CN108320514A - The analysis method that accident rate is influenced based on the Tobit highway route indexs returned - Google Patents
The analysis method that accident rate is influenced based on the Tobit highway route indexs returned Download PDFInfo
- Publication number
- CN108320514A CN108320514A CN201810300708.7A CN201810300708A CN108320514A CN 108320514 A CN108320514 A CN 108320514A CN 201810300708 A CN201810300708 A CN 201810300708A CN 108320514 A CN108320514 A CN 108320514A
- Authority
- CN
- China
- Prior art keywords
- accident rate
- section
- accident
- tobit
- vertical curve
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 61
- 230000000694 effects Effects 0.000 claims abstract description 32
- 206010039203 Road traffic accident Diseases 0.000 claims abstract description 26
- 230000005856 abnormality Effects 0.000 claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 21
- 238000007619 statistical method Methods 0.000 claims abstract description 8
- 230000001419 dependent effect Effects 0.000 claims description 13
- 238000010606 normalization Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 abstract description 6
- 238000009825 accumulation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000003556 assay Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Analytical Chemistry (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Chemical & Material Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Based on the analysis method that the Tobit highway route indexs returned influence accident rate, the present invention relates to the analysis methods that highway route index influences accident rate.When the purpose of the present invention is to solve using accident rate index analysis evaluation on traffic safety situation, the Accident analysis model of existing foundation be not applicable in, and when the phase of statistics is short or section divides shorter, the limited problem of accident rate value.Process is：One, it is based on linear index and section division is carried out to highway；Two, accident rate is calculated to all types of sections after division；Three, the accident rate of rejecting abnormalities value is obtained；Four, assignment is carried out to each independent variable；Five, three and four results are normalized；Six, parameter calibration is carried out to the expressway traffic accident rate analysis model returned based on Tobit, independent variable edge effect is asked based on parameter calibration result application Stata statistical analysis softwares.The present invention is used for expressway traffic accident rate impact analysis field.
Description
Technical field
The present invention relates to the analysis methods that highway route index influences accident rate.
Background technology
In traditional traffic accident analysis method, accident number, casualty accident number or accident casualty people are often used
The stochastic patterns variables such as number establish Accident analysis model, and Poisson distribution, negative binomial distribution and zero accumulation class Poisson usually can be used
Or negative binomial distribution is fitted hazard model.
However in practical applications, the report of most of Evaluation of Traffic Safety often use hundred million truck kilometer accident rates, million
The accident rates index such as vehicle accident rate describes traffic safety status, and accident rate achievement data is often more easy to obtain.In addition, accident rate sheet
Body just contains the accidents relation factor information such as the volume of traffic, road section length, and it is more objective to be used for assay traffic safety status.
When using accident rate index analysis evaluation on traffic safety situation, since accident rate belongs to random variable of continuous type, with
Accident number is different, therefore the thing established based on Poisson distribution, negative binomial distribution or the Poisson distribution of zero accumulation class, negative binomial distribution
Therefore analysis model be not applicable in.It is considered simultaneously when the phase of statistics is short or section divides shorter, it may in accident rate statistical data
There is a large amount of " 0 " value, at this time, it is believed that accident rate value is limited, belongs to limited dependent variable.
Invention content
When the purpose of the present invention is to solve using accident rate index analysis evaluation on traffic safety situation, existing foundation
Accident analysis model be not applicable in, and when counting that the phase is short or section divides shorter, accident rate value it is limited the problem of, and
It is proposed the analysis method influenced on accident rate based on the highway route index that Tobit is returned.
It is on the analysis method detailed process that accident rate influences based on the Tobit highway route indexs returned：
Step 1: carrying out section division to highway based on linear index；
Step 2: calculating accident rate to all types of sections after division；All types of section accident rates are expressway traffic accident
Rate analysis model dependent variable；
Step 3: handling obtained all types of section accident rate data, the accident rate of rejecting abnormalities value is obtained；
Step 4: selecting linear index and annual average daily traffic as the expressway traffic accident rate returned based on Tobit
The independent variable of analysis model, and assignment is carried out to each independent variable；
Step 5: the independent variable number after the assignment that the accident rate and step 4 of the rejecting abnormalities value obtained to step 3 obtain
According to being normalized；
Step 6: the accident rate and argument data of the rejecting abnormalities value after the normalized obtained according to step 5,
Parameter calibration is carried out to the expressway traffic accident rate analysis model returned based on Tobit, is based on parameter calibration result application Stata
Statistical analysis software seeks independent variable edge effect.
Beneficial effects of the present invention are：
The present invention is based on Tobit recurrence to propose a kind of method that analysis highway route index influences accident rate.
The present invention establishes Accident analysis model, influence of the analysis highway route index to accident rate, with biography using accident rate data
The Accident analysis model based on accident number of system is different；Solves existing use accident rate index analysis evaluation on traffic safety situation
When, the Accident analysis model of existing foundation is not in applicable problem.
The present invention is based on highway route indexs to the method for accidental rate analysis, can find out and hand over certain highway
The maximum road alignment factor of logical security implication, to which the section poor to alignment condition is administered into row major, to ensure
In expressway traffic safety control, consumption economic resources as few as possible and social resources maximize highway and hand over
The logical effect administered safely, is not influenced by section condition；Solves existing when the statistics phase is short or section divides shorter, accident
Rate value it is limited the problem of.
As shown in Table 7, the edge effect of annual average daily traffic is -0.043, illustrates the increasing with annual average daily traffic
Add, hundred million truck kilometer accident rates are gradually reduced.This also implies explanation, and accident number is not linearly increasing with the growth of the volume of traffic
, i.e., the growth rate of accident number will be less than the growth rate of the volume of traffic.The edge effect of road section length be 0.305, illustrate with
The increase of road section length, hundred million truck kilometer accident rates are increasing.The edge effect of horizontal curve curvature is 0.013, to the shadow of accident rate
Unobvious are rung, edge effect is minimum.The edge effect of vertical curve curvature is 0.070, illustrates the increase with vertical curve curvature, i.e.,
The reduction of radius of vertical curve, hundred million truck kilometer accident rates are increasing.The edge effect of longitudinal slope type is 0.135, and the longitudinal slope gradient is absolute
The edge effect of value is 0.100, illustrates that hundred million truck kilometer accident rate of descending section is higher than uphill way, and as the longitudinal slope gradient is exhausted
Growth to value, hundred million truck kilometer accident rates are consequently increased.
The absolute value of edge effect represents influence degree size of the condition to accident rate, wherein being influenced on accident rate
Maximum factor is road section length, and minimum is annual average daily traffic.Each influence factor is arranged from small to large according to influence degree
It is classified as annual average daily traffic, horizontal curve curvature, vertical curve curvature, the longitudinal slope gradient, longitudinal slope type, road section length.It follows that
This expressway traffic safety situation is influenced maximum to be road section length, i.e. road section length is longer, dangerous bigger.Next is
Longitudinal slope type, i.e. descending section are than uphill way danger.It is that the longitudinal slope gradient and gradient absolute value are bigger again, danger is got over
Greatly.Therefore, when carrying out traffic safety control to the highway, long downslope is administered first, can obtain maximum
Income.
Description of the drawings
Fig. 1 is flow chart of the present invention；
Fig. 2 is horizontal alignment section division methods schematic diagram of the present invention；
Fig. 3 is vertical alignment section division methods schematic diagram of the present invention.
Specific implementation mode
Specific implementation mode one：The highway route index of present embodiment returned based on Tobit is to accident rate shadow
Loud analysis method detailed process is：
Step 1: carrying out section division to highway based on linear index；
Step 2: calculating accident rate to all types of sections after division；All types of section accident rates are expressway traffic accident
Rate analysis model dependent variable；
Step 3: handle obtained all types of section accident rate data, obtain rejecting abnormalities value accident rate (because
Variable)；
Step 4: selecting appropriate linear index and annual average daily traffic as the highway returned based on Tobit
The independent variable of accidental rate analysis model, and assignment is carried out to each independent variable；
Step 5: the independent variable number after the assignment that the accident rate and step 4 of the rejecting abnormalities value obtained to step 3 obtain
According to being normalized；
Step 6: the accident rate and argument data of the rejecting abnormalities value after the normalized obtained according to step 5,
Parameter calibration is carried out to the expressway traffic accident rate analysis model returned based on Tobit, is based on parameter calibration result application Stata
Statistical analysis software seeks independent variable edge effect；
The independent variable edge effect that step 6 obtains is analyzed；
It is analyzed by the edge effect to the expressway traffic accident rate parameter of analytic model acquired, according to each linear item
The edge effect size of part factor arranges different alignment condition factors from big to small.The maximum alignment condition of edge effect
It is maximum to the expressway traffic safety influence degree, therefore in traffic safety control, improvement should be paid the utmost attention to and come
The alignment condition in forefront.
Specific implementation mode two：The present embodiment is different from the first embodiment in that：Line is based in the step 1
Shape index carries out section division to highway；Detailed process is：
The linear index is：Road section length (L), horizontal curve curvature (CH), vertical curve curvature (CV), longitudinal slope type (T) and
The longitudinal slope gradient (i)；
Horizontal alignment is divided into two class of linear section and Horizontal Curve Sections；
Vertical alignment is divided into the sections Zhi Po and vertical curve section；
The sections Zhi Po include two kinds of longitudinal slope types, respectively uphill way and descending section；
The division in vertical curve section is to be divided into vertical curve front half section and perpendicular song for boundary with knick point (point of slope change)
The line second half section；
If the tangent line longitudinal slope gradient of vertical curve front half section is timing, which is classified as vertical curve upward trend
Section；
If the tangent line top rake of vertical curve second half section is timing, which is classified as vertical curve upward trend
Section；
If the tangent line top rake of vertical curve front half section is negative, which is classified as vertical curve downhill path
Section；
If the tangent line top rake of vertical curve second half section is negative, which is classified as vertical curve downhill path
Section；
Horizontal alignment and vertical alignment are combined, following eight type section is obtained：
Road segment classification 1：Straight line-uphill way；
Road segment classification 2：Straight line-descending section；
Road segment classification 3：Straight line-vertical curve uphill way；
Road segment classification 4：Straight line-vertical curve descending section；
Road segment classification 5：Horizontal curve-uphill way；
Road segment classification 6：Horizontal curve-descending section；
Road segment classification 7：Horizontal curve-vertical curve uphill way；
Road segment classification 8：Horizontal curve-vertical curve descending section.
Other steps and parameter are same as the specific embodiment one.
Specific implementation mode three：The present embodiment is different from the first and the second embodiment in that：It is right in the step 2
All types of sections after division calculate accident rate；All types of section accident rates are expressway traffic accident rate analysis model dependent variable；
Detailed process is：
Accident rate selects hundred million truck kilometer accident rates, calculation formula as follows：
In formula, RjFor the accident rate on the j of section, secondary/hundred million truck kilometer；AmjIt is secondary for the accident number of the upper m of section j；
AADTmjFor the upper 1 year annual average daily traffic of section j, pcu/ days；LjFor the length of section j, km；N is statistic years.This
In j be all sections of all types sum, such as each type all includes 100 sections, and j is exactly 800.
Other steps and parameter are the same as one or two specific embodiments.
Specific implementation mode four：Unlike one of present embodiment and specific implementation mode one to three：The step 3
In obtained all types of section accident rate data are handled, obtain the accident rate (dependent variable) of rejecting abnormalities value；Specific mistake
Cheng Wei：
Contracting tail (Winsorize) processing is carried out to hundred million truck kilometer accident rates, is rejected 1% high (1% a height of be higher than 99%)
Accident rate extremum.
Other steps and parameter are identical as one of specific implementation mode one to three.
Specific implementation mode five：Unlike one of present embodiment and specific implementation mode one to four：The step 4
It is middle to select appropriate linear index and annual average daily traffic as the expressway traffic accident rate analysis model returned based on Tobit
Independent variable, and to each independent variable carry out assignment；Detailed process is：
The independent variable chosen based on the expressway traffic accident rate analysis model that Tobit is returned is annual average daily traffic
(AADT), road section length (L), horizontal curve curvature (CH), vertical curve curvature (CV), longitudinal slope type (T) and the longitudinal slope gradient (i), unit
Respectively pcu/ days, m, km-1、km-1, dimensionless and %；
Longitudinal slope categorical variable takes " 0 " or " 1 " value, belongs to dummy variable, and to distinguish climb and fall section, longitudinal slope type is divided into
Uphill way and descending section, uphill way take " 0 ", and descending section takes " 1 "；
The reason of selecting horizontal curve curvature and vertical curve curvature is can to facilitate the assignment of variable.Road segment classification 1 is to section class
(road segment classification 1 to road segment classification 4 is that plane and straight line combines section with vertical alignment to type 4, and radius of horizontal curve is unlimited at this time
Greatly, and horizontal curve curvature then can be taken as " 0 " value；) horizontal curve curvature takes " 0 "；Road segment classification 5 is to 8 horizontal curve curvature of road segment classification
The inverse for sweep of making even；
(in the section that longitudinal slope is combined with horizontal alignment, the radius of vertical curve of longitudinal gradient section is nothing to road segment classification 1,2,5,6
Limit is big, and vertical curve curvature then can be taken as " 0 ") vertical curve curvature takes " 0 "；Road segment classification 3,4,7,8 takes perpendicular for vertical curve curvature
The inverse of sweep；
In being divided in section, different road segment classifications has been adhered in uphill, downhill section separately, and therefore, the longitudinal slope gradient takes absolutely
To value.
Other steps and parameter are identical as one of specific implementation mode one to four.
Specific implementation mode six：Unlike one of present embodiment and specific implementation mode one to five：The step 5
In argument data after the obtained assignment of the accident rate of rejecting abnormalities value that obtains to step 3 and step 4 be normalized
Processing；
It is lateral comparison difference independent variable to the influence degree of accident rate, needs to eliminate the potential of respective characteristics of variables scale
It influences, therefore, it is necessary to the argument datas after the assignment obtained to step 4 to be normalized, and normalization formula is：
In formula,Value after being normalized for q class variables；XqFor q class variable original values；XqminIt is former for q class variables
Minimum value in initial value；XqmaxFor the maximum value in q class variable original values；P is independent variable species number, value 6；
The dependent variable of model, i.e. hundred million truck kilometer accident rates, should also be normalized；
The accident rate data (dependent variable) of the rejecting abnormalities value obtained to step 3 are normalized, i.e., to hundred million vehicles public affairs
In accident rate be normalized, normalization formula be：
In formula, X*Value after being normalized for hundred million truck kilometer accident rates；X is hundred million truck kilometer accident rate original values；XminIt is hundred million
Minimum value in truck kilometer accident rate original value；XmaxFor the maximum value in hundred million truck kilometer accident rate original values.
Other steps and parameter are identical as one of specific implementation mode one to five.
Specific implementation mode seven：Unlike one of present embodiment and specific implementation mode one to six：The step 6
The accident rate and argument data of rejecting abnormalities value after the middle normalized obtained according to step 5 are returned to being based on Tobit
The expressway traffic accident rate analysis model returned carries out parameter calibration, is based on parameter calibration result application Stata statistical analysis softwares
Seek independent variable edge effect；Detailed process is：
By all types of section accident rates and argument data importing Stata statistical analysis softwares after normalization, with normalizing
Dependent variable of all types of section accident rates as the expressway traffic accident rate analysis model returned based on Tobit after change, to return
One change after linear index and annual average daily traffic be used as the expressway traffic accident rate analysis model returned based on Tobit oneself
Variable carries out Tobit recurrence, and the expressway traffic accident rate parameter of analytic model calibration result for obtaining being returned based on Tobit (is become certainly
Amount)；
Independent variable edge effect is sought based on parameter calibration result application Stata statistical analysis softwares.
In fact, since the expressway traffic accident rate analysis model returned based on Tobit belongs to generalized linear model, model
Regression coefficient can not directly represent influence degree size of the independent variable to dependent variable, therefore also need to ask it to model independent variable
Respective edge effect；Edge effect can equally be solved by Stata softwares；
Other steps and parameter are identical as one of specific implementation mode one to six.
Beneficial effects of the present invention are verified using following embodiment：
Embodiment：
The embodiment for the analysis method that highway route index based on Tobit recurrence influences accident rate is specifically to press
It is carried out according to following steps：
The present embodiment has collected the casualty data and road alignment condition data of certain highway, is collected into 1557 altogether
Casualty data, and according to the geometric linear data of above-mentioned highway, using a kind of highway route returned based on Tobit
The analysis method that index influences accident rate, analysis influence maximum alignment condition to the expressway traffic safety situation.
Step 1: carrying out section division to highway based on linear index；
Section division is carried out to the highway based on geometric linear condition.It is bent that horizontal alignment is divided into linear section peace
Two class of part of path.The division of vertical curve be divided into vertical curve front half section and second half section using knick point as boundary, if front half section or after
Half section of tangent line longitudinal slope is to go up a slope, then half section of vertical curve is classified as vertical curve upward slope section, otherwise is classified as vertical curve lower slope section.
Horizontal and vertical alignment is combined, following eight kinds of sections can be obtained：Straight line-uphill way, straight line-descending section, straight line-vertical curve
Uphill way, straight line-vertical curve descending section, horizontal curve-uphill way, horizontal curve-descending section, on horizontal curve-vertical curve
Slope section, horizontal curve vertical curve descending section, number respectively road segment classification 1 to road segment classification 8.Section division result such as table 1
It is shown, 1082 sections are marked off altogether, these sections are exactly the sample for establishing accidental rate analysis model.
1 section division result of table
Step 2: calculating accident rate to all types of sections after division；
To the section after division, accident rate index calculating is carried out, accident rate selects hundred million truck kilometer accident rates.Calculation formula is shown in
Formula (1).
Step 3: handle obtained all types of section accident rate data, obtain rejecting abnormalities value accident rate (because
Variable)；
To reduce influence of the exceptional value of accident rate to evaluation result, hundred million truck kilometer accident rates are handled, reject 1%
High accident rate extremum.1033 effective links are finally obtained.After rejecting abnormalities value, accident rate on effective links and its
Linear data, traffic data statistical result are shown in Table 2.
2 accident rate of table and road alignment data statistics result
Step 4: selecting appropriate linear index and annual average daily traffic as the highway returned based on Tobit
The independent variable of accidental rate analysis model, and assignment is carried out to each independent variable；Detailed process is：
Longitudinal slope categorical variable takes " 0 " or " 1 " value, belongs to dummy variable, to distinguish climb and fall section, uphill way takes
" 0 ", descending section take " 1 ".Road segment classification 1 to road segment classification 4 is that plane and straight line combines section with vertical alignment, is put down at this time
Sweep is infinity, and horizontal curve curvature then can be taken as " 0 " value.Similarly, in the section that longitudinal slope is combined with horizontal alignment,
The radius of vertical curve of longitudinal gradient section is infinity, and vertical curve curvature then can be taken as " 0 ".The longitudinal slope gradient takes absolute value.It is linear
Index initial data is as shown in table 3, and the data after assignment are as shown in table 4.
3 linear index raw data sample of table
Note："-" is initial data null value item
Data instance after 4 linear index assignment of table
Step 5: the independent variable number after the assignment that the accident rate and step 4 of the rejecting abnormalities value obtained to step 3 obtain
According to being normalized；
It is lateral comparison difference independent variable to the influence degree of accident rate, needs to eliminate the potential of respective characteristics of variables scale
It influences, therefore, it is necessary to argument data is normalized.The dependent variable of model, i.e. hundred million truck kilometer accident rates, also should be into
Row normalized.Accident rate and each argument data statistical result are as shown in table 5 after normalization.
Accident rate and argument data statistical result after table 5 normalizes
Step 6: the accident rate and argument data of the rejecting abnormalities value after the normalized obtained according to step 5,
Parameter calibration is carried out to the expressway traffic accident rate analysis model returned based on Tobit, is based on parameter calibration result application Stata
Statistical analysis software seeks independent variable edge effect；
To the accident rate and accident impact factor data after normalization, parameter mark is carried out using Stata statistical analysis softwares
It is fixed, it the results are shown in Table 6.
6 model parameter calibration result of table
The result of calculation of edge effect is as shown in table 7.
7 independent variable edge effect result of calculation of table
Step 7: analyzing accidental rate analysis model parameter calibration result.
As shown in Table 6, in 1033 groups of data of peg model, 551 groups of untethered data, using " 0 " value as left side by
482 groups of the left side restricted data of limit limit.In addition to horizontal curve curvature and annual average daily traffic conspicuousness are poor, road section length,
Vertical curve curvature, longitudinal slope type, longitudinal slope gradient absolute value are significant under 95% confidence level, and probability value P is respectively less than
0.05。
As shown in Table 7, the edge effect of annual average daily traffic is negative, illustrates the increase with annual average daily traffic,
Hundred million truck kilometer accident rates are gradually reduced.This also implies explanation, accident number with the growth of the volume of traffic be not it is linearly increasing, i.e.,
The growth rate of accident number will be less than the growth rate of the volume of traffic.The edge effect of road section length is just, to illustrate with road section length
Increase, hundred million truck kilometer accident rates are increasing.Influence unobvious of the horizontal curve curvature to accident rate, edge effect are minimum.Perpendicular song
The edge effect of line curvature is just, to illustrate the increase with vertical curve curvature, the i.e. reduction of radius of vertical curve, hundred million truck kilometer accidents
Rate is increasing.Longitudinal slope type and the edge effect of longitudinal slope absolute value are just, to illustrate that hundred million truck kilometer accident rate of descending section wants high
In uphill way, and with the growth of longitudinal slope gradient absolute value, hundred million truck kilometer accident rates are consequently increased.
The absolute value of edge effect represents influence degree size of the condition to accident rate, wherein being influenced on accident rate
Maximum factor is road section length, and minimum is annual average daily traffic.Each influence factor is arranged from small to large according to influence degree
It is classified as annual average daily traffic, horizontal curve curvature, vertical curve curvature, the longitudinal slope gradient, longitudinal slope type, road section length.It follows that
This expressway traffic safety situation is influenced maximum to be road section length, i.e. road section length is longer, dangerous bigger.Next is
Longitudinal slope type, i.e. descending section are than uphill way danger.It is that the longitudinal slope gradient and gradient absolute value are bigger again, danger is got over
Greatly.Therefore, when carrying out traffic safety control to the highway, long downslope is administered first, can obtain maximum
Income.
The present invention can also have other various embodiments, without deviating from the spirit and substance of the present invention, this field
Technical staff makes various corresponding change and deformations in accordance with the present invention, but these corresponding change and deformations should all belong to
The protection domain of appended claims of the invention.
Claims (7)
1. the analysis method influenced on accident rate based on the Tobit highway route indexs returned, it is characterised in that：The side
Method detailed process is：
Step 1: carrying out section division to highway based on linear index；
The linear index is：Road section length L, horizontal curve curvature CH, vertical curve curvature CV, longitudinal slope type T and longitudinal slope gradient i；
Step 2: calculating accident rate to all types of sections after division；All types of section accident rates are expressway traffic accident rate point
Analyse model dependent variable；
Step 3: handling obtained all types of section accident rate data, the accident rate of rejecting abnormalities value is obtained；
Step 4: selecting linear index and annual average daily traffic as the expressway traffic accident rate analysis returned based on Tobit
The independent variable of model, and assignment is carried out to each independent variable；
Step 5: argument data after the assignment that the accident rate and step 4 of the rejecting abnormalities value obtained to step 3 obtain into
Row normalized；
Step 6: the accident rate and argument data of the rejecting abnormalities value after the normalized obtained according to step 5, to base
Parameter calibration is carried out in the expressway traffic accident rate analysis model that Tobit is returned, based on parameter calibration result application Stata statistics
Analysis software seeks independent variable edge effect.
2. the analysis method that accident rate is influenced based on the Tobit highway route indexs returned according to claim 1,
It is characterized in that：Section division is carried out to highway based on linear index in the step 1；Detailed process is：
The linear index is：Road section length L, horizontal curve curvature CH, vertical curve curvature CV, longitudinal slope type T and longitudinal slope gradient i；
Horizontal alignment is divided into two class of linear section and Horizontal Curve Sections；
Vertical alignment is divided into the sections Zhi Po and vertical curve section；
The sections Zhi Po include two kinds of longitudinal slope types, respectively uphill way and descending section；
The division in vertical curve section is to be divided into vertical curve front half section and vertical curve second half section using knick point as boundary；
If the tangent line longitudinal slope gradient of vertical curve front half section is timing, which is classified as vertical curve uphill way；
If the tangent line longitudinal slope gradient of vertical curve second half section is timing, which is classified as vertical curve uphill way；
If the tangent line longitudinal slope gradient of vertical curve front half section is negative, which is classified as vertical curve descending section；
If the tangent line longitudinal slope gradient of vertical curve second half section is negative, which is classified as vertical curve descending section；
Horizontal alignment and vertical alignment are combined, following eight type section is obtained：
Road segment classification 1：Straight line-uphill way；
Road segment classification 2：Straight line-descending section；
Road segment classification 3：Straight line-vertical curve uphill way；
Road segment classification 4：Straight line-vertical curve descending section；
Road segment classification 5：Horizontal curve-uphill way；
Road segment classification 6：Horizontal curve-descending section；
Road segment classification 7：Horizontal curve-vertical curve uphill way；
Road segment classification 8：Horizontal curve-vertical curve descending section.
3. the analysis side that the highway route index according to claim 1 or claim 2 returned based on Tobit influences accident rate
Method, it is characterised in that：Accident rate is calculated to all types of sections after division in the step 2；All types of section accident rates are height
Fast highway accident rate analysis model dependent variable；Detailed process is：
Accident rate selects hundred million truck kilometer accident rates, calculation formula as follows：
In formula, RjFor the accident rate on the j of section, secondary/hundred million truck kilometer；AmjIt is secondary for the accident number of the upper m of section j；AADTmj
For the upper 1 year annual average daily traffic of section j, pcu/ days；LjFor the length of section j, km；N is statistic years.
4. the analysis method that accident rate is influenced based on the Tobit highway route indexs returned according to claim 3,
It is characterized in that：Obtained all types of section accident rate data are handled in the step 3, obtain rejecting abnormalities value
Accident rate；Detailed process is：
The processing of contracting tail is carried out to hundred million truck kilometer accident rates, rejects 1% high accident rate extremum.
5. the analysis method that accident rate is influenced based on the Tobit highway route indexs returned according to claim 4,
It is characterized in that：Select linear index and annual average daily traffic public as the high speed returned based on Tobit in the step 4
The independent variable of road accidental rate analysis model, and assignment is carried out to each independent variable；Detailed process is：
The independent variable chosen based on the expressway traffic accident rate analysis model that Tobit is returned is annual average daily traffic AADT, road
Segment length L, horizontal curve curvature CH, vertical curve curvature CV, longitudinal slope type T and longitudinal slope gradient i, unit be respectively pcu/ days, m, km-1、km-1, dimensionless and %；
Longitudinal slope type is divided into uphill way and descending section, and uphill way takes 0, and descending section takes 1；
Road segment classification 1 to 4 horizontal curve curvature of road segment classification is taken as 0；Road segment classification 5 to 8 horizontal curve curvature of road segment classification is made even song
The inverse of line radius；
1,2,5,6 vertical curve curvature of road segment classification takes 0；3,4,7,8 vertical curve curvature of road segment classification takes the inverse of radius of vertical curve；
The longitudinal slope gradient takes absolute value.
6. the analysis method that accident rate is influenced based on the Tobit highway route indexs returned according to claim 5,
It is characterized in that：After the assignment that the accident rate and step 4 of the rejecting abnormalities value obtained to step 3 in the step 5 obtain
Argument data is normalized；
Argument data after the assignment obtained to step 4 is normalized, and normalization formula is：
In formula,Value after being normalized for q class variables；XqFor q class variable original values；XqminFor q class variable original values
In minimum value；XqmaxFor the maximum value in q class variable original values；P is independent variable species number, value 6；
The accident rate data of the rejecting abnormalities value obtained to step 3 are normalized, i.e., are carried out to hundred million truck kilometer accident rates
Normalized, normalization formula are：
In formula, X*Value after being normalized for hundred million truck kilometer accident rates；X is hundred million truck kilometer accident rate original values；XminFor hundred million vehicles public affairs
In minimum value in accident rate original value；XmaxFor the maximum value in hundred million truck kilometer accident rate original values.
7. the analysis method that accident rate is influenced based on the Tobit highway route indexs returned according to claim 6,
It is characterized in that：The accident rate of rejecting abnormalities value after the normalized obtained according to step 5 in the step 6 and from becoming
Data are measured, parameter calibration is carried out to the expressway traffic accident rate analysis model returned based on Tobit, is answered based on parameter calibration result
Independent variable edge effect is sought with Stata statistical analysis softwares；Detailed process is：
By all types of section accident rates and argument data importing Stata statistical analysis softwares after normalization, after normalization
Dependent variable of all types of section accident rates as the expressway traffic accident rate analysis model returned based on Tobit, with normalization
The independent variable of linear index and annual average daily traffic as the expressway traffic accident rate analysis model returned based on Tobit afterwards
Tobit recurrence is carried out, the expressway traffic accident rate parameter of analytic model calibration result returned based on Tobit is obtained；
Independent variable edge effect is sought based on parameter calibration result application Stata statistical analysis softwares.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810300708.7A CN108320514A (en) | 2018-04-04 | 2018-04-04 | The analysis method that accident rate is influenced based on the Tobit highway route indexs returned |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810300708.7A CN108320514A (en) | 2018-04-04 | 2018-04-04 | The analysis method that accident rate is influenced based on the Tobit highway route indexs returned |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108320514A true CN108320514A (en) | 2018-07-24 |
Family
ID=62896926
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810300708.7A Pending CN108320514A (en) | 2018-04-04 | 2018-04-04 | The analysis method that accident rate is influenced based on the Tobit highway route indexs returned |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108320514A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738591A (en) * | 2019-09-20 | 2020-01-31 | 哈尔滨工业大学（威海） | Method for calculating traffic safety benefit of climbing lane based on tendency value matching |
| CN111582707A (en) * | 2020-04-30 | 2020-08-25 | 华南理工大学 | Road safety analysis method and system based on three-dimensional space alignment of road |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101059851A (en) * | 2007-06-05 | 2007-10-24 | 天津市市政工程设计研究院 | Highway route evaluation method |
| CN101826258A (en) * | 2010-04-09 | 2010-09-08 | 北京工业大学 | Method for predicting simple accidents on freeways |
| CN103531023A (en) * | 2013-10-18 | 2014-01-22 | 北京世纪高通科技有限公司 | Data processing method and device |
| CN105374206A (en) * | 2015-12-09 | 2016-03-02 | 敏驰信息科技（上海）有限公司 | Active traffic demand management system and working method thereof |
| CN105608902A (en) * | 2016-03-28 | 2016-05-25 | 辽宁省交通科学研究院 | Expressway black spot identification system and method |
| CN106530171A (en) * | 2016-10-12 | 2017-03-22 | 长安大学 | Interchange type overpass security estimation method |
| CN107273340A (en) * | 2017-06-01 | 2017-10-20 | 南京邮电大学 | A kind of road traffic accident factor-analysis approach based on Logistic models |
-
2018
- 2018-04-04 CN CN201810300708.7A patent/CN108320514A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101059851A (en) * | 2007-06-05 | 2007-10-24 | 天津市市政工程设计研究院 | Highway route evaluation method |
| CN101826258A (en) * | 2010-04-09 | 2010-09-08 | 北京工业大学 | Method for predicting simple accidents on freeways |
| CN103531023A (en) * | 2013-10-18 | 2014-01-22 | 北京世纪高通科技有限公司 | Data processing method and device |
| CN105374206A (en) * | 2015-12-09 | 2016-03-02 | 敏驰信息科技（上海）有限公司 | Active traffic demand management system and working method thereof |
| CN105608902A (en) * | 2016-03-28 | 2016-05-25 | 辽宁省交通科学研究院 | Expressway black spot identification system and method |
| CN106530171A (en) * | 2016-10-12 | 2017-03-22 | 长安大学 | Interchange type overpass security estimation method |
| CN107273340A (en) * | 2017-06-01 | 2017-10-20 | 南京邮电大学 | A kind of road traffic accident factor-analysis approach based on Logistic models |
Non-Patent Citations (2)
| Title |
|---|
| PANAGIOTIS CH. ANASTASOPOULOS 等: "A multivariate tobit analysis of highway accident-injury-severity rates", 《ACCIDENT ANALYSIS AND PREVENTION》 * |
| 叶莲娜: "基于几何线形及路面状态的高速公路事故预测模型", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738591A (en) * | 2019-09-20 | 2020-01-31 | 哈尔滨工业大学（威海） | Method for calculating traffic safety benefit of climbing lane based on tendency value matching |
| CN111582707A (en) * | 2020-04-30 | 2020-08-25 | 华南理工大学 | Road safety analysis method and system based on three-dimensional space alignment of road |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109408557B (en) | Traffic accident cause analysis method based on multiple correspondences and K-means clustering | |
| CN104408916B (en) | Based on section speed, the road traffic running status appraisal procedure of data on flows | |
| CN101329734A (en) | License plate character recognition method based on K-L transform and LS-SVM | |
| CN103984939B (en) | A kind of sample visible component sorting technique and system | |
| CN109635852B (en) | User portrait construction and clustering method based on multi-dimensional attributes | |
| CN101763466B (en) | Biological information recognition method based on dynamic sample selection integration | |
| CN108320514A (en) | The analysis method that accident rate is influenced based on the Tobit highway route indexs returned | |
| CN112101159A (en) | Multi-temporal forest remote sensing image change monitoring method | |
| CN105931252A (en) | Ellipse rapid detection method based on geometric constraint | |
| CN106570076A (en) | Computer text classification system | |
| CN109191828B (en) | Traffic participant accident risk prediction method based on ensemble learning | |
| CN103500343A (en) | Hyperspectral image classification method based on MNF (Minimum Noise Fraction) transform in combination with extended attribute filtering | |
| CN112084716B (en) | Red tide prediction and early warning method based on eutrophication comprehensive evaluation | |
| CN103345575B (en) | A kind of data flow concept drift detection method and system | |
| CN111612334A (en) | Driving behavior risk rating judgment method based on Internet of vehicles data | |
| CN116612307A (en) | Solanaceae disease grade identification method based on transfer learning | |
| CN104732246B (en) | A kind of semi-supervised coorinated training hyperspectral image classification method | |
| CN117392853A (en) | Big data intelligent lane control system based on high in clouds | |
| CN113138979A (en) | High-speed service area truck service level evaluation method based on trajectory data | |
| CN109617864A (en) | A kind of website identification method and website identifying system | |
| CN105574363A (en) | Feature selection method based on SVM-RFE (Support Vector Machine-Recursive Feature Elimination) and overlapping degree | |
| CN109166093A (en) | A kind of detection method for image salient region | |
| CN116933139A (en) | Village classification method integrating membership value, grade and sequence | |
| CN111428064A (en) | Small-area fingerprint image fast indexing method, device, equipment and storage medium | |
| CN111415081A (en) | Enterprise data processing method and device |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180724 |
|
| WD01 | Invention patent application deemed withdrawn after publication |