect statistics three parts.
1. O, D extract in real time: be first ranked up gps data according to car number and date, time, Then O point or D point are judged according to the change of passenger carrying status field in gps data, such as: carrying shape In state, 1 is loadcarrying, and 0 is unloaded, then when the passenger carrying status of same car is become 0 from 1, for D；When same The passenger carrying status of one car from 0 become 1 time, for O；When finding O point, the most down find corresponding D Point；When finding D point, then corresponding O point is found in backtracking, mates OD point with this right.
2. section coupling: O, D of extracting point is converted to shapefile dot format, uses nearest section Method coupling section, i.e. search out the section that GPS point is nearest, and section numbering be assigned to GPS point.
3. OD based on section collects statistics: to matching the GPS OD point in section, segment number of satisfying the need is carried out Collect statistics, passenger carrying status is carried out counting statistics, obtain O based on section, D matrix.
1.3, realtime OD traffic volume forecast:
In real time OD traffic volume forecast is divided into three steps: first, determines basis OD matrix according to historical data；The Two, determine OD Matrix prediction value according to real time data；3rd, OD are counter estimates the OD friendship calculating prediction time Flux.
1. basis OD matrix is determined according to historical data
Historical data based on predicted time point and the proportionate relationship of history Road network traffic, be predicted.As wanted In prediction morning 8:00 road network each OD between the volume of traffic, by extracting each OD of road network to 8: The historical traffic amount of 00, and it is averaged value as historical data, bring in OD matrix.
2. OD Matrix prediction value is determined according to real time data
Forecasting Methodology based on lineary system theory specifically includes that linear statistical recurrence, historical trend method, time Sequence prediction method, Kalman prediction method, Smoothing Forecasting and adaptive weighting forecast model Method etc..The present embodiment carries out shortterm traffic flow as a example by Time Series Forecasting Methods Single moving average method Prediction.
If x_{1}, x_{2}, x_{3}..., x_{n}For specific O point, the life with t as time interval, in every period Become the volume of traffic, or specific D point, the attraction volume of traffic with t as time interval, in every period.
Therefore, it is a time series, and sample size is n.
The formula of moving average method is as follows:
Wherein, N is the data amount check that each rolling average comprises；It it is the Single moving average of T phase Value.
The effect of rolling average is smoothing data, eliminates some random disturbances, makes longterm trend reveal, Thus can be used for trend analysis and prediction.
Generally, if time series does not has obvious mechanical periodicity and Longterm change trend, the available T phase Single moving average value is as the predictive value of T+1 phase, it may be assumed that
Abbreviation obtains recurrence formula:
When n is large, amount of calculation can be greatly reduced with recurrence formula.
Therefore, according to OD between historical data, the volume of traffic in following time interval is predicted.
Generation traffic volume forecast value and the attraction of each D obtaining each O in OD matrix may finally be added up Traffic volume forecast value, by historical data base, it was predicted that around the moment, the traffic flow data in the range of certain time is pressed It is assigned in OD matrix according to origin and destination, as the basis matrix of OD matrix.Meanwhile, according to time series side Method and the traffic flow data in several moment before prediction time, extrapolate the origin and destination traffic volume forecast of prediction time Value, as the predictive value in OD matrix.
3. OD is counter estimates the realtime OD volume of traffic
This research uses the distribution of growth factor methed prognosis traffic volume.
The algorithm steps of growth factor methed is as follows:
The first step, makes calculation times m=0
Second step, in given present situation OD tableU in Tm and in the future OD table_{i}、V_{j}；
WhereinRepresent the m time when calculating, the volume of traffic between traffic zone i to traffic zone j；
Representing the m time when calculating, there is the volume of traffic in ith traffic zone；
Represent the m time when calculating, the attraction volume of traffic of jth traffic zone；
U_{i}Represent the generation volume of traffic in OD table in future；
V_{j}Represent the attraction volume of traffic in OD table in future；
3rd step, obtains occurring and the rate of increase attracting the volume of traffic of each community
4th step, seeks the approximation of the m+1 time distribution volume of traffic
5th step, convergence differentiates, it determines formula is:
In formula, U_{i}Represent the generation volume of traffic in OD table in future；
V_{j}Represent the attraction volume of traffic in OD table in future；
Represent that the m time calculating of i community occurs growth factor；
Represent that the m time calculating of j community attracts growth factor；
It is the iterative manner of growth factor, relevant with every kind of method；
ε represents any given error constant, typically takes 3%.
According to functionKind different, growth factor methed can be divided into normal growth factor methed (Unique Growth Factor Method), balanced growth Yfactor method Y (Average Growth Factor Method), the end special Law (Detroit Method), Follett, Mary Parker method (Fratar Method) and Buddhist Nice method (Furness Method).
The present invention uses balanced growth Yfactor method Y, and the method formula is simple and clear in calculating, and shortcoming is that convergence is slow, repeatedly In generation, computational accuracy was low often.Balanced growth coefficient is expressed as follows:
First approximation:
(4) arithmetic for realtime traffic flow distribution:
Generally traffic flow distribution method is divided into equilibrium assignmen and the big class of nonequilibrium distribution two in the world.For completely Meet the poised state of Wardrop principle definition, referred to as EqP method；For use heuristic or its The distribution model of his approximation method, the most nonequilibrium distribution method.
Obtained after realtime OD matrix by OD estimation model, then be respectively adopted 01 distribution method, increment divides Join method, straight average method etc. three kinds nonequilibrium allocation algorithm development behavior traffic flow distribution model, by new realtime OD Traffic growth rate, to road network, obtains the arithmetic for realtime traffic flow distribution situation of road network.
1. 01 distributes method
01 distribution method also referred to as has entirely completely without distribution method, is the most most basic Traffic growth rate method, in the U.S. In the traffic programme of city, Chicago, obtain application first.This algorithm has 2 features: one is to think the out of shape of section Time is the most relevant with the length in section, is a definite value, unrelated with the size of the volume of traffic on section, Bu Sui road The degree of crowding on road and be varied from；Two is to think that all drivers of OD of same group select identical Route.The main processes of calculation of this distribution method is to find shortest path.
01 distribution method is simple and efficiency is high.It is applicable in the rare remote districts of road, and general city road network Traffic growth rate in be not suitable for using this distribution method.Calculation procedure can be summarized as follows:
The first step, calculate each OD in network between shortest path；
Second step, is fully allocated to the volume of traffic between OD on corresponding shortest path.
Owing to 01 distribution method can not reflect crowded effect, it is mainly used for some uncongested road network, this distribution method For there is no the situation of network that the traffic capacity limits.Therefore, it is suggested that range is: road between intercity The unrestricted area of the traffic capacity can use；The traffic flow distribution of general urban road network should not use the party Method.In practice due to its simple and practical characteristic, generally as the basis of other various distribution techniques, increasing Reusability in the amount method such as distribution method and EqP method.
2. increment distribution method
Increment distribution method is the EqP method of a kind of approximation.It is that the OD volume of traffic is divided equally into some equal portions, Cyclically by the OD Traffic growth rate of each equal portions to network.The OD of cycle assignment one equal portions hands over each time Flux, on corresponding shortest path, often distributes the travel time the most just recalculating and updating each section, so The shortest path between each OD of network is recalculated afterwards by the travel time after updating.Subsequent cycle is pressed after updating Shortest path distribute the OD volume of traffic of next equal portions.Its calculation procedure is as follows:
The first step, initializes.It is divided equally into N equal portions, order by often organizing the OD volume of traffic
Second step, updates the travel time in section. Calculate shortest path between OD.
3rd step, increment distributes.Gained is calculated by second stepBy 01 distribution method by the OD traffic of 1/N AmountIt is assigned in network.So obtain one group of additional traffic amount
4th step, the volume of traffic adds up.Order
5th step, it is determined that.If n=N, stop calculating.Current link counting is i.e. last solution；As Really n ＜ N, makes n=n+1, returns second step.
Increment distribution method obtains complexity and solves accuracy all between 01 distribution method and EqP method. The result distributing method as N=1 with 01 is consistent；As N → ∞, its solution is consistent with the solution of EqP method. The method is simple, is all provided with using in many ripe business softwares, and degree of accuracy can be according to N's Size regulates, and is often applied in actual road network Traffic growth rate.But this distribution method is still It is a kind of approximation method, sometimes by section the least for too much Traffic growth rate to some capacity.
3. straight average method
Straight average method is a kind of roundrobin technique between increment distribution method and EqP method.It is Constantly adjustment has been assigned to the volume of traffic on each section and has moved closer to or arrive equilibrium assignmen.Often walking circulation In, carry out a 01 distribution according to the volume of traffic being assigned on each section and obtain the additional of one group of each section The volume of traffic.Then the additional traffic amount obtained in circulating with this with the distribution volume of traffic in section each in last circulation It is weighted the section distribution volume of traffic that is average and that obtain in this circulation.When the distribution in adjacent two circulations is handed over Flux equal to definite value time, can stop calculate, this equivalence is a value the least, approximates 1, i.e. it is believed that The distribution volume of traffic in adjacent two circulations is sufficiently close to.The distribution volume of traffic obtained in last circulation It it is i.e. the final volume of traffic.Its main algorithm step is as follows:
The first step, initializes.Carry out a 01 distribution according to the free travel time in each section, obtain each road The distribution volume of traffic of section
Second step, according to the distribution volume of traffic in current each sectionCalculate each Link Travel Time, and calculate OD Between shortest path.
3rd step, the Link Travel Time calculated according to second step and the OD volume of traffic carry out 01 distribution, Obtain additional traffic amount F in each section_{a}。
4th step, calculates the current amount of traffic in each section by the method for weighted mean methodAs follows:
5th step, ifThen loop termination, otherwise restarts 01 distribution, Perform circulation.
In the 4th step and the 5th step, weight coefficient θ and parameter K need to be defined by reckoner oneself.Parameter Number of times and the operation efficiency of circulation that can determine circulation is set.
Straight average method is the most simply to be suitable for, again closest to a kind of distribution method of EqP method；Often step is followed In ring, the value of weight coefficient θ is in strict accordance with mathematical programming model value, can be balanced the solution of distribution.
By above method, the service level of road network can be obtained with Realtime and Dynamic, for special traffic events (as interior Flood, vehicle accident etc.) traffic flow is dredged under situation offer decision support.
Abovementioned based on the antiarithmetic for realtime traffic flow distribution forecasting method pushed away of section OD, combine traffic zone and draw Point, shortterm traffic flow prediction, OD Prediction of Traffic Flow in Realtime, the aspect such as link flow distribution, it is achieved that to going out Row OD between the prediction of realtime traffic amount, and further according to the method for shortest path distribution, by OD pair Between realtime estimate Traffic growth rate in each section of road network, obtain realtime estimate traffic flow distribution feelings Condition.
Compared with conventional method, the invention has the beneficial effects as follows: one, section is considered as traffic zone by the present invention, The problem that go on a journey in current traffic zone, the attraction volume of traffic is bad with road network can be solved；Two, the present invention can With the cost of reduction traffic zone investigation, obtain going on a journey between traffic zone by real time traffic data, attract The volume of traffic, solves realtime traffic distribution forecast problem.
Based on above feature, the present invention announce based on the antiarithmetic for realtime traffic flow distribution forecasting method pushed away of section OD Can play a great role at aspects such as traffic flow realtime estimate, road condition assessment, also can be digital city and wisdom The development in city increases powerassisted.
Embodiment described above only have expressed the embodiment that the present invention is possible, and it describes more concrete and detailed, But therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that, for this area For those of ordinary skill, without departing from the inventive concept of the premise, it is also possible to make some deformation and improvement, These broadly fall into protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be with claims It is as the criterion.
Claims (16)
Priority Applications (1)
Application Number  Priority Date  Filing Date  Title 

CN201410410008.5A CN104183119B (en)  20140819  20140819  Based on the antiarithmetic for realtime traffic flow distribution forecasting method pushed away of section OD 
Applications Claiming Priority (1)
Application Number  Priority Date  Filing Date  Title 

CN201410410008.5A CN104183119B (en)  20140819  20140819  Based on the antiarithmetic for realtime traffic flow distribution forecasting method pushed away of section OD 
Publications (2)
Publication Number  Publication Date 

CN104183119A CN104183119A (en)  20141203 
CN104183119B true CN104183119B (en)  20160824 
Family
ID=51964122
Family Applications (1)
Application Number  Title  Priority Date  Filing Date 

CN201410410008.5A CN104183119B (en)  20140819  20140819  Based on the antiarithmetic for realtime traffic flow distribution forecasting method pushed away of section OD 
Country Status (1)
Country  Link 

CN (1)  CN104183119B (en) 
Families Citing this family (13)
Publication number  Priority date  Publication date  Assignee  Title 

CN104731963B (en) *  20150403  20180615  重庆邮电大学  Recommend method and system in a kind of gridding path based on car networking 
CN105006148B (en) *  20150706  20171031  同济大学  A kind of crossing turns to vehicle number estimation method and system 
CN106504535B (en) *  20161130  20181012  东南大学  A kind of trip distribution modeling method of combination Gravity Models and Fratar models 
CN108257378B (en) *  20161229  20200519  杭州海康威视数字技术股份有限公司  Traffic flow prediction method and device 
CN106951549A (en) *  20170327  20170714  重庆邮电大学  A kind of passenger's traffic path recognition methods based on track ICcard and mobile phone signaling data 
CN107038862B (en) *  20170504  20200214  东南大学  Traffic distribution correction method under tail number restriction condition 
CN107294794A (en) *  20170731  20171024  国网辽宁省电力有限公司  A kind of largescale ip communication service matrix estimation method and system 
CN107798877B (en) *  20171122  20201002  北京云星宇交通科技股份有限公司  Method and system for predicting traffic volume based on highway charging data 
CN108269399B (en) *  20180124  20201225  哈尔滨工业大学  Highspeed rail passenger flow demand prediction method based on road network passenger flow OD reverse thrust technology 
CN108711286A (en) *  20180529  20181026  重庆市交通规划研究院  A kind of Traffic growth rate method and system based on multisource car networking and mobile phone signaling 
CN109215341A (en) *  20180905  20190115  中山大学  A kind of bus trip data creation method comprising information of getting off 
CN109493449B (en) *  20181123  20200526  北京航空航天大学  Freight train cargo state estimation method based on freight train GPS track data and highspeed transaction data 
CN110363994B (en) *  20190725  20200825  北京建筑大学  Method for acquiring dynamic traffic volume of whole road sections of road network 
Family Cites Families (6)
Publication number  Priority date  Publication date  Assignee  Title 

JP3414843B2 (en) *  19930622  20030609  三菱電機株式会社  Transportation control device 
DE10022812A1 (en) *  20000510  20011122  Daimler Chrysler Ag  Method for determining the traffic situation on the basis of reporting vehicle data for a traffic network with trafficregulated network nodes 
CN101436345B (en) *  20081219  20100818  天津市市政工程设计研究院  System for forecasting harbor district road traffic requirement based on TransCAD macroscopic artificial platform 
CN101930668B (en) *  20090429  20120822  上海电器科学研究所（集团）有限公司  Processing method of road traffic OD (Optical Density) information collection system for license plate recognition 
US8798897B2 (en) *  20101101  20140805  International Business Machines Corporation  Realtime traffic analysis through integration of road traffic prediction and traffic microsimulation models 
CN102682598B (en) *  20120531  20140326  山东交通学院  Traffic information distribution method capable of eliminating oscillation 

2014
 20140819 CN CN201410410008.5A patent/CN104183119B/en active IP Right Grant
Also Published As
Publication number  Publication date 

CN104183119A (en)  20141203 
Similar Documents
Publication  Publication Date  Title 

Liang et al.  A deep reinforcement learning network for traffic light cycle control  
Kong et al.  Urban traffic congestion estimation and prediction based on floating car trajectory data  
Liang et al.  Deep reinforcement learning for traffic light control in vehicular networks  
Yao et al.  Short‐term traffic speed prediction for an urban corridor  
US9424515B2 (en)  Predicting taxi utilization information  
Hofleitner et al.  Arterial travel time forecast with streaming data: A hybrid approach of flow modeling and machine learning  
CN105513400B (en)  The method of Dynamic Programming trip route  
BenAkiva et al.  Dynamic model of peak period traffic congestion with elastic arrival rates  
Geroliminis et al.  Cordon pricing consistent with the physics of overcrowding  
Chang et al.  Dynamic multiinterval bus travel time prediction using bus transit data  
Chen et al.  Freeway performance measurement system: mining loop detector data  
CN103413443B (en)  Shortterm traffic flow forecasting method based on hidden Markov model  
CN102629418B (en)  Fuzzy kalman filteringbased traffic flow parameter prediction method  
CN103678917B (en)  A kind of realtime arrival time Forecasting Methodology of public transport based on simulated annealing  
Yuan et al.  Tdrive: driving directions based on taxi trajectories  
CN106971545B (en)  A kind of bus arrival time prediction technique  
CN104636828B (en)  Based on markovian public bicycles website supply and demand prediction method  
Mazloumi et al.  An integrated framework to predict bus travel time and its variability using traffic flow data  
CN103632212B (en)  System and method for predicating timevarying user dynamic equilibrium networkevolved passenger flow  
CN105513414A (en)  Parkinglot parking space predication and recommendation method based on realtime traffic query and cloud model  
CN105938572B (en)  A kind of more automatic guided vehicle paths planning methods of the pre antiinterference of logistics storage system  
CN104157139A (en)  Prediction method and visualization method of traffic jam  
Li et al.  Analytical hierarchy process using fuzzy inference technique for realtime route guidance system  
Gao et al.  Cognitive cost in route choice with realtime information: An exploratory analysis  
CN103295414A (en)  Bus arrival time forecasting method based on mass historical GPS (global position system) trajectory 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  
C14  Grant of patent or utility model  
GR01  Patent grant 