CN107679668A - The electric bicycle travel time prediction method of duration model based on risk - Google Patents

The electric bicycle travel time prediction method of duration model based on risk Download PDF

Info

Publication number
CN107679668A
CN107679668A CN201710958398.3A CN201710958398A CN107679668A CN 107679668 A CN107679668 A CN 107679668A CN 201710958398 A CN201710958398 A CN 201710958398A CN 107679668 A CN107679668 A CN 107679668A
Authority
CN
China
Prior art keywords
mrow
electric bicycle
travel time
risk
trip
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
Application number
CN201710958398.3A
Other languages
Chinese (zh)
Inventor
徐铖铖
邓翎
刘攀
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.)
Southeast University
Original Assignee
Southeast 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 Southeast University filed Critical Southeast University
Priority to CN201710958398.3A priority Critical patent/CN107679668A/en
Publication of CN107679668A publication Critical patent/CN107679668A/en
Pending legal-status Critical Current

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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • 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
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G06Q50/40—

Abstract

The invention discloses a kind of electric bicycle travel time prediction method of duration model based on risk.First, individual and Family characteristics and the whole day trip information of electric bicycle user is obtained；Secondly, parameter corresponding to different demarcation under different items of information is extracted；Then, by gross sample, this according to sex is divided into masculinity and femininity two parts；Then, the forecast model of masculinity and femininity electric bicycle travel time is built using the duration model of risk；The travel time of electric bicycle is predicted by the duration model of risk, and according to influence of the interpretation of result various factors for the electric bicycle travel time.The travel time of electric bicycle can be predicted using the inventive method, contributes to professional person to predict the use demand of electric bicycle, so as to formulate the policy for promoting city electric bicycle to use and carry out rural infrastructure planning.

Description

The electric bicycle travel time prediction method of duration model based on risk
Technical field
The present invention relates to a kind of Forecasting Methodology of electric bicycle travel time, more particularly to a kind of continued based on risk The electric bicycle travel time prediction method of time model.
Background technology
Compared with automobile, electric bicycle is a kind of environmentally friendly, sustainable mode of transportation, can be with using electric bicycle Help deterioration urban transportation and the pollution problem for solving such as traffic congestion, death by accident, energy expenditure and air quality；It is electronic Interdependence between bicycle utilization rate and each influence factor, important work is played in different traffic programme work With.Therefore, electric bicycle is an essential research topic in Chinese transportation field.
Found when consulting conventional research, study and all do not differentiate between masculinity and femininity mostly, but in fact, sex An important factor for being to determine travel behaviour, the body and psychological characteristics of women make their travel time to a certain extent with man Property it is different, in addition, most of research is all concentrated on the select permeability of electric bicycle pattern, few researchs consider from Drive a vehicle the time gone on a journey, and the travel time significantly affects for transport need, is to determine that one of transport need is important Factor.By the research to the electric bicycle travel time, the shadow that various factors is used bicycle can be best understood from Ring, and will be helpful to Non_signal intersection personage and predict the demand that electric bicycle uses, this is to formulate to promote electrical salf-walking Effective policy that car uses and the important prerequisite for carrying out rural infrastructure planning.
The content of the invention
Technical problem：The present invention provides a kind of electric bicycle travel time prediction of the duration model based on risk Method, this method can be used for analyzing the travel time of electric bicycle, help to formulate promotion city electric bicycle use Policy and carry out rural infrastructure planning.
Technical scheme：A kind of electric bicycle travel time of duration model based on risk of the present invention is pre- Survey method, comprises the following steps：
(1) individual and Family characteristics and their trip information of electric bicycle user is obtained；
(2) parameter corresponding to different demarcation under different items of information is extracted；
(3) by gross sample, this according to sex is divided into masculinity and femininity two parts；
(4) duration model based on risk, the masculinity and femininity electric bicycle travel time is modeled respectively；
(5) predict the travel time of electric bicycle, at the same according to model result analyze various factors for it is electronic from The influence of driving travel time.
The individual referred in the step (1) and Family characteristics and trip information mainly include：Traveler occupation, traveler Age, annual family income, automobile possess situation, trip purpose, trip distance, go out the beginning-of-line density of population, trip duration, trip Whether the terminal density of population, travel time are morning peak and the magnitude of traffic flow of origin and destination.
Parameter in the step (2) is arranged to：Traveler occupation be student, worker, official and other, its corresponding parameter For x1i、x2i、x3i、x4i；The traveler age is less than between 20 years old, 20 to 40 years old, between 40 to 50 years old and more than 50 years old, correspondence Parameter is x5i, x6i, x7i, x8i；Annual family income is less than 2000 RMB and more than 20000 RMB, and corresponding parameter is x9i、 x10i；The existing automobile of family, the coming five years can buy automobile, the following automobile and future of buying for 10 years will not buy automobile, and corresponding parameter is x11i、x12i、x13i、x14i；Trip purpose is work, go to school, do shopping, going home and other, corresponding parameter is x15i、x16i、x17i、 x18i、x19i；Trip purpose respectively works, gone to school, doing shopping, going home, other trip distances, and corresponding parameter is x20i、x21i、 x22i、x23i、x24i、x25i；Go out the beginning-of-line density of population more than 0.023 people/square metre and less than 0.023 people/square metre, corresponding ginseng Number is x26i、x27i；The travel destination density of population be more than 0.023 people/square metre and less than 0.023 people/square metre, corresponding parameter is x28i、x29i；Travel time is morning peak, and corresponding parameter is x30i；The magnitude of traffic flow of origin and destination, corresponding parameter is x31i；Other letters Breath, corresponding parameter xki；I represents i-th part of questionnaire.
The step (4) comprises the following steps：
(41) duration model based on risk, travel time t cumulative distribution function F (t) are as follows：
Wherein f (t) is travel time t probability density function, and S (t) is survival function, and it provides the duration and is more than t Probability, dangerous function h (t) gives the probability for terminating stroke in time t, and condition is that the trip does not have also before time t Terminate, equation is as follows：
(42) accelerated failure-time model is used to be solved as dangerous function to explain in the duration model based on risk The influence of variable is released, the expression formula of accelerated failure-time model is as follows：
h(ti)=h0(t×exp(-(β0+β1x1i+…βnxni)))×exp(-(β0+β1x1i+…βnxni))
Wherein h0 () is baseline hazard function, and n is the quantity of explanatory variable, and accelerated failure-time model can also be write as Following expression-form：
Ln (t)=β0+β1x1i+β2x2i+β3x3i+β4x4i+β5x5i+β6x6i+β7x7i+β8x8i+β9x9i+β10x10i+β11x11i+ β12x12i+β13x13i+β14x14i+β15x15i+β16x16i+β17x17i+β18x18i+β19x19i+β20x20i+β21x21i+β22x22i+β23x23i+ β24x24i+β25x25i+β26x26i+β27x27i+β28x28i+β29x29i+β30x30i+β31x31i+βkxki+θ+σεi
Wherein t represents travel time, εiIt is independent random error, σ is scale parameter, and θ is stochastic effects, βkIt is corresponding Coefficient；
(43) dangerous function is estimated using Weibull distribution, the probability density function and dangerous function of Wei Buer distributions are such as Under：
F (t | λ, p)=λ p (λ t)p-1e-(λt)p, h (t)=λpptp-1。
Wherein, introducing of the accelerated failure-time model to stochastic effects is for explaining the heterogeneous of whole sample.
Beneficial effect：The present invention compared with prior art, beneficial effects of the present invention：1st, the trip of electric bicycle is studied Time can be best understood from the influence that various factors is used electric bicycle, contribute to Non_signal intersection personage to predict electricity The demand that dynamic bicycle uses, and formulate effective policy and the important prerequisite of good basis facilities planning；2nd, by gross sample one's duty For the major class of masculinity and femininity two, model respectively, make prediction more accurate；3rd, present invention uses accelerated failure-time model to divide The travel time of electric bicycle is analysed, explanatory variable can be preferably caught and the electric bicycle travel time is directly affected； 4th, this method has used Weibull distribution to carry out calculated risk function, by Weibull, lognormal, normal state and exponential distribution this Four kinds of distributions carry out likelihood ratio test, as a result prove that the likelihood ratio of Weibull distribution is maximum, the distribution is for electrical salf-walking Car travel time data are most suitably used.
Brief description of the drawings
Fig. 1 is the FB(flow block) of the present invention；
Fig. 2 is the mean absolute percentage error of male's model；
Fig. 3 is the mean absolute percentage error of Female Model；
Fig. 4 is the male's AFT models for having stochastic effects；
Fig. 5 is male's AFT models of no stochastic effects；
Fig. 6 is the women AFT models for having stochastic effects；
Fig. 7 is the women AFT models of no stochastic effects.
Embodiment
With reference to Figure of description, the present invention is further illustrated.
Individual and Family characteristics and the trip information of electric bicycle user is obtained, is mainly included：Traveler occupation, go out Passerby's age, annual family income, automobile possess situation, trip purpose, trip distance, go out the beginning-of-line density of population, trip duration, Whether the travel destination density of population, travel time are morning peak and the magnitude of traffic flow of origin and destination.
Extract the parameter corresponding to the different demarcation under different items of information：Traveler occupation is student, worker, Guan Yuanhe Other, its corresponding parameter is x1i、x2i、x3i、x4i；The traveler age be less than between 20 years old, 20 to 40 years old, between 40 to 50 years old and More than 50 years old, corresponding parameter was x5i, x6i, x7i, x8i；Annual family income is less than 2000 RMB and more than 20000 RMB , corresponding parameter is x9i、x10i；The existing automobile of family, the coming five years can buy automobile, future can buy automobile in 10 years and future will not Automobile is bought, corresponding parameter is x11i、x12i、x13i、x14i；Trip purpose is work, go to school, do shopping, going home and other, corresponding parameter For x15i、x16i、x17i、x18i、x19i；Trip purpose respectively works, gone to school, doing shopping, going home, other trip distances, corresponding ginseng Number is x20i、x21i、x22i、x23i、x24i、x25i；Go out the beginning-of-line density of population more than 0.023 people/square metre and less than 0.023 people/ Square metre, corresponding parameter is x26i、x27i；The travel destination density of population be more than 0.023 people/square metre and less than 0.023 people/square Rice, corresponding parameter is x28i、x29i；Travel time is morning peak, and corresponding parameter is x30i；The magnitude of traffic flow of origin and destination, corresponding parameter For x31i；Other information, corresponding parameter xki；I represents i-th part of questionnaire.
By gross sample, this according to sex is divided into masculinity and femininity two parts.
Duration model based on risk, the masculinity and femininity electric bicycle travel time is modeled respectively：
Duration model based on risk, travel time t cumulative distribution function F (t) are as follows：
Wherein f (t) is travel time t probability density function, and S (t) is survival function, and it provides the duration and is more than t Probability, dangerous function h (t) gives the probability for terminating stroke in time t, and condition is that the trip does not have also before time t Terminate, equation is as follows：
Using accelerated failure-time model (AFT) as dangerous function, to explain in the duration model based on risk The influence of explanatory variable, the expression formula of accelerated failure-time model are as follows：
h(ti)=h0(t×exp(-(β0+β1x1i+…βnxni)))×exp(-(β0+β1x1i+…βnxni))
Wherein h0 () is baseline hazard function, and n is the quantity of explanatory variable, and accelerated failure-time model can also be write as Following expression-form：
Ln (t)=β0+β1x1i+β2x2i+β3x3i+β4x4i+β5x5i+β6x6i+β7x7i+β8x8i+β9x9i+β10x10i+β11x11i+ β12x12i+β13x13i+β14x14i+β15x15i+β16x16i+β17x17i+β18x18i+β19x19i+β20x20i+β21x21i+β22x22i+β23x23i+ β24x24i+β25x25i+β26x26i+β27x27i+β28x28i+β29x29i+β30x30i+β31x31i+βkxki+θ+σεi
Wherein t represents travel time, εiIt is independent random error, σ is scale parameter, and θ is stochastic effects, βkIt is corresponding Coefficient；
Estimate dangerous function using Weibull distribution, the probability density function and dangerous function that Wei Buer is distributed are as follows：
F (t | λ, p)=λ p (λ t)p-1e-(λt)p, h (t)=λpptp-1。
Wherein, introducing of the accelerated failure-time model to stochastic effects is for explaining the heterogeneous of whole sample.
Using 2007, the family's outgoing data that Chinese Shaoxin City is taken a broad survey, Shaoxing was to be located at East China Sea bank Typical medium-sized city, population is 90.85 ten thousand people, and the gross area is 59.96 square kilometres within 2007, has used 7320 parts to ask altogether Volume investigation.
The electric bicycle trip duration prediction model result of duration model structure based on risk is as shown in table 1.
Table 1 has the estimated result of the accelerated failure-time model of stochastic effects
Note:aStandard deviation；
bThe variable is unimportant in a model；
cWith reference to rank
It can be drawn the following conclusions by upper table：The estimation parameter of the trip distance of masculinity and femininity is all positive number, show with The increase of trip distance, electric bicycle travel time of masculinity and femininity traveler is likely to longer；For different Trip purpose, the parameter of trip distance is less than the trip distance parameter in Female Model in male's model, to this possible explanation It is that the trip speed of female travelers is lower than male traveler；Because the time of women electric bicycle trip is shorter than male, The distance of women electric bicycle trip is shorter than man；In addition, the parameter of operating distance is less than the parameter of other stroke distances, table Bright electric bicycle traveler, which is on duty in trip, more likely faster travel speed.
The parameter of the origin and destination volume of traffic is all positive, shows the duration of electric bicycle trip with starting point destination The increase of the volume of traffic and increase；However, the parameter in male's model is more than the parameter in Female Model, to this possible explanation It is that the duration that traffic congestion is gone on a journey to male's electric bicycle has a great influence, because male might have longer trip Distance；The origin and destination density of population is higher can cause the two-way trip of men and women travel time reduction because between these areas The road network traffic capacity is big, so as to reduce the running time of electric bicycle.
On the age of trip, parameter shows：The travel time of the male of 20 to 50 years old may be than more than 50 years old male Travel time it is shorter, on the contrary, the travel time of the women of 20 to 50 years old may be than the more than 50 years old women travel time more It is long, the explanation for this phenomenon be possible go out between elderly men and elderly woman row mode difference it is relevant, old man Property may like going on a journey for a long time because of retirement, and old women may be partial to be gone out in short term due to home duties OK.
On trip purpose variable, if trip purpose is work, then male's travel time can be longer than women, male Shopping parameter be negative, and this parameter is unimportant for women, illustrate time that male spends in shopping compared with It is few.
Morning peak and occupation can also influence the electric bicycle travel time, for men and women, if in the morning peak period Electric bicycle trip is carried out, the time specifically gone on a journey is more likely longer, and this is probably due to the traffic congestion of morning peak phase； Masculinity and femininity worker's electric bicycle travel time is longer, because they may stay in suburb.
In order to assess the estimated performance of the model, examined using mean absolute percentage error (MAPE) relative to trip The error of view of time measured value, MAPE are defined as：
Wherein tA(i) actual value of ith observation, t are representedp(i) predicted value of ith observation is represented, Fig. 2 and Fig. 3 are provided It is different go out line duration accelerated failure-time (AFT) model MAPE values, the MAPEs of stochastic effects AFT models is less than The MAPEs of AFT models, and to various durations without Random Effect.
Conventional researcher used 50%, 20% and 10% MAPE threshold values represent reasonable, good and high accuracy this Several ranks, assess the estimated performance of the duration model based on risk respectively, men and women's stochastic effects AFT models in table 1 Overall MAPE is respectively 10.4% and 11.8%, illustrates the AFT models of exploitation and has preferable precision of prediction；In order to compare, together When also calculate no stochastic effects AFT models MAPEs, be as a result 39.5% and 40.8%, therefore, include stochastic effects The estimated performance of AFT models can be improved, Fig. 4, Fig. 5, Fig. 6, Fig. 7 give the AFT models with and without stochastic effects Actual value and predicted value figure, also indicate that stochastic effects AFT models can provide more preferable estimated performance.

Claims (5)

1. a kind of electric bicycle travel time prediction method of duration model based on risk, it is characterised in that including Following steps：
(1) individual and Family characteristics and their trip information of electric bicycle user is obtained；
(2) parameter corresponding to different demarcation under different items of information is extracted；
(3) by gross sample, this according to sex is divided into masculinity and femininity two parts；
(4) duration model based on risk, the masculinity and femininity electric bicycle travel time is modeled respectively；
(5) travel time of electric bicycle is predicted, while various factors is analyzed for electric bicycle according to model result The influence of travel time.
2. the electric bicycle travel time prediction method of the duration model according to claim 1 based on risk, Characterized in that, the individual referred in the step (1) and Family characteristics and trip information mainly include：Traveler occupation, go out Passerby's age, annual family income, automobile possess situation, trip purpose, trip distance, go out the beginning-of-line density of population, trip duration, Whether the travel destination density of population, travel time are morning peak and the magnitude of traffic flow of origin and destination.
3. the electric bicycle travel time prediction method of the duration model according to claim 1 based on risk, Characterized in that, the parameter in the step (2) is arranged to：Traveler occupation be student, worker, official and other, it is corresponding Parameter is x1i、x2i、x3i、x4i；The traveler age is less than between 20 years old, 20 to 40 years old, between 40 to 50 years old and more than 50 years old, Corresponding parameter is x5i, x6i, x7i, x8i；Annual family income is less than 2000 RMB and more than 20000 RMB, corresponding parameter x9i、x10i；The existing automobile of family, the coming five years can buy automobile, the following automobile and future of buying for 10 years will not buy automobile, corresponding ginseng Number is x11i、x12i、x13i、x14i；Trip purpose is work, go to school, do shopping, going home and other, corresponding parameter is x15i、x16i、 x17i、x18i、x19i；Trip purpose respectively works, gone to school, doing shopping, going home, other trip distances, and corresponding parameter is x20i、 x21i、x22i、x23i、x24i、x25i；Go out the beginning-of-line density of population more than 0.023 people/square metre and less than 0.023 people/square metre, it is right It is x to answer parameter26i、x27i；The travel destination density of population be more than 0.023 people/square metre and less than 0.023 people/square metre, corresponding ginseng Number is x28i、x29i；Travel time is morning peak, and corresponding parameter is x30i；The magnitude of traffic flow of origin and destination, corresponding parameter is x31i；Its His information, corresponding parameter xki；I represents i-th part of questionnaire.
4. the electric bicycle travel time prediction method of the duration model according to claim 1 based on risk, Characterized in that, the step (4) comprises the following steps：
(41) the continuous time model held based on risk, travel time t cumulative distribution function F (t) are as follows：
<mrow> <mi>F</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <msubsup> <mo>&amp;Integral;</mo> <mn>0</mn> <mi>t</mi> </msubsup> <mi>f</mi> <mrow> <mo>(</mo> <mi>T</mi> <mo>)</mo> </mrow> <mi>d</mi> <mi>T</mi> <mo>=</mo> <mi>Pr</mi> <mrow> <mo>(</mo> <mi>T</mi> <mo>&amp;le;</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mn>1</mn> <mo>-</mo> <mi>Pr</mi> <mrow> <mo>(</mo> <mi>T</mi> <mo>&gt;</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mn>1</mn> <mo>-</mo> <mi>S</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>,</mo> <mn>0</mn> <mo>&lt;</mo> <mi>x</mi> <mo>&lt;</mo> <mi>&amp;infin;</mi> </mrow>
Wherein f (t) is travel time t probability density function, and S (t) is survival function, and it is general more than t that it provides the duration Rate, dangerous function h (t) give the probability for terminating stroke in time t, and condition is that the trip is not over before time t, Equation is as follows：
<mrow> <mi>h</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>lim</mi> <mrow> <mi>&amp;Delta;</mi> <mi>t</mi> </mrow> </munder> <mfrac> <mrow> <mi>Pr</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>&amp;le;</mo> <mi>T</mi> <mo>&amp;le;</mo> <mi>t</mi> <mo>+</mo> <mi>&amp;Delta;</mi> <mi>t</mi> <mo>|</mo> <mi>T</mi> <mo>&amp;GreaterEqual;</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mi>&amp;Delta;</mi> <mi>t</mi> </mrow> </mfrac> <mo>=</mo> <mfrac> <mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mn>1</mn> <mo>-</mo> <mi>F</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>=</mo> <mfrac> <mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mi>S</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>;</mo> </mrow>
(42) accelerated failure-time model is used to explain change in the duration model based on risk as dangerous function to explain The influence of amount, the expression formula of accelerated failure-time model are as follows：
h(ti)=h0(t×exp(-(β0+β1x1i+…βnxni)))×exp(-(β0+β1x1i+…βnxni))
Wherein h0 () is baseline hazard function, and n is the quantity of explanatory variable, and accelerated failure-time model can also be write as following Expression-form：
Ln (t)=β0+β1x1i+β2x2i+β3x3i+β4x4i+β5x5i+β6x6i+β7x7i+β8x8i+β9x9i+β10x10i+β11x11i+β12x12i+ β13x13i+β14x14i+β15x15i+β16x16i+β17x17i+β18x18i+β19x19i+β20x20i+β21x21i+β22x22i+β23x23i+β24x24i+ β25x25i+β26x26i+β27x27i+β28x28i+β29x29i+β30x30i+β31x31i+βkxki+θ+σεi
Wherein t represents travel time, εiIt is independent random error, σ is scale parameter, and θ is stochastic effects, βkIt is to be accordingly Number；
(43) dangerous function is estimated using Weibull distribution, the probability density function and dangerous function that Wei Buer is distributed are as follows：
<mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>|</mo> <mi>&amp;lambda;</mi> <mo>,</mo> <mi>p</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>&amp;lambda;</mi> <mi>p</mi> <msup> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mi>t</mi> <mo>)</mo> </mrow> <mrow> <mi>p</mi> <mo>-</mo> <mn>1</mn> </mrow> </msup> <msup> <mi>e</mi> <mrow> <mo>-</mo> <msup> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mi>t</mi> <mo>)</mo> </mrow> <mi>p</mi> </msup> </mrow> </msup> <mo>,</mo> <mi>h</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <msup> <mi>&amp;lambda;</mi> <mi>p</mi> </msup> <msup> <mi>pt</mi> <mrow> <mi>p</mi> <mo>-</mo> <mn>1</mn> </mrow> </msup> <mo>.</mo> </mrow>
5. the electric bicycle travel time prediction method of the duration model according to claim 4 based on risk, Characterized in that, the introducing of the stochastic effects is used for explaining the heterogeneity of whole sample.
CN201710958398.3A 2017-10-16 2017-10-16 The electric bicycle travel time prediction method of duration model based on risk Pending CN107679668A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710958398.3A CN107679668A (en) 2017-10-16 2017-10-16 The electric bicycle travel time prediction method of duration model based on risk

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710958398.3A CN107679668A (en) 2017-10-16 2017-10-16 The electric bicycle travel time prediction method of duration model based on risk

Publications (1)

Publication Number Publication Date
CN107679668A true CN107679668A (en) 2018-02-09

Family

ID=61140302

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710958398.3A Pending CN107679668A (en) 2017-10-16 2017-10-16 The electric bicycle travel time prediction method of duration model based on risk

Country Status (1)

Country Link
CN (1) CN107679668A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109243174A (en) * 2018-09-05 2019-01-18 昆明理工大学 A kind of mixing bicycle traffic wave calculation method based on spatial perception
CN110782706A (en) * 2019-11-06 2020-02-11 腾讯科技（深圳）有限公司 Early warning method and device for driving risk of intelligent vehicle
CN115830872A (en) * 2022-12-23 2023-03-21 东南大学 Method for judging influence of accident on bicycle use elasticity

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110082636A1 (en) * 2006-03-03 2011-04-07 Inrix, Inc. Dynamic time series prediction of future traffic conditions
CN103279802A (en) * 2013-04-17 2013-09-04 吉林大学 Method for predicting daily activity-travel time of commuter
US20150006071A1 (en) * 2013-06-27 2015-01-01 Sap Ag Mobile trip planner and live route update system
CN104766146A (en) * 2015-04-24 2015-07-08 陆化普 Traffic demand forecasting method and system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110082636A1 (en) * 2006-03-03 2011-04-07 Inrix, Inc. Dynamic time series prediction of future traffic conditions
CN103279802A (en) * 2013-04-17 2013-09-04 吉林大学 Method for predicting daily activity-travel time of commuter
US20150006071A1 (en) * 2013-06-27 2015-01-01 Sap Ag Mobile trip planner and live route update system
CN104766146A (en) * 2015-04-24 2015-07-08 陆化普 Traffic demand forecasting method and system

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
XU, CHENGCHENG 等: "Analyzing Travel Time and Frequency of E-bike Trips by Men and Women Using Hazard-Based Duration and Zero-Inflated Negative Binomial Models", 《TRANSPORTATION RESEARCH BOARD 94TH ANNUAL MEETING》 *
康国祥 等: "基于风险分析的交通事件持续时间预测", 《同济大学学报(自然科学版)》 *
李明: "基于风险模型的城市居民购物出发时间分布规律分析", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 *
隽志才 等: "基于持续时间模型的日出行时间分析", 《交通运输系统工程与信息》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109243174A (en) * 2018-09-05 2019-01-18 昆明理工大学 A kind of mixing bicycle traffic wave calculation method based on spatial perception
CN109243174B (en) * 2018-09-05 2021-11-19 昆明理工大学 Hybrid bicycle traffic wave calculation method based on spatial perception
CN110782706A (en) * 2019-11-06 2020-02-11 腾讯科技（深圳）有限公司 Early warning method and device for driving risk of intelligent vehicle
CN110782706B (en) * 2019-11-06 2022-05-06 腾讯科技（深圳）有限公司 Early warning method and device for driving risk of intelligent vehicle
CN115830872A (en) * 2022-12-23 2023-03-21 东南大学 Method for judging influence of accident on bicycle use elasticity
CN115830872B (en) * 2022-12-23 2023-09-01 东南大学 Method for judging elastic influence of accident occurrence on bicycle use

Similar Documents

Publication Publication Date Title
CN104217250B (en) A kind of urban rail transit new line based on historical data opens passenger flow forecasting
CN107895283A (en) A kind of businessman&#39;s volume of the flow of passengers big data Forecasting Methodology based on Time Series
CN107103392A (en) A kind of identification of bus passenger flow influence factor and Forecasting Methodology based on space-time Geographical Weighted Regression
CN113902011A (en) Urban rail transit short-time passenger flow prediction method based on cyclic neural network
CN107729938B (en) Rail station classification method based on bus connection radiation zone characteristics
CN113177657B (en) Rail transit passenger flow prediction method and device
Zhao et al. Identification of land-use characteristics using bicycle sharing data: A deep learning approach
CN108629457B (en) Travel prediction mode and method and device for building prediction model
CN108133302A (en) A kind of public bicycles potential demand Forecasting Methodology based on big data
CN107563540A (en) A kind of public transport in short-term based on random forest is got on the bus the Forecasting Methodology of the volume of the flow of passengers
CN106910199A (en) Towards the car networking mass-rent method of city space information gathering
CN107679668A (en) The electric bicycle travel time prediction method of duration model based on risk
CN109785618A (en) Short-term traffic flow prediction method based on combinational logic
Yang et al. The study of pedestrian accessibility to rail transit stations based on KLP model
CN104361255B (en) It is a kind of to improve cellular automata urban sprawl analogy method
de Jong et al. The driving factors of passenger transport
Faber et al. The role of travel-related reasons for location choice in residential self-selection
Liu et al. Understanding urban bus travel time: Statistical analysis and a deep learning prediction
CN103020733A (en) Method and system for predicting single flight noise of airport based on weight
Li et al. Influence of weather conditions on the intercity travel mode choice: A Case of Xi’an
CN107316096A (en) A kind of track traffic one-ticket pass passenger amount of entering the station Forecasting Methodology
Sun et al. The relative roles of different land-use types in bike-sharing demand: A machine learning-based multiple interpolation fusion method
Ma et al. Short-term subway passenger flow prediction based on gcn-bilstm
CN108053053A (en) The non-commuting travel forecasting approaches of electric bicycle based on space autoregression model
Gao et al. Method of Predicting Passenger Flow in Scenic Areas Considering Multisource Traffic Data.

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20180209

RJ01 Rejection of invention patent application after publication