CN108830509A - One kind is cruised taxi transport power scale dynamic adjusting method - Google Patents

One kind is cruised taxi transport power scale dynamic adjusting method Download PDF

Info

Publication number
CN108830509A
CN108830509A CN201810754996.3A CN201810754996A CN108830509A CN 108830509 A CN108830509 A CN 108830509A CN 201810754996 A CN201810754996 A CN 201810754996A CN 108830509 A CN108830509 A CN 108830509A
Authority
CN
China
Prior art keywords
transport power
power scale
taxi
cruising
transport
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.)
Granted
Application number
CN201810754996.3A
Other languages
Chinese (zh)
Other versions
CN108830509B (en
Inventor
叶晓飞
刘文丽
黄正锋
郑彭军
金宇明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ningbo University
Original Assignee
Ningbo University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Ningbo University filed Critical Ningbo University
Priority to CN201810754996.3A priority Critical patent/CN108830509B/en
Publication of CN108830509A publication Critical patent/CN108830509A/en
Application granted granted Critical
Publication of CN108830509B publication Critical patent/CN108830509B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06Q10/06315—Needs-based resource requirements planning or analysis

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Traffic Control Systems (AREA)

Abstract

It cruises taxi transport power scale dynamic adjusting method the invention discloses one kind.The present invention is based on taxi information managing system operation data, comprehensively consider the average daily train number of bicycle, kilometres utilization, business revenue is horizontal, bicycle is averagely operated duration, single poor waiting time, the indexs such as net about vehicle business share ratio, vehicle transport power scale adjustment regression tree model of cruising is constructed using data mining-traditional decision-tree, it is proposed the cruise threshold value and ordering relation of vehicle dynamic adjustment mechanism and key index, with the taxi transport power scale adjustment index that solves the problems, such as to cruise without unified system, and each key index threshold value without specific quantitative criterion and lack index between importance ranking relationship the problem of.Compared with prior art, the present invention can carry out scientific and reasonable sequence to the importance of each key index relevant to taxi transport power scale, to can get more accurate taxi transport power scale forecast value, and then adjusts theoretical foundation and decision support are provided for city vehicle transport power scale of cruising.

Description

One kind is cruised taxi transport power scale dynamic adjusting method
Technical field
It cruises taxi transport power scale dynamic adjusting method the present invention relates to one kind, belongs to road traffic engineering technology neck Domain.
Background technique
Formulating reasonable taxi transport power scale adjustment system is that government implements macro readjustment of direction to taxi trade structure Important means is the important leverage alleviated industry contradiction and promote taxi trade benign development.It is existing that taxi transport power is advised Mould prediction is usually directed to ten thousand people's owning amounts, taxi accounts for many indexs such as public transport share ratio, kilometres utilization or rate of empty ride And threshold value, but taxi transport power scale adjustment index of cruising has no unified system, and each key index threshold value is without specific quantitative Importance ranking relationship between standard and shortage index, it is difficult to which the adjustment of vehicle transport power scale provides support to cruise.
Summary of the invention
The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and to provide one kind cruise taxi transport power rule Mould dynamic adjusting method can carry out scientific and reasonable sequence to the importance of each key index relevant to taxi transport power scale, To can get more accurate taxi transport power scale forecast value, and then adjusts theory is provided for city vehicle transport power scale of cruising Foundation and decision support.
The present invention specifically uses following technical scheme to solve above-mentioned technical problem：
One kind is cruised taxi transport power scale dynamic adjusting method, is included the following steps：
Step A, initial predicted is carried out to taxi transport power using equilibrium of supply and demand predicted method, obtains transport power scale initial predicted Value；
Step B, calculate with transport power scale external influence factors data corresponding to the transport power scale initial prediction, And by being compared to the ideal indicator of the transport power scale external influence factors data to transport power scale initial prediction It is adjusted, obtains transport power scale intermediate predictor；
Step C, the transport power scale intermediate predictor and current transport power scale influnecing factor data are inputted back Return decision tree prediction model, obtain the final predicted value of transport power scale, and according to the final predicted value of transport power scale to cruising out Transport power scale of hiring a car is adjusted；The regression tree prediction model be using cruise taxi transport power scale historical data as Decision variable is obtained using corresponding history transport power scale influnecing factor data as characteristic variable by training in advance.
Preferably, the transport power scale external influence factors data include at least one of following data：It cruises taxi Ten thousand people's owning amount of vehicle, taxi of cruising account for public transport share ratio, cruise taxi and net about vehicle business share ratio, taxi of cruising multiply The objective waiting time.
It is further preferred that the ideal indicator of the transport power scale external influence factors data is specific as follows：
Ten thousand people's owning amount of taxi of cruising is greater than 21；
Taxi of cruising accounts for public transport share ratio less than 18%；
Taxi and net about vehicle business share ratio cruise less than 10:6.54；
Whether cab-getter's waiting time cruise less than 11.38 minutes.
It is further preferred that described by being compared with the ideal indicator of the transport power scale external influence factors data Relatively transport power scale initial prediction is adjusted, it is specific as follows：
(1) transport power scale initial prediction is inputted；
(2) whether judgement ten thousand people's owning amount of taxi of cruising is greater than 21；It is to continue in next step；It is no, then it needs to increase transport power Scale 15, new transport power scale initial prediction is obtained, is returned (1)；
(3) whether judgement taxi of cruising accounts for public transport share ratio less than 18%；It is to continue in next step；It is no, then it needs to increase and transport Power scale 15, new transport power scale initial prediction is obtained, is returned (1)；
(4) whether judgement cruises taxi and net about vehicle business share ratio less than 10:6.54；It is to continue in next step；It is no, It then needs to increase transport power scale 15, obtains new transport power scale initial prediction, return (1)；
(5) whether judgement cruises cab-getter's waiting time less than 11.38 minutes；It is, then by final transport power scale Numerical value is exported as transport power scale intermediate predictor；It is no, then it needs to increase transport power scale 15, it is initially pre- to obtain new transport power scale Measured value returns (1).
Preferably, the transport power scale influnecing factor data include at least one of following data：It cruises taxi Car kilometer utilization rate, hundred kilometers of business revenues of taxi of cruising, the average daily train number of taxi bicycle of cruising, taxi bicycle of cruising daily are sought It receives, taxi bicycle of cruising averagely waits duration, taxi bicycle of cruising daily is operated duration.
Preferably, the regression tree prediction model uses the automatic cross-diagnosis device (Chi- in card side of variance analysis Squared Automatic Interaction Detector, CHAID) training obtains algorithm in advance.
Preferably, the specific formula of the equilibrium of supply and demand predicted method is as follows：
Wherein, N is transport power scale initial prediction, and unit is；LHaveIt cruises taxi always effective mileage travelled for the whole city, Unit is ten thousand Km；T was averaged the service time for one day taxi of cruising in the middle, unit h；K is the rate of empty ride of taxi of cruising；V It is averaged operating speed for taxi of cruising, unit Km/h.
Compared with prior art, technical solution of the present invention has the advantages that：
(1) complexity that the present invention is adjusted for taxi transport power scale of cruising, with taxi information managing system operation Based on data, comprehensively consider that the average daily train number of bicycle, kilometres utilization, business revenue be horizontal, bicycle is averagely operated duration, single poor waiting The indexs such as duration, net about vehicle business share ratio are constructed vehicle transport power scale adjustment of cruising using data mining-traditional decision-tree and returned Tree-model proposes the cruise threshold value and ordering relation of vehicle dynamic adjustment mechanism and key index, cruises vehicle transport power scale for city Adjustment provides theoretical foundation and decision support.
(2) it solves the problems, such as to cruise taxi transport power scale adjustment index without unified system, solves each key and refer to The problem of threshold value is marked without specific quantitative criterion and lacks the importance ranking relationship between index, while in view of competing with net about vehicle Relationship is striven, the adjustment of vehicle transport power scale provides support to cruise.
Detailed description of the invention
Fig. 1 is the flow diagram of taxi transport power scale one specific embodiment of dynamic adjusting method of the invention of cruising；
Fig. 2 is that the parameter of CHAID algorithm corresponds to exemplary diagram；
Fig. 3 is the structural schematic diagram of regression tree prediction model；
Fig. 4 is an example of regression tree prediction model.
Specific embodiment
In view of the shortcomings of the prior art, thinking of the invention be based on taxi information managing system operation data, it is comprehensive The average daily train number of conjunction consideration bicycle, kilometres utilization, business revenue are horizontal, bicycle is averagely operated duration, single poor waiting time, net about vehicle industry The indexs such as share ratio of being engaged in are constructed vehicle transport power scale adjustment regression tree model of cruising using data mining-traditional decision-tree, propose to patrol The threshold value and ordering relation of tourist bus dynamic adjustment mechanism and key index, with solve cruise taxi transport power scale adjustment index without The problem of unified system and each key index threshold value without specific quantitative criterion and lack the importance ranking pass between index The problem of being.
Taxi transport power scale dynamic adjusting method of cruising proposed by the invention, specifically includes following steps：
Step A, initial predicted is carried out to taxi transport power using equilibrium of supply and demand predicted method, obtains transport power scale initial predicted Value；
Step B, calculate with transport power scale external influence factors data corresponding to the transport power scale initial prediction, And by being compared to the ideal indicator of the transport power scale external influence factors data to transport power scale initial prediction It is adjusted, obtains transport power scale intermediate predictor；
Step C, the transport power scale intermediate predictor and current transport power scale influnecing factor data are inputted back Return decision tree prediction model, obtain the final predicted value of transport power scale, and according to the final predicted value of transport power scale to cruising out Transport power scale of hiring a car is adjusted；The regression tree prediction model be using cruise taxi transport power scale historical data as Decision variable is obtained using corresponding history transport power scale influnecing factor data as characteristic variable by training in advance.
Wherein, transport power scale external influence factors data and transport power scale influnecing factor data can be according to the actual situation Selection；Preferably, the transport power scale external influence factors data include at least one of following data：It cruises taxi ten thousand People's owning amount, taxi of cruising account for public transport share ratio, cruise taxi and net about vehicle business share ratio, the cab-getter that cruises etc. To the time；The transport power scale influnecing factor data include at least one of following data：It cruises and hires out car kilometer benefit With rate, hundred kilometers of business revenues of taxi of cruising, the average daily train number of taxi bicycle of cruising, the average daily business revenue of taxi bicycle of cruising, cruise Duration that taxi bicycle averagely waits duration, taxi bicycle of cruising daily is operated.For the sake of description, part shadow hereinafter Ring " taxi of cruising " before being omitted in factor title.
Preferably, the ideal indicator of the transport power scale external influence factors data is specific as follows：
Ten thousand people's owning amount of taxi of cruising is greater than 21；
Taxi of cruising accounts for public transport share ratio less than 18%；
Taxi and net about vehicle business share ratio cruise less than 10:6.54；
Whether cab-getter's waiting time cruise less than 11.38 minutes.
The regression tree prediction model can be used existing C4.5 algorithm, C5.0 algorithm, CHAID algorithm, CART and calculate Method and QUEST algorithm etc., training obtains the preferred CHAID algorithm of the present invention in advance.
For the ease of public understanding, below with a specific embodiment and in conjunction with attached drawing come to technical solution of the present invention into Row is further described：
The process of the method for taxi transport power scale dynamic adjustment of cruising in the present embodiment is as shown in Figure 1, specifically include Following steps：
(1) city is obtained to cruise the relevant achievement data of taxi；Data content divides external influence factors and internal influence Factor；External influence factors have：Vehicle guaranteeding organic quantity, urban district permanent resident population, taxi account for public transport share ratio, cruises vehicle and net about Vehicle business share ratio etc.；Influnecing factor has：Kilometres utilization, passenger's average latency, taxi it is effective traveling in Journey, hundred kilometers of business revenues, the average daily train number of bicycle, the average daily business revenue of bicycle etc.；
(2) according to the index of correlation data of the taxi of cruising of step (1) acquisition, with equilibrium of supply and demand predicted method to taxi Vehicle transport power is predicted, predicts transport power scale；The specific formula of equilibrium of supply and demand predicted method in the present embodiment is as follows：
Wherein, N is transport power scale initial prediction, and unit is；LHaveIt cruises taxi always effective mileage travelled for the whole city, Unit is ten thousand Km；T was averaged the service time for one day taxi of cruising in the middle, unit h；K is the rate of empty ride of taxi of cruising；V It is averaged operating speed for taxi of cruising, unit Km/h；
(3) the transport power scale numerical value for obtaining step (2) equilibrium of supply and demand predicted method is defeated as transport power scale initial prediction Enter；
(4) judge whether ten thousand people's owning amounts are greater than 21；It is to continue in next step；It is no, then it needs to increase transport power scale 15, Obtain new transport power scale initial prediction, return step (3)；
(5) judge that whether taxi accounts for public transport share ratio less than 18%；It is to continue in next step；It is no, then it needs to increase transport power rule Mould 15, obtain new transport power scale initial prediction, return step (3)；
(6) whether judgement cruises vehicle and net about vehicle business share ratio less than 10:6.54；It is to continue in next step；It is no, then it needs Increase transport power scale 15, obtains new transport power scale initial prediction, return step (3)；
(7) judge passenger waiting time whether less than 11.38 minutes；It is, then using final transport power scale numerical value as fortune The output of power scale intermediate predictor；It is no, then it needs to increase transport power scale 15, obtains new transport power scale initial prediction, return Step (3)；
(8) current kilometres utilization, passenger's average latency, effective mileage travelled of taxi, hundred kilometers are obtained The transport power scale influnecing factor data such as the average daily train number of business revenue, bicycle, the average daily business revenue of bicycle, and with obtained transport power scale Intermediate predictor inputs regression tree prediction model together, obtains the final predicted value of transport power scale；
(9) the transport power scale value finally predicted is exported；
(10) it obtains transport power scale Adjusted Option, and continues the supply and demand situation new to market and detect, more new data.
For regression tree prediction model in the present embodiment using the training building of CHAID algorithm, which is the prior art, The mapping relations of correlated variables such as the following table 1：
1 parameter interpretation table of table
Fig. 2 is that the parameter of CHAID algorithm corresponds to exemplary diagram, and the basic process that algorithm is realized is as follows：
1. firstly, confirmation node m：
Assuming that a certain key index such as kilometres utilization corresponds to decision node m (corresponding index as shown in Figure 2), another XmFor X (all sample training data of 2 index value of corresponding diagram) reach the subset of node m, i.e., it is the satisfaction of x ∈ X from tree root to node All x of all decision point conditions of m, then
2. then, the mean square error E of calculate node m estimated valuem, obtain division threshold value, and complete branch：
The mean square error of estimated value:
Wherein
Wherein, gmFor the estimated value in node m, calculation method is as follows:
If Em<θr(wherein, θrFor acceptable error), then a tree node is created, g is storedm；
Otherwise, reach the data further division of node m, so that the error and minimum of branch.It (on each node, seeks The division threshold value for looking for the attribute and numerical attribute that minimize error, then recursively proceeds as described above.)
Enable XmjFor XmTake branch's j subset：Definition
Enable gmjIt is the estimated value for reaching the branch j of node m
And the error after dividing is
For arbitrarily dividing, the reduction of error is provided by the difference of formula (2) and formula (6).Find the reduction for maximizing error Division direction is equivalent to formula (6) and is minimized.Mean square error is a kind of possible error function, another worst error
It can guarantee that the error of any example is all not more than given threshold value using worst error.Acceptable error threshold is The function of complexity；Its value is smaller, and the tree risk that is bigger and being excessively fitted of generation is bigger, and value is bigger, and fitting is insufficient and mistake A possibility that light splitting is sliding is bigger.
When each decision node uses all critical index as input dimension, then Linear Multivariable node definition is
From root to the successive nodes further division example on the path of leaf, and leaf node define it is more in the input space Face body, one by one to weight wmjIt is finely adjusted to reduce statistical significance index, selects to reduce dimension by subset and reduce node Complexity.
3. finally, obtaining decision tree structure figure.
The regression tree prediction model of similar Fig. 3 finally can be obtained.
In order to verify effect of the present invention, with Ningbo City's data instance, to be adjusted to taxi transport power scale dynamic of cruising, tool Steps are as follows for body：
1) data are acquired.City is obtained to cruise the relevant achievement data of taxi；Data content includes：Point external action because Element and influnecing factor；External influence factors have：Vehicle guaranteeding organic quantity, urban district permanent resident population, taxi account for public transport share ratio, It cruises vehicle and net about vehicle business share ratio etc..Influnecing factor has：Kilometres utilization, passenger's average latency, taxi Effective mileage travelled, hundred kilometers of business revenues, the average daily train number of bicycle, the average daily business revenue of bicycle etc.；
2) according to the index of correlation data of the taxi of cruising of step 1) acquisition, with equilibrium of supply and demand predicted method to taxi Transport power is predicted that concrete outcome is as follows：
=effective mileage travelled (1110138)/(taxi is averaged service time (13h) * (1- rate of empty ride in one day (30%)) * taxi is averaged overall trip speed (28km/h) * 0.9)=4841
3) it is inputted according to the transport power scale numerical value that step 2 equilibrium of supply and demand predicted method obtains as transport power scale initial value；
Transport power scale forecast value 4841 ()；
4) judge whether ten thousand people's owning amount of taxi is greater than 21；
Ningbo City's taxi total amount 4841,220.04 ten thousand people of urban district permanent resident population；
Calculation method：Ten thousand people's owning amount of taxi=taxi number ()/urban district permanent resident population (ten thousand people)=4841/ 220.04=22>21, continue in next step.
5) judge that whether taxi accounts for public transport share ratio less than 18%；
Ningbo City's taxi accounts for public transport share ratio=17.2%<18%, continue in next step.
6) whether judgement cruises vehicle and net about vehicle business share ratio less than 10:6.54；
Ningbo City cruises vehicle and net about vehicle business share ratio=10:7.35<10:6.54, continue in next step.
7) judge passenger's average latency whether less than 11.38 minutes；
Passenger average latency=10.46 minute<11.38 minutes, continue in next step.
8) current kilometres utilization, passenger's average latency, effective mileage travelled of taxi, hundred kilometers of battalion are obtained The transport power scale influnecing factor data such as receipts, the average daily train number of bicycle, the average daily business revenue of bicycle, and in obtained transport power scale Between predicted value input regression tree prediction model together, obtain the final predicted value of transport power scale；
Wherein, the algorithm using CHAID algorithm as building regression tree, data are using in January, 2014 in August, 2017 Ningbo City's taxi operation data carry out modeling analysis, and the corresponding data source of key index goes out in Passenger Transport Authorities of Ningbo City It hires a car operation and management information system, comprising 1317 groups of effective sample data (about one day one group), wherein 856 groups of training sample, about It accounts for 65% to construct for decision tree, 35% is used for model testing.The regression tree prediction model finally constructed is as shown in Figure 4.
Obtain Ningbo City cruise taxi the current adjustment period each index value it is as follows：
Kilometres utilization is 0.6699；
Hundred kilometers of business revenues are 235.53 yuan；
The average daily train number of bicycle is 40.80 times；
The average daily business revenue of bicycle is 910.67 yuan；
It is 209.49 seconds a length of when single poor averagely waiting；
A length of 602.85 points when bicycle is daily operated；
It is obtained by preceding step, transport power scale is 4841+0=4841, value is inputted in decision tree structure, according to Fig. 3 Recurrence tree construction index is judged：
Node 1：The kilometres utilization of taxi be 0.6699 in (0.6179,0.6616] in section, therefore predict fortune Power needs 4955 taxis, then adjusts window and opens, launches transport power (4955-4841=114)；
Node 4：A length of 602.85 when the bicycle of taxi is daily operated<=616.27606, therefore predict that transport power needs 4977 Taxi then adjusts window and opens, launches transport power (4977-4841=136)；
Node 7：The average daily train number of bicycle is 40.80>39.73000, therefore predict that transport power needs 4990 taxis, then adjust Whole window is opened, and is launched transport power (4990-4841=149)；
Therefore, it need to finally launch transport power (114+136+149=399), adjustment transport power scale is 4841+399=5240 ?.
9) transport power scale value 5240 finally predicted are exported.
10) transport power scale Adjusted Option is obtained：Taxi transport power 399 need to be launched, transport power scale is 5240.Continue pair The new supply and demand situation in market is detected, more new data.

Claims (7)

  1. The taxi transport power scale dynamic adjusting method 1. one kind is cruised, which is characterized in that include the following steps：
    Step A, initial predicted is carried out to taxi transport power using equilibrium of supply and demand predicted method, obtains transport power scale initial prediction；
    Step B, calculate with transport power scale external influence factors data corresponding to the transport power scale initial prediction, and lead to It crosses and is compared to carry out transport power scale initial prediction with the ideal indicator of the transport power scale external influence factors data Adjustment, obtains transport power scale intermediate predictor；
    Step C, the transport power scale intermediate predictor is returned with current transport power scale influnecing factor data input and is determined Plan tree prediction model obtains the final predicted value of transport power scale, and according to the final predicted value of transport power scale to taxi of cruising Transport power scale is adjusted；The regression tree prediction model is using taxi transport power scale historical data of cruising as decision Variable is obtained using corresponding history transport power scale influnecing factor data as characteristic variable by training in advance.
  2. 2. method as described in claim 1, which is characterized in that the transport power scale external influence factors data include following data At least one of：Cruise ten thousand people's owning amount of taxi, taxi of cruising account for public transport share ratio, cruise taxi and net about vehicle industry Business share ratio is cruised cab-getter's waiting time.
  3. 3. method as claimed in claim 2, which is characterized in that the ideal indicator of the transport power scale external influence factors data has Body is as follows：
    Ten thousand people's owning amount of taxi of cruising is greater than 21；
    Taxi of cruising accounts for public transport share ratio less than 18%；
    Taxi and net about vehicle business share ratio cruise less than 10:6.54；
    Whether cab-getter's waiting time cruise less than 11.38 minutes.
  4. 4. method as claimed in claim 3, which is characterized in that it is described by with the transport power scale external influence factors data Ideal indicator is compared to be adjusted transport power scale initial prediction, specific as follows：
    (1) transport power scale initial prediction is inputted；
    (2) whether judgement ten thousand people's owning amount of taxi of cruising is greater than 21；It is to continue in next step；It is no, then it needs to increase transport power scale 15, new transport power scale initial prediction is obtained, is returned (1)；
    (3) whether judgement taxi of cruising accounts for public transport share ratio less than 18%；It is to continue in next step；It is no, then it needs to increase transport power rule Mould 15, new transport power scale initial prediction is obtained, is returned (1)；
    (4) whether judgement cruises taxi and net about vehicle business share ratio less than 10:6.54；It is to continue in next step；It is no, then it needs Increase transport power scale 15, obtain new transport power scale initial prediction, returns (1)；
    (5) whether judgement cruises cab-getter's waiting time less than 11.38 minutes；It is, then by final transport power scale numerical value It is exported as transport power scale intermediate predictor；It is no, then it needs to increase transport power scale 15, obtains new transport power scale initial predicted Value returns (1).
  5. 5. method as described in claim 1, which is characterized in that the transport power scale influnecing factor data include following data At least one of：It cruises taxi kilometres utilization, hundred kilometers of business revenues of taxi of cruising, the average daily vehicle of taxi bicycle of cruising When secondary, the average daily business revenue of taxi bicycle of cruising, taxi bicycle of cruising averagely wait duration, taxi bicycle of cruising daily is operated It is long.
  6. 6. method as described in claim 1, which is characterized in that the regression tree prediction model uses the card side of variance analysis Automatic cross-diagnosis device algorithm training in advance obtains.
  7. 7. method as described in claim 1, which is characterized in that the specific formula of the equilibrium of supply and demand predicted method is as follows：
    Wherein, N is transport power scale initial prediction, and unit is；LHaveIt cruises taxi always effective mileage travelled for the whole city, unit For ten thousand Km；T was averaged the service time for one day taxi of cruising in the middle, unit h；K is the rate of empty ride of taxi of cruising；V is to patrol Swim out of average operating speed of hiring a car, unit Km/h.
CN201810754996.3A 2018-07-11 2018-07-11 Method for dynamically adjusting capacity scale of touring taxi Active CN108830509B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810754996.3A CN108830509B (en) 2018-07-11 2018-07-11 Method for dynamically adjusting capacity scale of touring taxi

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810754996.3A CN108830509B (en) 2018-07-11 2018-07-11 Method for dynamically adjusting capacity scale of touring taxi

Publications (2)

Publication Number Publication Date
CN108830509A true CN108830509A (en) 2018-11-16
CN108830509B CN108830509B (en) 2021-11-23

Family

ID=64136840

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810754996.3A Active CN108830509B (en) 2018-07-11 2018-07-11 Method for dynamically adjusting capacity scale of touring taxi

Country Status (1)

Country Link
CN (1) CN108830509B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112990610A (en) * 2021-05-06 2021-06-18 北京工业大学 Method for predicting taxi capacity demand of railway station based on multiple linear regression

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103761585A (en) * 2014-01-27 2014-04-30 北京市交通运行监测调度中心 Airport continuous transport dynamic matching, transport capacity monitoring and early warning and intelligent dispatching method
CN104021668A (en) * 2014-06-26 2014-09-03 中国科学院自动化研究所 Public traffic supply-demand detection and prediction system and method
CN107274665A (en) * 2017-07-31 2017-10-20 多维新创（北京）技术有限公司 Bus transport capacity resource method and system for planning
CN107679746A (en) * 2017-09-29 2018-02-09 飞友科技有限公司 A kind of method for calculating flight transport power

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103761585A (en) * 2014-01-27 2014-04-30 北京市交通运行监测调度中心 Airport continuous transport dynamic matching, transport capacity monitoring and early warning and intelligent dispatching method
CN104021668A (en) * 2014-06-26 2014-09-03 中国科学院自动化研究所 Public traffic supply-demand detection and prediction system and method
CN107274665A (en) * 2017-07-31 2017-10-20 多维新创（北京）技术有限公司 Bus transport capacity resource method and system for planning
CN107679746A (en) * 2017-09-29 2018-02-09 飞友科技有限公司 A kind of method for calculating flight transport power

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
刘嘉琪等: "互联网时代出租车供需匹配及补贴方案确定", 《经济数学》 *
吴昊等: "基于供需平衡的出租车规模确定方法", 《洛阳理工学院学报》 *
朱珍龙: "出租车运营模式与运力规模优化研究", 《中国优秀硕士学位论文全文数据库》 *
洪杰等: "基于因素分析的出租车运力规模预测", 《宁波大学学报》 *
雷旻虎: "城市出租汽车数量管制研究", 《中国优秀硕士学位论文全文数据库•工程科技Ⅱ辑》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112990610A (en) * 2021-05-06 2021-06-18 北京工业大学 Method for predicting taxi capacity demand of railway station based on multiple linear regression
CN112990610B (en) * 2021-05-06 2021-08-20 北京工业大学 Method for predicting taxi capacity demand of railway station based on multiple linear regression

Also Published As

Publication number Publication date
CN108830509B (en) 2021-11-23

Similar Documents

Publication Publication Date Title
CN111260221B (en) Traffic facility bearing capacity evaluation method based on dynamic model and oriented to city global situation
Zhou et al. Bus arrival time calculation model based on smart card data
CN103632212B (en) System and method for predicating time-varying user dynamic equilibrium network-evolved passenger flow
CN102496076B (en) Macroscopic, mid-scope and microscopic multilevel urban parking demand prediction model integrated system
WO2018023331A1 (en) System and method for real-time evaluation of service index of regular public buses
Li et al. Prediction of electric bus energy consumption with stochastic speed profile generation modelling and data driven method based on real-world big data
Chen et al. Uncertainty in urban mobility: Predicting waiting times for shared bicycles and parking lots
CN105279572A (en) City track traffic passenger flow density index calculating and releasing system
CN106408984B (en) Urban traffic flow induction system based on Internet of things
CN111275965A (en) Real-time traffic simulation analysis system and method based on internet big data
CN111598317A (en) Generation optimization method for customized train operation scheme of railway
Csiszár et al. Method for analysis and prediction of dwell times at stops in local bus transportation
CN110415513B (en) Method, system and electronic device for publishing bus lane service index
Ding et al. The optimization of airport management based on collaborative optimization of flights and taxis
CN106327867A (en) Bus punctuality prediction method based on GPS data
Yu et al. Research on High‐Speed Railway Freight Train Organization Method considering Different Transportation Product Demands
Wang et al. An Ensemble Learning Model for Short‐Term Passenger Flow Prediction
Zhu et al. XGBoost‐Based Travel Time Prediction between Bus Stations and Analysis of Influencing Factors
Wang et al. A simulation-based metro train scheduling optimization incorporating multimodal coordination and flexible routing plans
Cui et al. Study on the selection model of staying adjustment bus lines along rail transit
CN108830509A (en) One kind is cruised taxi transport power scale dynamic adjusting method
CN111723871B (en) Estimation method for real-time carriage full load rate of bus
Luo et al. Regression model for daily passenger volume of high-speed railway line under capacity constraint
Deng et al. Research on bus passenger traffic forecasting model based on gps and ic card data
Tian et al. Method for determining the valid travel route of railways based on generalised cost under the syncretic railway network

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
GR01 Patent grant
GR01 Patent grant