CN109886508A - Taxi high benefit based on GPS big data seeks objective multiple objective programming algorithm - Google Patents
Taxi high benefit based on GPS big data seeks objective multiple objective programming algorithm Download PDFInfo
- Publication number
- CN109886508A CN109886508A CN201910237145.6A CN201910237145A CN109886508A CN 109886508 A CN109886508 A CN 109886508A CN 201910237145 A CN201910237145 A CN 201910237145A CN 109886508 A CN109886508 A CN 109886508A
- Authority
- CN
- China
- Prior art keywords
- taxi
- grid
- objective
- data
- period
- 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
Landscapes
- Traffic Control Systems (AREA)
Abstract
A kind of taxi high benefit based on GPS big data seeks objective multiple objective programming algorithm.It includes: the running data and financial data that (1) obtains taxi from GPS big data, and driving trace and financial information are matched；(2) grid dividing is carried out to city district；(3) running data of taxi and financial data are matched with grid, obtains the total record number by bus occurred in each grid and the total amount that record by bus occurs, establishes multiple objective function；(4) region for the transaction record that taxi generates and each grid are matched and obtains the relation between supply and demand of taxi and passenger in each grid and obtains obtaining the objective average latency, be introduced into and correct multiple objective function；(5) it is obtained according to revised multiple objective function and recommends mesh coordinate.Present invention combination GPS big data analyzes the index factor for influencing traveller's benefit, takes into account performance indicator and transport power index, effectively promotes the income of taxi driver and improves urban transportation transport power.
Description
Technical field
The present invention relates to big data field, in particular to a kind of taxi high benefit based on GPS big data seeks the more mesh of visitor
Mark planning algorithm.
Background technique
For urban transportation in addition to public transport, subway, taxi plays very important role in the trip of people.But it is existing
Often occur such awkward scene in reality: being that taxi expires street and turns on one side, another side is that passenger beats less than vehicle；Go out on one side
It hires a car and flocks together, another side is that a vehicle does not have yet.Such case with the nets about vehicle platform such as " drop drip ", " first vapour " appearance
Take on a new look, but commerial vehicle it is unloaded after still remain and blindly travel to this period of order, vehicle bundle in hot spot region pushes away etc. now
As, for passenger, still remain the waiting time of calling a taxi it is too long the problems such as, urban transportation efficiency has to be hoisted.
Urban Residential Trip has randomness, and different time, the Passenger's distribution of different zones are also uneven, and can be with
The development in city, the expansion of road and the variation of surrounding enviroment and quickly change, by experience realize trip supply and demand it is accurate
Docking is impossible, it is necessary to is just able to achieve by the research that collects and analyzes of big data.The collection of taxi GPS big data is
Real-time and transparent, it has recorded pick-up time, time getting off, the Entrucking Point, place of getting off, mileage travelled, expense of passenger
Etc. information, this be excavate passenger's random behavior behind trip rule, realize with taxi operation benefit, transport power and traffic resource
Equiblibrium mass distribution is Multiple Optimization target, formulates real-time high-efficiency and seeks objective strategy and provides the foundation data.
Currently, seeking objective strategy study by GPS data from taxi development both at home and abroad, but the taxi of this quasi-tradition seeks objective plan
Bigger majority only focuses on unit transport power index, seldom considers unit performance indicator.It is many often to will appear passenger carrying capacity in this way, but
It is very high but to very long the time spent in next carrying that every list amount of money takes in not high instead or single carrying income less, leads to list
Position decline in benefits.
Summary of the invention
The recommendation in objective strategy concern passenger carrying capacity or carrying path is sought in order to solve current taxi in background technique, but is not had
There is the problem of promoting carrying benefit, the present invention provides a kind of going out based on GPS big data for taking into account performance indicator and transport power index
High benefit of hiring a car seeks objective multiple objective programming algorithm.
The technical solution adopted by the present invention to solve the technical problems is: a kind of taxi based on GPS big data is efficient
Benefit seeks objective multiple objective programming algorithm, comprising the following steps:
Step 1) data prediction: the historical data of taxi, including running data and wealth are obtained from GPS big data
Business data, according to taxi mark by driving trace and corresponding financial information phase in the running data and financial data
Matching；
Step 2) region division: grid dividing is carried out to city district, each grid is denoted as (i, j), i=1,2 ... m, j
=1,2 ... n；
Each taxi is completed the running data after matching and the net after financial data and division by step 3)
Lattice match, and extract total record number N by bus that (i, j) a grid within the t period occursij(t) and occur by bus record
Total amount Aij(t), multiple objective function is established
Wherein, (i, j) is the net region that taxi is currently located, and (s, k) is the net region that taxi is gone to, Nsk
It (t) is the total quantity by bus occurred in t period (s, k) a net region, TijskTo go to from a net region (i, j)
The running time of a net region (s, k), Tsk(t) for the t period occurs a net region (s, k) in ride record be averaged
Running time, tijskIt is the t period from a net region (i, j), reaches the period of a net region (s, k)；
Step 4) correct multiple objective function: by taxi in GPS big data get on the bus record and empty wagons record and each grid phase
Matching obtains the average latency TA that unloaded taxi in each grid obtains visitor, i.e., empty wagons is in t period, the net region (s, k)
Obtain the objective time
1) it is lined up mode of getting on the bus (fixation such as airport is got on the bus a little), a newly-increased empty wagons obtains the objective waiting time
2) mode of getting on the bus at random (on-demand stop etc.), a newly-increased empty wagons obtain the objective waiting time
Wherein,
Nsk(t): for t period being averaged to get on the bus and record number in (s, k) grid
NEsk(t): recording number in the average empty wagons of (s, k) grid for the t period (if without empty wagons data, can use note of getting off
It is quasi- to record digital-to-analogue)
It obtains objective average latency TA by described and introduces and correct multiple objective function
Step 5) is obtained according to revised multiple objective function recommends mesh coordinate (s, k).
The step 4) further includes introducing constraint condition to carry out constraint, the constraint packet to multiple objective function
It includes:
1) operating range constrains:
Wherein, dijskFor (i, j) a grid to (s, k) a grid distance, UxIndicate corresponding in the grid once weft
Physical length, UyIndicate in the corresponding physical length of the grid once warp, any latitude it is once the same through line length, and
Once the length of weft was related to the longitude θ of administrative region, and had Ux=Uycos(θ)；xij, yijFor in a region (i, j)
Heart point longitude and latitude, xsk, yskFor (s, k) a regional center point longitude and latitude；
2) running time constrains:
Wherein, vijIndicate the average speed for the record by bus that the t period occurs in a net region (i, j)；
3) it reaches period constraint: setting out and hire a car in the t period from a net region (i, j), spend TijskHour arrives
Up to a net region (s, k), period when arrival is tijsk, then
tijsk=mod ([t+Tijsk],24)
Wherein, [x] expression is rounded downwards x, and mod (M, N) refers to M to the remainder of N；
The multiple objective function obtains Model for Multi-Objective Optimization after the constraint:
The beneficial effects of the present invention are: the present invention, by carrying out slice division to the time, the web area that satisfies the need carries out gridding
It divides, calculates the volume of the flow of passengers distribution, benefit distribution and the carrying probability of success of arbitrary mess arbitrary period, go to mesh in conjunction with taxi
The time cost for marking region, establishes Optimized model using high benefit, supply-demand structure as multiple target, acquires and most preferably seeks visitor's recommendation point,
The taxi high benefit for forming complete set seeks objective Generalization bounds, promotes the income of taxi driver and improves urban transportation fortune
Power.
Detailed description of the invention
Fig. 1 is that the high benefit of the multiple objective programming of the embodiment of the present invention seeks objective Generalization bounds algorithm flow chart.
Fig. 2, Fig. 3 and Fig. 4 are respectively that the Wall Street region zero load taxi of the embodiment of the present invention seeks objective recommendation results table, emperor
Back zone shopping center region zero load taxi seeks objective recommendation results table, Brooklyn Lincoln's terraced fields park areas zero load taxi is sought
Objective recommendation results table.
Fig. 5 is that the unloaded taxi of the different zones of the embodiment of the present invention seeks objective recommendation results spatial distribution map.
Fig. 6 is that hot spot region flow characteristics figure is recommended in the period variation of the embodiment of the present invention.
Fig. 7 is Wenzhou GPS data from taxi basic condition table of the embodiment of the present invention.
Fig. 8 is the volume of the flow of passengers timing distribution figure of the embodiment of the present invention.
Fig. 9 is the benefit timing distribution figure of the embodiment of the present invention.
Figure 10 is that the vehicle of the embodiment of the present invention seeks visitor's recommendation Efficiency Comparison table.
Specific embodiment
Embodiments of the present invention is further illustrated with reference to the accompanying drawing:
In the embodiment of the present invention, as shown in Figure 1, this algorithm, by carrying out slice division to the time, the web area that satisfies the need carries out
Gridding divides, and the volume of the flow of passengers distribution, benefit distribution and the carrying probability of success of arbitrary mess arbitrary period is calculated, in conjunction with taxi
The time cost for going to target area establishes Optimized model using high benefit, supply-demand structure as multiple target, acquires and most preferably seeks visitor and push away
It recommends and a little forms the taxi high benefit of complete set and seek objective Generalization bounds.
This algorithm specifically includes the following steps:
Step 1) data prediction: the historical data of taxi, including running data and wealth are obtained from GPS big data
Business data, according to taxi mark by driving trace and corresponding financial information phase in the running data and financial data
Matching；
Step 2) region division: grid dividing is carried out to city district, grid dividing can be there are two types of method, first is that by single
Position distance is divided, and unit distance is set as 1km*1km；Another kind is divided by n*m, both feasible, but the latter is more
Add flexibly.Each grid of division is denoted as (i, j), remembers xij, yij, i=1,2 ... m, j=1,2 ... n are (i, j) a grid
The longitude and latitude of regional center point, xmin,ymin,xmax,ymaxIndicate New York administrative region minimum latitude, longitude and maximum longitude and latitude, it can
The size for obtaining each space lattice is Δ x* Δ y,
(i, j) a regional center point longitude and latitude are as follows:
Grid distance determines starting point to cost the time spent in recommending point, is to promote high benefit to seek objective tactful precision
Key index.Grid distance is defined as the distance of two regional centers by we because Ordinary Rd be all south-north direction or
Person is east-west, so the calculating of the distance can be approximately to prolong longitudinal range difference plus prolonging latitude direction range difference.Then
(i, j) a grid can be obtained to (s, k) a grid distance is defined as:
dijsk=| xij-xsk|Ux+|yij-ysk|Uy
Wherein, UxIt indicates in the corresponding physical length of the grid once weft, UyIt indicates corresponding in the grid once warp
Physical length.Arbitrarily latitude was once the same through line length, and once the length of weft was related to the longitude θ of administrative region,
And there is Ux=Uycos(θ)。
Step 3) establishes multiple objective function: for taxi driver's individual, unloaded taxi seeks objective target to improve list
Position time income and quantity of received orders, wherein performance indicator is main target, and transport power index is by-end, and unloaded taxi is sought
Objective strategy is converted into multi-objective problem, and multiple objective function is characterized as below:
Wherein, (i, j) is the net region that taxi is currently located, and (s, k) is the net region that taxi is gone to, Aij
(t) indicate that the total amount recorded by bus, N occur in a net region (i, j) for the t periodsk(t) indicate that t period (s, k) is a
The total quantity by bus occurred in net region, TijskIt indicates to go to a net region (s, k) from a net region (i, j)
Running time, Tsk(t) indicate that the average running time recorded by bus, t occur in a net region (s, k) for the t periodijskTable
Show that the t period from a net region (i, j), reaches the period of a net region (s, k)；
Step 4) corrects multiple objective function: as above-mentioned, it is possible that taxi is recommended the serious superfluous region of transport power,
It causes taxi to flock together phenomenon, therefore we introduce the average latency TA that unloaded taxi obtains visitor, remembers TAskIt (t) is the t period
The objective time is obtained in the net region (s, k), is determined by the supply and demand ratio in the region.Is defined as:
1) it is lined up mode of getting on the bus (fixation such as airport is got on the bus a little), a newly-increased empty wagons obtains the objective waiting time
2) mode of getting on the bus at random (on-demand stop etc.), a newly-increased empty wagons obtain the objective waiting time
Wherein,
Nsk(t): for t period being averaged to get on the bus and record number in (s, k) grid
NEsk(t): recording number in the average empty wagons of (s, k) grid for the t period (if without empty wagons data, can use note of getting off
It is quasi- to record digital-to-analogue).
Multiple objective function is modified
And multiple objective function is constrained using constraint condition, comprising:
1) operating range constrains:
2) running time constrains:
vijIndicate the average speed for the record by bus that the t period occurs in a net region (i, j)；
3) period constraint is reached:
Due to taxi in the t period from a net region (i, j), spend TijskHour reaches (s, k) a net
Lattice region, period when arrival are tijsk, therefore have
tijsk=mod ([t+Tijsk],24)
Wherein, [x] expression is rounded downwards x, and mod (M, N) refers to M to the remainder of N；
It is excellent to have obtained the personal multiple target for seeking objective strategy under above-mentioned constraint condition for comprehensive benefit target and transport power target
Change model:
Finally, the recommendation area grid that can take into account performance indicator and transport power index for taxi driver can be obtained
Coordinate.
Verify below with reference to the taxi car data of New York to model: analysis taxi GPS big data is ridden
The longitude and latitude range of record is that [- 74.03,40.57] arrive [- 73.76,40.89], i.e., area is 16.75 miles × 22.09 miles
Rectangle, then city space is divided into the grid of 20*20, i.e., 0.84 mile × 1.10 miles of region, by from south to north, from
West is successively numbered to east.Representational 3 regions are chosen, crowded, traffic congestion Wall Street is respectively as follows:,
(9,2) grid；The queen area shopping center of medium, the single objective high efficiency of the volume of the flow of passengers, in (11,12) grid；The cloth of volume of the flow of passengers rareness
Luke Lin Linken terraced fields park, in (7,8) grid.Model solution obtains following high benefit and seeks objective recommendation results, such as Fig. 2, Fig. 3 and
Shown in Fig. 4, the intensive Wall Street region of the volume of the flow of passengers, unit time transport power is strong, it is sufficient to lost revenue caused by traffic congestion is made up,
Therefore it seeks visitor's scheduling progress short distance substantially centered on itself grid and seeks visitor；The medium queen area shopping center of the volume of the flow of passengers is in passenger flow
The stable period seeks visitor in itself and periphery grid, and in the period of passenger flow decline, such as at night 24 points are arrived 5:00 AM, then pushes away
Recommend two airports for going to high benefit；Brooklyn Lincoln's terraced fields park of volume of the flow of passengers rareness, in addition to peak period is in neighboring area
Visitor is sought, other times push away 10 substantially and recommend JFK Airport.
Objective scheduling result is sought in order to more intuitively show, objective hot spot destination distribution situation is sought in Fig. 5 displaying.With " unloaded just
Seek visitor to business district " intuition it is different, to recommend the highest region of index be two airports, followed by closely La Guardia airport
Recreation ground and Brooklyn Bridge neighboring area, the stream of people recommend temperature and periphery substantially to maintain an equal level in Manhattan region the most intensive.
It is two airports, followed by closely the recreation ground on La Guardia airport and Brooklyn Bridge neighboring area, the stream of people are the most intensive
Manhattan region recommend temperature and periphery substantially to maintain an equal level.
Further temporally slice can obtain the recommendation hot spot feature of any unit time, such as Fig. 6, can reflect with
The flow characteristics of hot spot region is recommended in period variation, if 6:00 AM is in the comprehensive effect of Manhattan and other administrative area junctions
Benefit is highest, and can then obtain highest expected benefit toward JFK Airport region at 11 o'clock of morning, has above-mentioned chart to believe
Breath support, blindly seek objective problem has obtained very good solution to taxi driver.
It is respectively present in two databases in taxi running data and financial data, and information is not to exactly match
In the case of, the present invention is as follows to process of data preprocessing with Wenzhou City's taxi data instance:
Step 1) taxi driving trace sample decimation:
Such as Fig. 7, the frequency acquisition of Wenzhou taxi running data is 15 seconds, and a day data amount is up to 20,000,000 records,
Under the premise of guaranteeing trace information integrality, seminar carries out double sampling, approximate structure with 10 minutes frequencies, to source data
Driving trace of hiring a car is produced, calculating speed is promoted.It comprises the concrete steps that:
(1) trace information is ranked up according to identification of the vehicle, current time；
(2) trace information of certain taxi is successively taken, is judged and sampling instant point (5,15,25,35,45,55 minutes)
Gap whether become larger, if becoming larger a upper trace information be the record to be used；
(3) (2) are repeated, until terminating.
Step 2) integrates financial data, constructs complete trip of taxi record:
(1) from taxi trace information, the record that passenger carrying status changes is extracted, generates Status Change file；
(2) from Status Change file, two records that are connected are taken, state is successively " carrying ", " empty wagons ", constitutes 1 row
Cheng Jilu, and taxi travel information is generated, including time of getting on or off the bus, longitude and latitude of getting on or off the bus；
(3) travel information is matched with financial information, and integration generates complete trip record.
In integration process, since taxi running data and financial data are from two sets of different systems, exist each other
Larger difference not can be carried out simple matching, such as:
1. there are mismatch cases for taxi mark, on inspection, the taxi mark of both data about 5% is mismatched, former
Because being mainly mistake inputs, how defeated, leakage is defeated etc.；
2. time precision of getting on or off the bus is inconsistent, the chronomere of running data is " second ", and the chronomere of financial data is
" dividing "；
3. the time of financial data acquires from terminal device, otherness is very big；The time unification of running data is by server
It generates, therefore the two gap is larger.
Using the two, there are certain relevances, such as:
1. in running data and financial data, the running time of same time stroke is of substantially equal, i.e., respectively get on or off the bus the time
Difference is approximately equal；
2. the time offset with taxi is essentially identical, i.e. the identical running data of taxi mark and financial data
There are essentially identical time delays；
3. there are correlations with " mileage travelled " in financial data for the longitude and latitude distance (i.e. linear distance) of running data
Based on above signature analysis, the matching scheme of use is as follows:
(1) it determines matching range, only data of the pick-up time difference in 1 hour is matched；
(2) Analytic Hierarchy Process Model is used, is selected from financial data and running data record (marking highest the most matched
Record) integrated, in AHP model, we make running time, shift time and mileage travelled and longitude and latitude correlation
For 3 criterion of rule layer, it is named as c1, c2, c3, judgment matrix
GPS data after pretreatment is finished does the timing diagram of the volume of the flow of passengers and benefit according to space-time analysis method respectively,
As shown in Figure 8 and Figure 9, benefit is bottomed out in two periods of morning peak and evening peak, point with New York trip space-time diagram
Analysis result is the same, and benefit and traffic condition have very big relationship, while 0 to 5 point of Wenzhou morning adds 20% fees policy
And the major reason of the period benefit high position is shown in figure.
GPS data is counted, vehicle most effective in one day is chosen and makees model validation test, test vehicle is real
The driving Efficiency Comparison specifying information that the driving benefit and model on border are recommended is recorded in Figure 10, by model validation is defined as:
Wherein, A2(t) value indicates to recommend expected benefit obtained, A1(t) value expression does not receive to recommend (real
Border traveling) benefit obtained, the benefit enhancing rate on the day of can calculating the vehicle is 127.18%.
Every technical staff's notice: of the invention although the present invention is described according to above-mentioned specific embodiment
Invention thought be not limited in the invention, any repacking with inventive concept will all be included in this patent protection of the patent right
In range.
Claims (3)
1. a kind of taxi high benefit based on GPS big data seeks objective multiple objective programming algorithm, which is characterized in that including following step
It is rapid:
Step 1) data prediction: the historical data of taxi, including running data and financial number are obtained from GPS big data
According to according to taxi mark by driving trace and corresponding financial information phase in the running data and financial data
Match；
Step 2) region division: carrying out grid dividing to city district, each grid be denoted as (i, j), i=1,2 ... m, j=1,
2 ... n；
Each taxi is completed the running data after matching and financial data and the grid phase after division by step 3)
Total record number N by bus that (i, j) a grid within the t period occurs is extracted in matchingij(t) and occur by bus record total gold
Volume Aij(t), multiple objective function is established
Wherein, (i, j) is the net region that taxi is currently located, and (s, k) is the net region that taxi is gone to, NskIt (t) is t
The total quantity by bus occurred in period (s, k) a net region, TijskTo go to (s, k) from a net region (i, j)
The running time of a net region, Tsk(t) the average traveling recorded by bus occurs in a net region (s, k) for the t period
Time, tijskIt is the t period from a net region (i, j), reaches the period of a net region (s, k)；
Step 4) corrects multiple objective function: taxi in GPS big data got on the bus into record and empty wagons record matches with each grid,
Obtain the average latency TA that unloaded taxi in each grid obtains visitor, i.e. empty wagons obtaining in t period, the net region (s, k)
The objective time,
It obtains objective average latency TA by described and introduces and correct multiple objective function
Step 5) is obtained according to revised multiple objective function recommends mesh coordinate (s, k).
2. the taxi high benefit according to claim 1 based on GPS big data seeks objective multiple objective programming algorithm, feature
Be: the average latency described in step 4) includes:
1) it is lined up mode of getting on the bus (fixation such as airport is got on the bus a little), a newly-increased empty wagons obtains the objective waiting time
2) mode of getting on the bus at random (on-demand stop etc.), a newly-increased empty wagons obtain the objective waiting time
Wherein,
Nsk(t): for t period being averaged to get on the bus and record number in (s, k) grid
NEsk(t): recording number in the average empty wagons of (s, k) grid for the t period (if without empty wagons data, can record number with getting off
Simulation).
3. the taxi high benefit according to claim 2 based on GPS big data seeks objective multiple objective programming algorithm, feature
Be: the step 4) further includes introducing constraint condition to carry out constraint to multiple objective function, and the constraint includes:
1) operating range constrains:
Wherein, dijskFor (i, j) a grid to (s, k) a grid distance, UxIt indicates in the corresponding reality of the grid once weft
Border length, UyIndicate in the corresponding physical length of the grid once warp, any latitude it is once the same through line length, and once
The length of weft is related to the longitude θ of administrative region, and has Ux=Uycos(θ)；xij, yijFor (i, j) a regional center point
Longitude and latitude, xsk, yskFor (s, k) a regional center point longitude and latitude；
2) running time constrains:
Wherein, vijIndicate the average speed for the record by bus that the t period occurs in a net region (i, j)；
3) it reaches period constraint: setting out and hire a car in the t period from a net region (i, j), spend TijskHour reach the
A net region (s, k), period when arrival are tijsk, then
tijsk=mod ([t+Tijsk],24)
Wherein, [x] expression is rounded downwards x, and mod (M, N) refers to M to the remainder of N；
The multiple objective function obtains Model for Multi-Objective Optimization after the constraint:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910237145.6A CN109886508A (en) | 2019-03-27 | 2019-03-27 | Taxi high benefit based on GPS big data seeks objective multiple objective programming algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910237145.6A CN109886508A (en) | 2019-03-27 | 2019-03-27 | Taxi high benefit based on GPS big data seeks objective multiple objective programming algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109886508A true CN109886508A (en) | 2019-06-14 |
Family
ID=66934645
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910237145.6A Pending CN109886508A (en) | 2019-03-27 | 2019-03-27 | Taxi high benefit based on GPS big data seeks objective multiple objective programming algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109886508A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110490393A (en) * | 2019-09-24 | 2019-11-22 | 湖南科技大学 | Objective route planning method, system and medium are sought in conjunction with the taxi of experience and direction |
| CN111127879A (en) * | 2019-12-09 | 2020-05-08 | 湖南大学 | Urban traffic flow prediction method based on generative countermeasure network |
| CN111882092A (en) * | 2020-06-16 | 2020-11-03 | 广东工业大学 | Taxi vehicle searching method suitable for shared trip |
| JP2021064180A (en) * | 2019-10-15 | 2021-04-22 | 株式会社Ｍｏｂｉｌｉｔｙ Ｔｅｃｈｎｏｌｏｇｉｅｓ | Charge determination device, charge determination system, and charge determination method |
| CN115620525A (en) * | 2022-12-16 | 2023-01-17 | 中国民用航空总局第二研究所 | Short-time traffic passenger demand prediction method based on time-varying dynamic Bayesian network |
| WO2024054153A1 (en) * | 2022-09-05 | 2024-03-14 | Grabtaxi Holdings Pte. Ltd. | Method and system for estimating a time of arrival of a driver at a location |
-
2019
- 2019-03-27 CN CN201910237145.6A patent/CN109886508A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110490393A (en) * | 2019-09-24 | 2019-11-22 | 湖南科技大学 | Objective route planning method, system and medium are sought in conjunction with the taxi of experience and direction |
| CN110490393B (en) * | 2019-09-24 | 2022-05-31 | 湖南科技大学 | Taxi passenger-searching route planning method, system and medium combining experience and direction |
| JP2021064180A (en) * | 2019-10-15 | 2021-04-22 | 株式会社Ｍｏｂｉｌｉｔｙ Ｔｅｃｈｎｏｌｏｇｉｅｓ | Charge determination device, charge determination system, and charge determination method |
| JP7337644B2 (en) | 2019-10-15 | 2023-09-04 | Ｇｏ株式会社 | Fee determination device, fee determination system and fee determination method |
| CN111127879A (en) * | 2019-12-09 | 2020-05-08 | 湖南大学 | Urban traffic flow prediction method based on generative countermeasure network |
| CN111127879B (en) * | 2019-12-09 | 2021-09-07 | 湖南大学 | Urban traffic flow prediction method based on generative countermeasure network |
| CN111882092A (en) * | 2020-06-16 | 2020-11-03 | 广东工业大学 | Taxi vehicle searching method suitable for shared trip |
| WO2024054153A1 (en) * | 2022-09-05 | 2024-03-14 | Grabtaxi Holdings Pte. Ltd. | Method and system for estimating a time of arrival of a driver at a location |
| CN115620525A (en) * | 2022-12-16 | 2023-01-17 | 中国民用航空总局第二研究所 | Short-time traffic passenger demand prediction method based on time-varying dynamic Bayesian network |
| CN115620525B (en) * | 2022-12-16 | 2023-03-10 | 中国民用航空总局第二研究所 | Short-time traffic passenger demand prediction method based on time-varying dynamic Bayesian network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109886508A (en) | Taxi high benefit based on GPS big data seeks objective multiple objective programming algorithm | |
| Zhu et al. | Understanding spatio-temporal heterogeneity of bike-sharing and scooter-sharing mobility | |
| Powell et al. | Towards reducing taxicab cruising time using spatio-temporal profitability maps | |
| Ma et al. | T-share: A large-scale dynamic taxi ridesharing service | |
| Tu et al. | Optimizing the locations of electric taxi charging stations: A spatial–temporal demand coverage approach | |
| Zhang et al. | Sensing the pulse of urban refueling behavior | |
| CN108053240B (en) | Method and device for generating vehicle-mounted advertisement delivery bus route scheme | |
| CN110033177B (en) | Taxi high-benefit intelligent collective dispatching method based on GPS big data | |
| CN103680127A (en) | A method for calculating signal lamp control road intersection delays through the utilization of low sampling rate floating vehicle data | |
| Foissaud et al. | Free-floating e-scooter usage in urban areas: A spatiotemporal analysis | |
| CN112749842B (en) | Escort path planning method and device | |
| Qu et al. | Location optimization for urban taxi stands based on taxi GPS trajectory big data | |
| CN112801552A (en) | Traffic big data mining and intelligent analysis-based network appointment and cruise supervision method | |
| Bursa et al. | Travel behavior on vacation: transport mode choice of tourists at destinations | |
| CN110458530A (en) | A kind of tourism plateform system and terminal towards tourist | |
| CN110223514B (en) | Urban traffic running state analysis method and device and electronic equipment | |
| Chiariotti et al. | Bike sharing as a key smart city service: State of the art and future developments | |
| Wamburu et al. | Ride substitution using electric bike sharing: feasibility, cost, and carbon analysis | |
| Liqun et al. | Research on taxi drivers' passenger hotspot selecting patterns based on GPS data: A case study in Wuhan | |
| Cui et al. | Usage demand forecast and quantity recommendation for urban shared bicycles | |
| Phiboonbanakit et al. | How does taxi driver behavior impact their profit? Discerning the real driving from large scale GPS traces | |
| CN109696174A (en) | Weather indicating means, device and the equipment of stroke | |
| Risimati et al. | Exploring the applicability of location based services to determine the state routes transport networks integratedness in the City of Johannesburg | |
| Zhou et al. | The measurement method of spatiotemporal accessibility of electric vehicle charging stations in the dynamic time-dependent urban environment | |
| CN110046752B (en) | Multi-level grid division method for complex geographic environment |
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: 20190614 |
|
| RJ01 | Rejection of invention patent application after publication |