CN106022577A - On-line balanced scheduling for passenger vehicle - Google Patents
On-line balanced scheduling for passenger vehicle Download PDFInfo
- Publication number
- CN106022577A CN106022577A CN201610312143.5A CN201610312143A CN106022577A CN 106022577 A CN106022577 A CN 106022577A CN 201610312143 A CN201610312143 A CN 201610312143A CN 106022577 A CN106022577 A CN 106022577A
- Authority
- CN
- China
- Prior art keywords
- passenger carrying
- carrying vehicle
- user
- taxi
- passenger
- 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 claims abstract description 76
- 238000009826 distribution Methods 0.000 claims abstract description 39
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 238000011156 evaluation Methods 0.000 claims abstract description 18
- 230000001105 regulatory effect Effects 0.000 claims description 7
- 238000010586 diagram Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 abstract description 5
- 230000000694 effects Effects 0.000 abstract description 2
- 238000009825 accumulation Methods 0.000 description 8
- 230000029305 taxis Effects 0.000 description 6
- 238000007781 pre-processing Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000009412 basement excavation Methods 0.000 description 2
- 230000033228 biological regulation Effects 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000005303 weighing Methods 0.000 description 2
- 230000006583 body weight regulation Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000005304 joining Methods 0.000 description 1
- 230000033001 locomotion Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000004080 punching Methods 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses an on-line balanced scheduling for a passenger vehicle, so that an effect of comprehensive consideration of unbalance of incomes of the passenger vehicles and the waiting time efficiency is realized. The method comprises: pretreatment is carried out on a city map and historical passenger vehicle track data; a range is reduced by using a range thinning algorithm and request predistribution is carried out by using a passenger vehicle income difference and a user waiting time evaluation function; conflict checking is carried out on a pre-distribution result; if a conflict exists, processing is carried out; and a balancing distribution result is scheduled and distributed to a request user. With the range thinning algorithm, the time spent for searching a passenger vehicle is reduced. Because an adjusting factor alpha is arranged in the evaluation function, importance adjustment between the user waiting time efficiency and the income difference is realized. Because an experiment is carried out on a real data set and comparison between an existing method and a global searching method is carried out, the result demonstrates that the income differences between passenger vehicles can be reduced substantially, the time efficiency is guaranteed, and balanced scheduling for passenger vehicles is realized. The method is used for balanced scheduling of passenger vehicles.
Description
Technical field
The invention belongs to Computer Applied Technology field, relate generally to data mining and passenger carrying vehicle distribution, specifically
Ground says it is the online equalization scheduling method of a kind of passenger carrying vehicle.Carrying for taxi with special train of calling a taxi is dispatched and is divided
Join.
Background technology
Along with the development of technology of Internet of things, GPS device and various sensor are widely used the life of people
In.Tachymeter such as taxi and the alignment system of smart mobile phone, navigation system and road etc..In a large number
Motion track information can be collected into, and the most substantial amounts of information in these track datas, by this
The a little analysis of data, excavations, can use the information obtained in real life, thus promote quality of life.
Such as move the rational traffic route of design data, and the city calculating etc. risen recently by the history of people
Deng.But owing to data are the hugest and value density is low, extract from these mass datas the most efficiently
Useful information, is to have actual application prospect and construction value.
By excavating the distribution that passenger carrying vehicle is carried out reasonably, equalizes by passenger carrying vehicle historical track information, from
And reducing the handling capacity of passengers difference between passenger carrying vehicle is a significant problem, therefore, it may have researching value.
As occurred in that many taxi drivers in current social because the reason of very few income creates miscellaneous
Problem, and by reducing the handling capacity of passengers difference between them so that their income becomes stable, thus reduce
The generation of these problems.
In prior art, the dispatching distribution for passenger carrying vehicle substantially can be divided into two classes, and the first kind is directly to count
Calculate the Euclidean distance of all passenger carrying vehicles and client, then select nearest passenger carrying vehicle to distribute to client；The
Two classes are then the situations considering share-car, under conditions of multiple passengers meet and take advantage of altogether, distribute a carrying for it
Vehicle；But these all do not have to consider from the situation of multiple passenger carrying vehicles, and the first kind is with time efficiency as target,
Equations of The Second Kind is then with the handling capacity of passengers maximizing a passenger carrying vehicle as target.These all do not account for multiple carrying
During vehicle, how to reduce their handling capacity of passengers difference.Such as, in actual life, each taxi owner is each
Month will hand over an expense to taxi company, and the income of some car owner may only many than the expense submitted one
Point, at this moment they can only increase the working time and earn and take more money, and even some car owner may change other
Work, at this moment in order to improve the stability of taxi work, reduces their operating pressure, it is necessary to reduce them
The difference of income, but simple it is intended merely to so that minimum this of income difference can reduce the experience of client, such as
Waiting time is long, so needing to do a power in the income difference of taxi with the waiting time efficiency of client
Weighing apparatus.In a word, prior art is simply from Customer waiting time efficiency or the handling capacity of passengers of one passenger carrying vehicle of maximization
Single from the standpoint of, the technical problem to many passenger carrying vehicle balance dispatchings can not be solved.
Summary of the invention
Present invention aims to prior art and lack the problem equalizing consideration on the whole, propose a kind of equal
The online equalization scheduling method of the passenger carrying vehicle of weighing apparatus passenger carrying vehicle amount difference and period of reservation of number.
Technical solution
The present invention is the online equalization scheduling method of a kind of passenger carrying vehicle, it is characterised in that include control centre
Stand, user, waits to dispatch buses, and connects station, control centre, user, on-line scheduling net to be dispatched buses
Network platform, concrete scheduling process comprises the following steps that
Step 1, to city map and passenger carrying vehicle historical trajectory data pretreatment, is extracted from diagram data truly
One sub-map datum in area, first counts the passenger carrying vehicle historical trajectory data of this area, further according to history rail
Mark data statistics goes out passenger carrying vehicle average time through every section；
The online balance dispatching of step 2 passenger carrying vehicle starts, and user sends request, and station, control centre receives request；
The predistribution of the online balance dispatching of step 3 passenger carrying vehicle, station, control centre according to the request of each user,
The scope thinning algorithm of utilization reduces the scope of search passenger carrying vehicle, and when utilizing passenger carrying vehicle income to wait with client
Between evaluation function each passenger carrying vehicle in the range of this is evaluated, the carrying minimum by obtaining evaluation function
Vehicle is fixed tentatively as distributing to dispatching buses of this request user；
Step 4, for the allocation result produced in pre-allocation process, carries out conflict and checks, check for many
Individual user request is assigned same passenger carrying vehicle, and checks for a user and be assigned many
Passenger carrying vehicle, if there is no collision problem, utilizes produce in pre-allocation process to equalize passenger carrying vehicle income
With the allocation result of period of reservation of number, ask dispatching distribution passenger carrying vehicle for each user, if there is conflict
Problem, redirects execution step 5；
Step 5, if there is collision problem, carries out clash handle, and station, control centre is according to clash handle principle first
Solving collision problem, then passenger carrying vehicle, and comprehensive predistribution are redistributed in the user's request for clashing
Journey does not has the allocation result of conflict, asks dispatching distribution passenger carrying vehicle for each user.
The present invention by being analyzed excavation to historical trajectory data, under the efficiency ensureing Customer waiting time,
The passenger carrying vehicle in city is carried out equilibrium assignment, thus reduces the income difference between passenger carrying vehicle, so this
Bright one side, can guarantee that the waiting time of user is short, on the other hand greatly reduces the receipts between passenger carrying vehicle
Enter difference.
Advantage and good effect
In the present invention, the online equalization scheduling method of passenger carrying vehicle has the advantage that
(1) present invention is by excavating true passenger carrying vehicle historical trajectory data, and simulates on true map
Experiment, meets reality scene.
(2) present invention devises scope thinning algorithm, compare global search, be significantly reduced searching passenger vehicle
Time, and the result consistent with global search can be obtained.
(3) method in the present invention is compared with prior art, the overall thinking distribution to passenger carrying vehicle,
And equalized passenger carrying vehicle income difference and period of reservation of number efficiency.
(4) method in the present invention is provided with a regulatory factor α, can be in period of reservation of number efficiency and load
Importance between passenger vehicle income difference is adjusted.
Accompanying drawing explanation
Fig. 1 is the flow chart of passenger carrying vehicle distribution portion real-time on line；
Fig. 2 is the flow chart of city map and the preprocessing part of passenger carrying vehicle historical trajectory data；
Fig. 3 is the experimental result picture of standard deviation of returns and distribution number of times relation；
Fig. 4 is the experimental result picture of period of reservation of number and distribution number of times relation；
Fig. 5 is the experimental result picture of distribution time and distribution number of times relation；
Fig. 6 is the experimental result picture of standard deviation of returns and α relation；
Fig. 7 is the experimental result picture of period of reservation of number and α relation.
Detailed description of the invention
Below in conjunction with the accompanying drawings with embodiment to the detailed description of the invention.
Embodiment 1.
The present invention is the online equalization scheduling method of a kind of passenger carrying vehicle, and method includes in scheduling during using
Center station, user, waits to dispatch buses, and connects station, control centre, user, on-line scheduling to be dispatched buses
The network platform, sees Fig. 1, and concrete passenger carrying vehicle online balance dispatching process comprises the following steps that
Step 1 to city map and passenger carrying vehicle historical trajectory data pretreatment, passenger carrying vehicle include taxi or
Call a taxi special train, from the diagram data truly in San Francisco, extract a sub-map datum in area, first count this ground
The passenger carrying vehicle historical trajectory data in district, the historical trajectory data further according to San Francisco counts passenger carrying vehicle process
The average time in every section.
The online balance dispatching of step 2 passenger carrying vehicle starts, and user sends request, and station, control centre receives request.
The predistribution of the online balance dispatching of step 3 passenger carrying vehicle, station, control centre according to the request of each user,
The scope thinning algorithm of utilization reduces the scope of search passenger carrying vehicle, and when utilizing passenger carrying vehicle income to wait with client
Between evaluation function each passenger carrying vehicle in the range of this is evaluated, the carrying minimum by obtaining evaluation function
Vehicle is fixed tentatively as distributing to dispatching buses of this request user.
Step 4, for the allocation result produced in pre-allocation process, carries out conflict and checks, check for many
Individual user request is assigned same passenger carrying vehicle, and checks for a user and be assigned many
Passenger carrying vehicle, if there is no collision problem, utilizes produce in pre-allocation process to equalize passenger carrying vehicle income
With the allocation result of period of reservation of number, ask dispatching distribution passenger carrying vehicle for each user, if there is conflict
Problem, redirects execution step 5.
Step 5, if there is collision problem, carries out clash handle, please because having multiple user at short notice
Asking, additionally the factor of area size may have multiple user to ask in the same time, please when there is multiple user
When asking, it is possible to ask to distribute same passenger carrying vehicle for multiple users, additionally there is likely to be a user please
Seeking allocated many passenger carrying vehicles, so needing these user's requests that distribution conflict occurs are processed, adjusting
Degree central station first solves collision problem according to clash handle principle, and then the user's request for clashing divides again
Join the allocation result not having conflict in passenger carrying vehicle, and comprehensive pre-allocation process, ask scheduling point for each user
Join passenger carrying vehicle.
The present invention considers on the whole from passenger carrying vehicle, it is proposed that a kind of equilibrium passenger carrying vehicle income difference and user etc.
Treat the online equalization scheduling method of passenger carrying vehicle of time, on the one hand, can guarantee that the waiting time of user is short, another
Aspect greatly reduces the income difference between passenger carrying vehicle.
Embodiment 2.
The online equalization scheduling method of passenger carrying vehicle is with embodiment 1, and in this example, passenger carrying vehicle is taxi, wherein
Pretreatment to city map Yu taxi historical trajectory data in step 1, sees Fig. 2, preprocessing process bag
Include:
1.1. read in city map data, in this example, the specifically city map data in San Francisco, and carry
Take a sub-map datum in this city.
1.2. obtain the city plan G=(E, I) of a connection according to sub-map datum, wherein E represents a road
Section, I represents cross point, section, and Ek=(Ii,Ij), i.e. one section is to be determined by two cross points.
1.3. read in the taxi historical trajectory data in San Francisco, and extract the taxi history rail of corresponding subregion
Mark data.
1.4. count this region according to historical trajectory data and have the section E of taxi processi, and count through
Section EiAll taxis, and time used by each taxiThen calculate through section Ei's
The average time of all taxisThe average time calculated is passed through by the present invention as taxi should
Section EiTime, wherein n represents through section EiThe quantity of taxi, j represents the numbering hired a car,
I represents the numbering in section.
1.5. count according to historical trajectory data and there is no the section E of taxi track dataj, arrange
V=30km/h, as the speed of each taxi, t, as through the time spent by this section, calculates
T=dist/v, wherein dist represents the length in this section, draws the taxi time through this section.
The present invention is by excavating true taxi historical trajectory data, and carries out simulation experiment on true map,
Meet reality scene.
Embodiment 3.
The online equalization scheduling method of passenger carrying vehicle is taxi with passenger carrying vehicle in embodiment 1-2, this example, its
In, the pre-allocation process of the online balance dispatching of taxi described in step 3 includes:
3.1. the Origin And Destination that station, control centre is asked according to user, calculates shortest path, and according to taxi
Car unit price calculates the profit that each user asks to produce.
3.2. the taxi taxi finding Euclidean distance the shortest is asked at station, control centre according to each useri,0, wherein
Subscript i represents that user asks numbering, and subscript 0 represents that the taxi asking Euclidean distance the shortest apart from this user is compiled
Number, and by single source single-point shortest path first structure taxii,0To the shortest time path of corresponding user, if should
The time in path is ti,0。
3.3. station, control centre is according to the request of each user, and the scope thinning algorithm of utilization reduces to search out hires a car
Scope.Specifically station, control centre is asked according to each user, and taxi obtained abovei,0With ti,0, profit
Radius is calculated with the operation of scope thinning algorithmReduce searching the most reasonable
The scope of taxi, it is assumed that be calculated for the first time and narrow down in the range of Di, wherein i=1, then DiBe with
User's request is the center of circle, and radius is the region of r, and wherein v represents the speed hired a car, and α represents a weight
Regulatory factor, is a real number, can be manually set according to demand in concrete operations, this example chooses α=1,
tProfit0Represent taxii,0Up to the present accumulative income, minTProfit represents (initial model in present scope
Enclose for whole region) the accumulative income of minimum in all taxis, above-mentioned accumulation income is along with taxi not
Disconnected being allocated can be gradually increased, and concrete increase form is, if certain taxi is assigned to user's request,
Then this taxi accumulation income (time initialized each taxi accumulation income is all 0) plus should
The profit that user produces, as the accumulation income that this taxi is new.
(3.3.a). determination range DiThe most whether there is the taxi that accumulative income is minTProfit, if it does,
Then stop the operation of scope thinning algorithm, scope D finally narrowed down to, and redirect execution (3.4)；
(3.3.b). determination range DiThe most whether there is the taxi that accumulative income is minTProfit, if do not deposited
, continuing to update minTProfit is scope DiThe accumulative income of minimum in interior taxi, and proceed model
Enclose thinning algorithm operation, make i=i+1, obtain new scope Di, redirect execution (3.3.a).
(3.4). utilize the taxi income evaluation function with Customer waiting time to each taxi in the range of this
It is evaluated, the taxi obtaining evaluation function minimum is fixed tentatively as distributing to dispatching buses of this request user.
Specifically station, control centre is according to evaluation functionIn computer capacity D often
The EVA value of individual taxi, fixes tentatively taxi minimum for EVA value and distributes to this user, wherein EVAjRepresent
The EVA value of jth taxi, tProfitjRepresent jth taxi accumulative income up to the present；
Δti,j=ti,j-ti,0It it is jth the taxi extra latency that causes i-th user；α is that a weight is adjusted
The joint factor, is used for arranging extra latency to EVAjThe impact of value, this example chooses α=1, andExplanation
Extra latency is to EVAjThe exponentially that affects of value rises, when extra latency is increasing, and EVAj
Value can increase quickly, this guarantees time efficiency, thus this evaluation function can equalize taxi income with
Period of reservation of number.
Method in the present invention compared with prior art, the overall thinking distribution to taxi, and equalizing
Taxi income difference and period of reservation of number efficiency, be also provided with a regulatory factor α in the present invention, when
Focus on be taxi income difference time, less α is set, and when emphasis be period of reservation of number efficiency time,
Bigger α is set, by arranging different α values, can take in taxi in period of reservation of number efficiency
It is adjusted between the importance of gap.
Embodiment 4.
The online equalization scheduling method of passenger carrying vehicle is taxi with passenger carrying vehicle in embodiment 1-3, this example, step
The clash handle process of the 5 online balance dispatchings of taxi includes:
5.1. clash handle principle 1, when multiple users request is assigned identical taxi, by profit
Big user asks preferentially to distribute to this taxi, because the EVA value of this taxi is minimum, illustrates compared with other
Taxi is optimum；If but multiple users are processed simultaneously, once produce conflict, unassigned taxi
The user of car is accomplished by redistributing taxi, and the inspection that the most once conflicts, because secondary punching may be produced
Prominent, the most repeatedly conflict, and this problem can be converted into and be preferably the big user's request of profit and distribute and hire a car,
Directly user is asked the profit size descending sort produced according to it, the most successively it is distributed, the most both accorded with
The treatment principle that syzygies is prominent, does not haves situation about repeatedly conflicting yet.
5.2. clash handle principle 2, when a user asks allocated many taxis, should distribute arrival should
The taxi that user's request time is the shortest, i.e. under conditions of EVA value is identical, time efficiency is preferential.
The present invention has carried out rational process to the collision problem produced in taxi pre-allocation process, eliminates point
During joining, multiple users are allocated same taxi, and same user asks allocated many taxis
Collision problem, create the allocation result finally more equalized.
Embodiment 5.
The online equalization scheduling method of passenger carrying vehicle is special train of calling a taxi with passenger carrying vehicle in embodiment 1-4, this example,
This example is more complete from one again, more details, and the present invention is explained by the more process for implementation again, tool
Body process includes:
One, the city map of the present invention walks with the concrete of the preprocessing part of special train historical trajectory data of calling a taxi
Suddenly it is expressed as follows:
(1). read in the city map data in San Francisco, and extract a sub-map datum in this city.
(2). obtain the city plan G=(E, I) of a connection according to sub-map datum, wherein E represents a road
Section, I represents cross point, a section, and Ek=(Ii,Ij), i.e. one section is to be determined by two cross points.
(3). read in the special train historical trajectory data of calling a taxi in San Francisco, and the special train of calling a taxi extracting corresponding subregion is gone through
History track data.
(4). counting this region according to historical trajectory data has the section E of special train process of calling a taxii, and count through
Cross section EiAll special trains of calling a taxi, and each calls a taxi the time used by special trainThen calculate through
Section EiAverage time of all special trains of calling a taxiAs special train of calling a taxi through this section Ei's
Time, wherein n represents through section EiThe quantity of special train of calling a taxi, j represents the numbering of special train of calling a taxi, i
Represent the numbering in section.
(5). the section E of special train track data of not calling a taxi is counted according to historical trajectory dataj, arrange
V=30km/h, as the speed of each special train of calling a taxi, t, as through the time spent by this section, calculates
T=dist/v, wherein dist represents the length in this section, draws the special train time through this section of calling a taxi.
Two, the concrete steps of special train distribution portion of calling a taxi real-time on the line of the present invention are expressed as follows:
(1). input user's request, station, control centre receives user's request.
(2). the Origin And Destination that station, control centre is asked according to user, calculate shortest path, and according to calling a taxi
Special train unit price calculates the profit that each user asks to produce.
(3). the special train taxi that calls a taxi finding Euclidean distance the shortest is asked at station, control centre according to each useri,0, wherein
Subscript i represents that user asks numbering, and subscript 0 represents asks the shortest special train of calling a taxi of Euclidean distance apart from this user
Numbering, and by single source single-point shortest path first structure taxii,0To the shortest time path of corresponding user, if
The time in this path is ti,0。
(4). station, control centre is asked according to each user, and taxi obtained abovei,0With ti,0, utilize scope thin
Change algorithm operating and calculate radiusReduce and find special train of the most rationally calling a taxi
Scope, it is assumed that be calculated for the first time and narrow down in the range of Di, wherein i=1, then DiIt is please with user
Asking as the center of circle, radius is the region of r, and wherein v represents the speed of special train of calling a taxi, and α represents a weight regulation
The factor, is a real number, can be manually set according to demand in concrete operations, this example chooses α=1, tProfit0
Represent taxii,0Up to the present accumulative income, minTProfit represents that in present scope, (initial range is whole
Individual region) the accumulative income of all minimums called a taxi in special train, above-mentioned accumulation income is continuous along with special train of calling a taxi
Allocated meeting is gradually increased, and concrete increase form is, if certain special train of calling a taxi is assigned to user's request,
Then this special train of calling a taxi accumulation income (time initialized each special train of calling a taxi accumulation income is all 0) add
The profit that this user upper produces, as the accumulation income that this special train of calling a taxi is new.
(4.a). determination range DiThe most whether there is the special train of calling a taxi that accumulative income is minTProfit, if it does,
Then stop the operation of scope thinning algorithm, scope D finally narrowed down to, and jump to (5)；
(4.b). determination range DiThe most whether there is the special train of calling a taxi that accumulative income is minTProfit, if do not deposited
, continuing to update minTProfit is scope DiThe accumulative income of the minimum in special train of inside calling a taxi, and proceed
Scope thinning algorithm operates, and makes i=i+1, obtains new scope Di, redirect execution (4.a).
(5). station, control centre is according to evaluation functionIn computer capacity D often
Call a taxi the EVA value of special train, distribute to this user, wherein EVA by tentative for minimum for EVA value special train of calling a taxij
Represent that jth is called a taxi the EVA value of special train, tProfitjRepresent that jth is called a taxi special train up to the present accumulative
Income；Δti,j=ti,j-ti,0It is that jth special train of calling a taxi causes the extra latency of i-th user；α is one
Individual weight regulatory factor, is used for arranging extra latency to EVAjThe impact of value, andIllustrate extra etc.
Treating that the exponentially that affects of EVA value is risen by the time, when extra latency is increasing, EVA value can increase
Add quickly, this guarantees time efficiency, so this evaluation function can equalize call a taxi special train income and user
Waiting time.
(6). station, control centre carries out conflict and checks the allocation result in pre-allocation process, and particular exam includes,
Check for multiple user request and same call a taxi special train is assigned, and check whether user's quilt
It is assigned with many special trains of calling a taxi, if there is no collision problem, redirects execution (7), if there is collision problem,
Carry out clash handle.
(6a). clash handle principle 1, when multiple users to have asked recommended identical call a taxi special train time, by profit
Maximum user asks preferentially to distribute to this special train of calling a taxi, because this is called a taxi, the EVA value of special train is minimum, explanation
Its relatively other special trains of calling a taxi are optimum；If but each user is processed simultaneously, once produce conflict, not
The user of allocated special train of calling a taxi is accomplished by redistributing a special train of calling a taxi, and redistribute carry out one again
Secondary conflict checks, because secondary conflict may be produced, the most repeatedly conflicts, and this problem can be converted into excellent
First ask distribution to be called a taxi special train for the big user of profit, directly user's request is dropped according to its profit size produced
Sequence sorts, and distributes it the most successively, now meets the treatment principle both conflicted, also do not have and repeatedly conflict
Situation.
(6b). clash handle principle 2, when user request exist multiple special train of calling a taxi have identical EVA and
Minimum, now answer preferential recommendation fastest to reach user request special train of calling a taxi, i.e. under conditions of EVA is identical,
In time efficiency preferentially.
(7). equilibrium call a taxi special train income and the user that station, control centre produces with clash handle according to predistribution
The allocation result of waiting time, asks distribution to be called a taxi special train for each user.
The method of the present invention is applicable not only to the on-line scheduling of taxi, the online tune of the special train that applies also for calling a taxi
Degree, by excavating special train historical trajectory data of truly calling a taxi, and carries out simulation experiment on true map, meets
Reality scene, compared with prior art, the overall thinking distribution to special train of calling a taxi, and consider simultaneously
The waiting time of user.
Embodiment 6.
The online equalization scheduling method of passenger carrying vehicle is with embodiment 1-5, below in conjunction with the accompanying drawings with embodiment to this
Bright it is described further.
Utilize the present invention to the urban road data in San Francisco and taxi historical trajectory data simulation experiment, carry out
Taxi equilibrium assignment, thus solve taxi income inequality and the problem of period of reservation of number efficiency, should
Data set from data hall (http://www.datatang.com/data/15935,
Http:// www.datatang.com/data/15731), collect the road, main cities that have recorded San Francisco respectively
Circuit-switched data, with the GPS in 500 taxis 25 days (on June 10,17 days to 2008 May in 2008)
Track data.Urban road information spinner will be by section, road section length, road junction and longitude and latitude composition, goes out
Track data of hiring a car mainly is made up of taxi ID, taxi state, time shaft and longitude and latitude.Whole road
Road network data include 174956 cross points and 223001 sections, whole history taxi track number altogether
More than 100 ten thousand records are comprised altogether according to collection.
Utilizing the present invention to complete taxi and carry out equilibrium assignment, workflow is shown in accompanying drawing 1 and accompanying drawing 2, concrete
Implement step as follows:
Data prediction part:
1) read road network data, store with node by section, and choose the sub regions in road network,
Obtaining a section is 28308, and road circuit node is the region of 19905.
2) read taxi track data, and calculate taxi through being somebody's turn to do according to the taxi recording gauge of selected areas
The time in Zhong Meitiao section, region.
Calculating section on line:
1) import the data that preprocessing process obtains, be simulated experiment, α=0.1 is set, at random in this district
Territory produces 500 taxis, and constantly randomly generates user's request.
2) station, control centre is asked according to each user, finds nearest taxi, and obtains a local
Short time path, then carries out scope thinning algorithm and reduces and search out the scope hired a car.
3) each taxi in the range of this is evaluated by station, control centre according to evaluation function, carries out pre-point
Join.
4) station, control centre carries out conflict inspection to the result of pre-allocation process, if there is no conflict, is directly
Each user asks dispatching distribution taxi, if there is conflict, performs 5.
5) station, control centre first solves conflict according to clash handle principle, and then the user for clashing asks weight
Newly distribute and hire a car, and comprehensive pre-allocation process does not has the allocation result of conflict, ask scheduling for each user
Distribute and hire a car.
The present invention is because having considered taxi income inequality and the problem of period of reservation of number efficiency, logical
Cross above step, complete the equilibrium assignment to taxi, then by nearest with directly recommendation for the method for the present invention
The method of taxi and the method for global search compare, the method the most directly recommending nearest taxi,
Pursue is time efficiency, and the method for global search is the feelings of method not range thinning algorithm of the present invention
Condition, directly at the taxi that gamut search is optimum, the method that in Fig. 3-Fig. 7, curve 1 represents the present invention,
Curve 2 represents the method directly recommending nearest taxi, and curve 3 represents the method for global search, Fig. 3
It is the experimental result picture of standard deviation of returns and distribution number of times relation, it can be seen from figure 3 that along with the increase of distribution number of times,
Directly recommend the method for nearest taxi that the standard deviation of returns between taxi can be made to become increasing, and this
The method of invention but ensure that standard deviation of returns is basicly stable, consistent with the result of the method for global search, Fig. 4
Being the experimental result picture of period of reservation of number and distribution number of times relation, as seen from Figure 4, the method for the present invention is flat
Little difference, general phase is only had with directly recommending the method for nearest taxi on the most each period of reservation of number
Differing from about 1 minute, Fig. 5 is the experimental result picture of distribution time and distribution number of times relation, from figure 5 it can be seen that this
The method of invention compare on the dispatching distribution time method of global search want short a lot because the method for the present invention
Employ scope thinning algorithm, reduce the scope hired a car that searches out, thus decrease the time of distribution, such as,
When distributing number of times and being 50, the time phase difference of average every sub-distribution 200ms, when distribution number of times gets more and more
Time, total distribution time (average every sub-distribution time is multiplied by distribution number of times) can be made to differ increasing.Figure
6 is standard deviation of returns and the experimental result picture of α relation and period of reservation of number and α relation with Fig. 7 respectively
Experimental result picture, from Fig. 6 and Fig. 7, when arranging different α values, the method for the present invention can produce
Different experimental results, when α more hour, no matter passenger carrying vehicle is taxi or for special train of calling a taxi, it takes in mark
Quasi-difference will be the least, but period of reservation of number will be more and more longer；When α=0, the most do not consider user etc.
Treat the time, now make taxi standard deviation of returns minimum, but the waiting time is the longest；When α is increasing
Time, standard deviation of returns also can be increasing, but period of reservation of number can be shorter and shorter.This explanation is when needing guarantor
Card taxi between take in difference hour, but for period of reservation of number again can not oversize time, therefore can select
Select less α；And when when being to wait for the time of emphasis, bigger α value can be set.The present invention is by repeatedly
Experiment repeatedly, under this experiment condition, it is achieved that equilibrium assignment, equalized taxi income difference and user
Waiting time efficiency, it is possible to the importance between taxi income gap and period of reservation of number efficiency is carried out
Regulation.
In brief, the invention discloses the online equalization scheduling method of a kind of passenger carrying vehicle, to passenger carrying vehicle
On the basis of the historical trajectory data of magnanimity carries out mining analysis, carry out reasonably, the passenger carrying vehicle of equilibrium distributes,
Thus reduce the income difference between passenger carrying vehicle.Compared with existing passenger carrying vehicle distribution method, the present invention from
One new from the point of view of the distribution of passenger carrying vehicle, not only consider time efficiency, but also consider passenger carrying vehicle
Between income difference, solve the problem that passenger carrying vehicle income inequality considers with waiting time efficiency.
First complicated calculating to city map and passenger carrying vehicle historical trajectory data pretreatment, is transferred to line by the present invention
Lower part, then stores result of calculation, in order to the real-time calculating section on Support Line；Refine by scope
Algorithm reduces the scope, by passenger carrying vehicle income difference with Customer waiting time evaluation function to request predistribution；Right
Predistribution result is done conflict and is checked have, make clash handle；Request user is given by equilibrium assignment result dispatching distribution.
Scope thinning algorithm in the present invention, decreases search passenger carrying vehicle time overhead.Evaluation function is arranged
Regulatory factor α, it is achieved that the regulation of importance between period of reservation of number efficiency and income difference.By truly
Data set on test, and enter with directly recommending the method for nearest passenger carrying vehicle and the method for global search
Row compares, and demonstrates the income difference utilizing the method for the present invention can be significantly reduced between passenger carrying vehicle, and
Also ensure that period of reservation of number efficiency, it is achieved that the balance dispatching to passenger carrying vehicle.
Claims (4)
1. the online equalization scheduling method of a passenger carrying vehicle, it is characterised in that include station, control centre,
User, passenger carrying vehicle to be dispatched, and connect station, control centre, and user, the online tune of passenger carrying vehicle to be dispatched
The degree network platform, concrete scheduling process comprises the following steps that
Step 1, to city map and passenger carrying vehicle historical trajectory data pretreatment, is extracted from diagram data truly
One sub-map datum in area, first counts the passenger carrying vehicle historical trajectory data of this area, further according to history rail
Mark data statistics goes out passenger carrying vehicle average time through every section；
The online balance dispatching of step 2 passenger carrying vehicle starts, and user sends request, and station, control centre receives request；
The predistribution of the online balance dispatching of step 3 passenger carrying vehicle, station, control centre according to the request of each user,
The scope thinning algorithm of utilization reduces the scope of search passenger carrying vehicle, and when utilizing passenger carrying vehicle income to wait with client
Between evaluation function each passenger carrying vehicle in the range of this is evaluated, the carrying minimum by obtaining evaluation function
Vehicle is fixed tentatively as distributing to dispatching buses of this request user；
Step 4, for the allocation result produced in pre-allocation process, carries out conflict and checks, check for many
Individual user request is assigned same passenger carrying vehicle, and checks for a user and be assigned many
Passenger carrying vehicle, if there is no collision problem, utilizes produce in pre-allocation process to equalize passenger carrying vehicle income
With the allocation result of period of reservation of number, ask dispatching distribution passenger carrying vehicle for each user, if there is conflict
Problem, redirects execution step 5；
Step 5, if there is conflict, carries out clash handle, and station, control centre first solves according to clash handle principle
Conflict, then the user's request for clashing is redistributed in passenger carrying vehicle, and comprehensive pre-allocation process and is not had
The allocation result of conflict, asks dispatching distribution passenger carrying vehicle for each user.
The online equalization scheduling method of passenger carrying vehicle the most according to claim 1, it is characterised in that step
The 1 pair of city map includes with the pretreatment of passenger carrying vehicle historical trajectory data:
(1.1). read in city map data, and extract a sub-map datum in this city；
(1.2). obtain the city plan G=(E, I) of a connection according to sub-map datum, wherein E represents a road
Section, I represents cross point, section, and Ek=(Ii,Ij), i.e. one section is to be determined by two cross points；
(1.3). read in passenger carrying vehicle historical trajectory data, and extract the passenger carrying vehicle historical track of corresponding subregion
Data；
(1.4). counting this region according to historical trajectory data has the section E of passenger carrying vehicle processi, and count
Through section EiAll passenger carrying vehicles, and time used by each passenger carrying vehicleThen calculate through
Cross section EiAverage time of all passenger carrying vehiclesAs passenger carrying vehicle through this section EiTime
Between, wherein n represents through section EiThe quantity of passenger carrying vehicle, j represents the numbering of passenger carrying vehicle, and i represents
The numbering in section；
(1.5). the section E not having passenger carrying vehicle track data is counted according to historical trajectory dataj, arrange
V=30km/h, as the speed of each passenger carrying vehicle, t, as through the time spent by this section, calculates
T=dist/v, wherein dist represents the length in this section, draws the passenger carrying vehicle time through this section.
The online equalization scheduling method of passenger carrying vehicle the most according to claim 1, it is characterised in that step
The pre-allocation process of the 3 online balance dispatchings of passenger carrying vehicle includes:
(3.1). the Origin And Destination that station, control centre is asked according to user, calculate shortest path, and according to load
Passenger vehicle unit price calculates the profit that each user asks to produce；
(3.2). the passenger carrying vehicle taxi finding Euclidean distance the shortest is asked at station, control centre according to each useri,0, its
Middle subscript i represents that user asks numbering, and subscript 0 represents asks the shortest passenger vehicle of Euclidean distance apart from this user
Numbering, and by single source single-point shortest path first structure taxii,0To the shortest time path of corresponding user,
If the time in this path is ti,0；
(3.3). station, control centre is asked according to each user, and taxi obtained abovei,0With ti,0, utilize scope
Thinning algorithm operation calculates radiusReduce the most reasonable passenger vehicle of searching
Scope, it is assumed that narrow down in the range of Di, wherein i=1, then DiIt is with user's request as the center of circle, half
Footpath is the region of r, and wherein v represents the speed of passenger carrying vehicle, and α represents a weight regulatory factor, is one
Real number, t Pr ofit0Represent taxii,0Up to the present accumulative income, min T Pr ofit represents institute in present scope
There is the accumulative income of the minimum in passenger carrying vehicle；
(3.3.a). determination range DiThe most whether there is the passenger carrying vehicle that accumulative income is min T Pr ofit, if deposited
, then stop the operation of scope thinning algorithm, scope D finally narrowed down to, and jump to (3.4)；
(3.3.b). determination range DiThe most whether there is the passenger carrying vehicle that accumulative income is min T Pr ofit, if not
Existing, continuing to update min T Pr ofit is scope DiThe accumulative income of minimum in interior passenger carrying vehicle, and continue into
Line range thinning algorithm operates, and makes i=i+1, obtains new scope Di, redirect execution (3.3.a)；
(3.4). station, control centre is according to evaluation functionIn computer capacity D
The EVA value of each passenger carrying vehicle, fixes tentatively passenger carrying vehicle minimum for EVA value and distributes to this user, wherein
EVAjRepresent the EVA value of jth passenger carrying vehicle, t Pr ofitjRepresent jth passenger carrying vehicle up to the present
Accumulative income；Δti,j=ti,j-ti,0It it is jth the passenger carrying vehicle extra latency that causes i-th user；α is
One weight regulatory factor, is used for arranging extra latency to EVAjThe impact of value, andIllustrate extra
Waiting time is to EVAjThe exponentially that affects of value rises, when extra latency is increasing, and EVAjValue meeting
Increase quickly.
The online equalization scheduling method of passenger carrying vehicle the most according to claim 1, it is characterised in that step
The clash handle process of the 5 online balance dispatchings of passenger carrying vehicle includes:
(5.1). clash handle principle 1, when multiple users passenger carrying vehicle that to have asked recommended identical, by profit
The user of profit maximum asks preferentially to distribute to this passenger carrying vehicle, because the EVA value of this passenger carrying vehicle is minimum, says
Bright relatively other passenger carrying vehicles are optimum；If but multiple users are processed simultaneously, once produce conflict, not
The user of allocated passenger carrying vehicle is accomplished by redistributing passenger carrying vehicle, and the inspection that the most once conflicts, because
Secondary conflict may be produced, the most repeatedly conflict, and this problem can be converted into and be preferably the user that profit is big
Request distribution passenger carrying vehicle, directly asks the profit size descending sort produced according to it, the most successively by user
It is distributed, had the most both met the treatment principle of conflict, and also do not had situation about repeatedly conflicting；
(5.2). clash handle principle 2, when a user asks allocated many passenger carrying vehicles, should be assigned to
Reaching the passenger carrying vehicle that this user's request time is the shortest, i.e. under conditions of EVA value is identical, time efficiency is preferential.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610312143.5A CN106022577A (en) | 2016-05-12 | 2016-05-12 | On-line balanced scheduling for passenger vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610312143.5A CN106022577A (en) | 2016-05-12 | 2016-05-12 | On-line balanced scheduling for passenger vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106022577A true CN106022577A (en) | 2016-10-12 |
Family
ID=57099334
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610312143.5A Pending CN106022577A (en) | 2016-05-12 | 2016-05-12 | On-line balanced scheduling for passenger vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106022577A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109523189A (en) * | 2018-11-29 | 2019-03-26 | 北京首汽智行科技有限公司 | A kind of vehicle dispatching method and system |
| CN109784659A (en) * | 2018-12-18 | 2019-05-21 | 东软集团股份有限公司 | Processing method, device, storage medium and the electronic equipment of service request |
| CN109859458A (en) * | 2019-01-17 | 2019-06-07 | 深圳市泰比特科技有限公司 | A kind of vehicle dispatching method and system based on vehicle big data |
| CN109997346A (en) * | 2017-06-23 | 2019-07-09 | 北京嘀嘀无限科技发展有限公司 | Service dispatch system and method based on user behavior |
| CN111723276A (en) * | 2019-03-19 | 2020-09-29 | 天津五八到家科技有限公司 | Task response system, method and device and electronic equipment |
| TWI720380B (en) * | 2017-12-15 | 2021-03-01 | 大陸商北京嘀嘀無限科技發展有限公司 | Systems and methods for optimizing an online on-demand service |
| CN113159348A (en) * | 2020-01-22 | 2021-07-23 | 丰田自动车株式会社 | Control device, system, non-transitory computer-readable medium, terminal device, and user assistance method |
| CN115457793A (en) * | 2022-08-31 | 2022-12-09 | 东风悦享科技有限公司 | Method and system for non-uniform departure of starting station in bus dispatching |
| CN113159348B (en) * | 2020-01-22 | 2024-05-31 | 丰田自动车株式会社 | User-assisted control device, system, non-transitory computer-readable medium, terminal device, and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855753A (en) * | 2011-06-30 | 2013-01-02 | 高德软件有限公司 | Method and platform for taxi scheduling based on real-time traffic |
| CN103177575A (en) * | 2013-03-07 | 2013-06-26 | 上海交通大学 | System and method for dynamically optimizing online dispatching of urban taxies |
| CN104134342A (en) * | 2013-05-04 | 2014-11-05 | 李艳友 | Intelligent analyzing and processing system and method for taxi dispatching |
-
2016
- 2016-05-12 CN CN201610312143.5A patent/CN106022577A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855753A (en) * | 2011-06-30 | 2013-01-02 | 高德软件有限公司 | Method and platform for taxi scheduling based on real-time traffic |
| CN103177575A (en) * | 2013-03-07 | 2013-06-26 | 上海交通大学 | System and method for dynamically optimizing online dispatching of urban taxies |
| CN104134342A (en) * | 2013-05-04 | 2014-11-05 | 李艳友 | Intelligent analyzing and processing system and method for taxi dispatching |
Non-Patent Citations (1)
| Title |
|---|
| 张和生 等: "利用GPS数据估计路段的平均行程时间", 《吉林大学学报（工学版）》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109997346A (en) * | 2017-06-23 | 2019-07-09 | 北京嘀嘀无限科技发展有限公司 | Service dispatch system and method based on user behavior |
| TWI720380B (en) * | 2017-12-15 | 2021-03-01 | 大陸商北京嘀嘀無限科技發展有限公司 | Systems and methods for optimizing an online on-demand service |
| US11455582B2 (en) | 2017-12-15 | 2022-09-27 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for optimizing an online on-demand service |
| CN109523189A (en) * | 2018-11-29 | 2019-03-26 | 北京首汽智行科技有限公司 | A kind of vehicle dispatching method and system |
| CN109523189B (en) * | 2018-11-29 | 2022-08-30 | 北京首汽智行科技有限公司 | Vehicle scheduling method and system |
| CN109784659A (en) * | 2018-12-18 | 2019-05-21 | 东软集团股份有限公司 | Processing method, device, storage medium and the electronic equipment of service request |
| CN109859458A (en) * | 2019-01-17 | 2019-06-07 | 深圳市泰比特科技有限公司 | A kind of vehicle dispatching method and system based on vehicle big data |
| CN109859458B (en) * | 2019-01-17 | 2020-06-30 | 深圳市泰比特科技有限公司 | Vehicle scheduling method and system based on vehicle big data |
| CN111723276A (en) * | 2019-03-19 | 2020-09-29 | 天津五八到家科技有限公司 | Task response system, method and device and electronic equipment |
| CN111723276B (en) * | 2019-03-19 | 2023-10-27 | 天津五八到家科技有限公司 | Task response system, method and device and electronic equipment |
| CN113159348A (en) * | 2020-01-22 | 2021-07-23 | 丰田自动车株式会社 | Control device, system, non-transitory computer-readable medium, terminal device, and user assistance method |
| CN113159348B (en) * | 2020-01-22 | 2024-05-31 | 丰田自动车株式会社 | User-assisted control device, system, non-transitory computer-readable medium, terminal device, and method |
| CN115457793A (en) * | 2022-08-31 | 2022-12-09 | 东风悦享科技有限公司 | Method and system for non-uniform departure of starting station in bus dispatching |
| CN115457793B (en) * | 2022-08-31 | 2023-06-23 | 东风悦享科技有限公司 | Method and system for non-uniform departure of origin stations in bus dispatching |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106022577A (en) | On-line balanced scheduling for passenger vehicle | |
| Alisoltani et al. | Can dynamic ride-sharing reduce traffic congestion? | |
| Chen et al. | Solving the first‐mile ridesharing problem using autonomous vehicles | |
| Bischoff et al. | City-wide shared taxis: A simulation study in Berlin | |
| Stocker et al. | Shared automated mobility: early exploration and potential impacts | |
| Barth et al. | Simulation model performance analysis of a multiple station shared vehicle system | |
| CN109087030B (en) | Method for realizing C2C universal mobile crowdsourcing, universal mobile crowdsourcing server and system | |
| Shen et al. | Dynamic ridesharing | |
| CN106875734B (en) | A kind of method and device of push parking route | |
| Guo et al. | Vehicle rebalancing with charging scheduling in one-way car-sharing systems | |
| CN111861086B (en) | Resource allocation method and system | |
| Celiński et al. | REAL TIME MODEL FOR PUBLIC TRANSPORTATION MANAGEMENT. | |
| Chen | Management of a shared, autonomous, electric vehicle fleet: vehicle choice, charging infrastructure & pricing strategies | |
| Costa et al. | A simulation-optimization model for analyzing a demand responsive transit system for last-mile transportation: A case study in São Paulo, Brazil | |
| Jakimavičius et al. | Assessing multiple criteria for rapid bus routes in the public transport system in Vilnius | |
| Li et al. | Factors affecting bike-sharing behaviour in Beijing: price, traffic congestion, and supply chain | |
| Gugler et al. | Smart city strategy of dubai | |
| CN109583634A (en) | A kind of take-away Distribution path selection method based on Modern Portfolio Theory | |
| Schlote et al. | On closed-loop bicycle availability prediction | |
| Lu et al. | Demand-responsive transport for students in rural areas: A case study in vulkaneifel, germany | |
| Vallée et al. | Maximizing the number of served requests in an online shared transport system by solving a dynamic DARP | |
| Alemi et al. | Simulation of ridesourcing using agent-based demand and supply regional models: potential market demand for first-mile transit travel and reduction in vehicle miles traveled in the San Francisco Bay Area. | |
| Romero-Torres et al. | Measurement of the quality of service of the informal transportation mode mototaxi in Mexico | |
| CN113053156B (en) | Intelligent bus station addressing method based on bus radius method | |
| Li et al. | Using non-cooperative game theory for taxi-sharing recommendation systems |
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 |
Application publication date: 20161012 |
|
| WD01 | Invention patent application deemed withdrawn after publication |