38
Development of Accident Prediction Development of Accident Prediction Models for the Highways of Thailand Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Embed Size (px)

Citation preview

Page 1: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Development of Accident Prediction Models for the Development of Accident Prediction Models for the Highways of ThailandHighways of Thailand

Development of Accident Prediction Models for the Development of Accident Prediction Models for the Highways of ThailandHighways of Thailand

Lalita Thakali

Transportation Engineering

Page 2: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Outline of PresentationOutline of Presentation

Statements of Problem Objective Methodology Preliminary analysisPreliminary analysis Model developmentModel development Identification of hazardous Identification of hazardous

location location Conclusion Recommendation

Page 3: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Statement of Statement of ProblemsProblems

18%18% of the annual road accidents occurs in highway of Thailand.

Trend of accident in highways of Thailand

Budget allocation for road safety for highways of Thailand

Year Accidents Fatalities Injuries

Property Damage (THB)

2001 15,341 2,212 12,712 352,851,000

2002 15,066 2,265 13,285 445,236,000

2003 15,171 2,023 12,984 464,248,000

2004 18,547 2,324 18,381 425,623,000

2005 16,287 2,169 15,300 405,248,000

Year Budget (in Million Baht)2002 1,400.0002003 1,400.0002004 1,770.0002005 1,644.999

(The annual report (2005) of the Bureau of Traffic Safety ) (The annual report (2005) of the Bureau of Traffic Safety )

In 2002 the economic losses due to road accidents was estimated to be in approximately 115932 million baht, or 2.13%2.13% of the GDP

Page 4: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Causes of Accidents in Causes of Accidents in ThailandThailand

63%

1%

3% 6%

20%5%

Road & EnvironmentRoad & EnvironmentVehicleVehicle

HumanHuman

4%

Descriptive model1

Predictive modelPredictive model2

Risk model3

Accident consequences model

4

How to Address Road Safety Problem

By Accident Modeling

Page 5: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Objective of StudyObjective of Study

Identify existing accident characteristicaccident characteristic

To develop a generalized accident prediction models generalized accident prediction models for highways using different statistical techniques.

To identify hazardous locationshazardous locations

1

2

3

Page 6: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

MethodoloMethodologygy Literatures Review

Site Site SelectionSelection

Data Collection

•DoH historical Accident Data•DoH traffic data•Metrological data•Video data

Explanatory Variables(Xij)

Homogenous Section

i= 1,2….ni= 1,2….n

l1l1 l2l2 ln-1ln-1 lnln

Monthly accident data (λij)•Accident•Fatality•Injury•Property damage

Identification of possible Variables

Preliminary data analysis (Characteristic of accident

& severities

Page 7: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Site SelectionSite Selection

1. Why route no 4

YearAccidents

(highways)

Accidents (major route

1,2,3,4)

Accidents (route 4)

% of accidents

(major route)

% of accidents (route 4 w.r.t. major routes)

2001 15341 3228 800 21.04 24.783

2002 15066 3142 869 20.85 27.658

2003 15171 2982 949 19.66 31.824

2004 18547 3534 993 19.05 28.098

2005 16287 3016 861 18.52 28.548

2006 10597 2077 552 19.60 26.577

Average 15168 2997 837 19.79 27.91

2. Why route no 4 in Ratcha Buri & Nakhonpathom

Year Accident Fatalit

y Injury

Property Damage

2001 29.9 15.1 10.4 15.6

2002 23.5 10.0 8.6 13.8

2003 19.0 5.0 5.2 7.3

2004 34.4 11.4 24.0 10.3

2005 29.6 10.0 14.9 7.3

2006 29.2 3.0 22.3 5.9

Average 27.59 9.10 14.24 10.04

3. Relatively high no of AADT high no of AADT count stationscount stations

27.59%27.59% of accident occurs in selected site, where it covers only 8.56%8.56% of Total road length of route no 4

Page 8: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Study AreaStudy Area

NakhonpathNakhonpathomomRatcha Ratcha

BuriBuri

Total length = Total length = 117.93 km117.93 km

Route no 4

Page 9: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

MethodoloMethodologygy Literatures Review

Site Selection

Data Data CollectionCollection

•DoH historical Accident Data•DoH traffic data•Metrological data•Video data

Explanatory Variables(Xij)

Homogenous Section

i= 1,2….ni= 1,2….n

l1l1 l2l2 ln-1ln-1 lnln

Monthly accident data (λij)•Accident•Fatality•Injury•Property damage

Identification of possible Variables

Preliminary data analysis (Characteristic of accident

& severities

Page 10: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Data Data CollectionCollection

1. DoH historical Accident Data2. DoH traffic data3. Metrological data4. Video data

No of lane (4,1))Types of median

(4,1)Shoulders

available (4)No of curves (4)No of

intersection (4)No of access (4)

AADT (2) % of heavy

vehicle (2)

Rainfall (3)

Month

Total accident Fatality Injury Property damage

DOH (1)

Data: Year 2001 to 2006

2001- 2006

Page 11: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

MethodoloMethodologygy Literatures Review

Site Selection

Data Collection

•DoH historical Accident Data•DoH traffic data•Metrological data•Video data

Explanatory Variables(Xij)

Homogenous Section

i= 1,2….ni= 1,2….n

l1l1 l2l2 ln-1ln-1 lnln

Monthly accident data (λij)•Accident•Fatality•Injury•Property damage

Identification of Identification of possible Variables possible Variables

Preliminary data Preliminary data analysis analysis

(Characteristic of (Characteristic of accident & severitiesaccident & severities

Page 12: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Accident Rate Accident Rate (MVK)(MVK)

Countries

Canada France Germany Italy UK USA Bahrain Egypt Oman Yemen

0.01 0.02 0.02 0.01 0.01 0.001 0.002 0.44 0.04 0.11

Average Fatality rate in this study area is much higher than the rate in other countries. . seven seven times greater than that of Egypt

Fatality rate - 3.08 per MVK

Objective1

Page 13: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Causes of Causes of AccidentAccident Location of Location of

AccidentAccident

•Accident•Fatality•Injury•Property damage

Objective1

The exceeding of max speed is mostly due to the human-vehicle and its interaction with the geometric features of the road- this could be addressed in the model with the inclusion of geometric variables

Page 14: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Weather related Weather related AccidentAccident

Vehicle Vehicle involvementinvolvement

Vehicles Total %Pedestrian 21 1.04Bicycle 3 0.15Tricycle 2 0.10Motorcycle 392 19.34Trimotcycle 480 23.68Passenger car 605 29.85Light bus 59 2.91Light truck 98 4.83

Heavy vehicle (HV)Heavy bus 218 10.75

Medium truck 1 0.05Heavy truck 92 4.54Farm vehicle 56 2.76

Weather Accident % Fatality % Injury % PD %

Clear 1018 74 59 92 421 65 14295 87Fog 1 0 0 0 0 0 22 0Rain 125 9 0 0 40 6 1654 10Other 237 17 5 8 187 29 398 2Total 1381 100 64 100 648 100 16369 100

Surface Condition

Accident % Fatality % Injury % PD %

Dry 990 72 59 92 394 61 14045 86Dirty 1 0 0 0 0 0.00 27 0Wet 125 9 0 0 40 7 1628 10Other 265 19 5 8 214 33 669 4Total 1381 100 64 100 648 100 16369 100

Note: PD= Property Damage (1000 baht)

Objective1

HV only 16% of total number of accidents while it represents 22.39% of total traffic volume

Page 15: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Accident distribution based on Accident distribution based on month month

Objective1

Page 16: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

MethodoloMethodologygy Literatures Review

Site Selection

Data Collection

•DoH historical Accident Data•DoH traffic data•Metrological data•Video data

Explanatory Explanatory VariablesVariables

(Xij)(Xij)

Homogenous Homogenous SectionSection

i= 1,2….ni= 1,2….n

l1l1 l2l2 ln-1ln-1 lnln

Monthly accident data Monthly accident data (λij)(λij)•AccidentAccident•FatalityFatality•InjuryInjury•Property damageProperty damage

Identification of possible Variables

Preliminary data analysis (Characteristic of accident

& severities

Page 17: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Variables Units

AADT Number (*1000)

HV %

Lane Number

Length km

Access (A’) Number/km

Intersection(I’) Number/km

Curve (C’) Number/km

Rain (R) mm

Variables Category

Median (MD)

Divided (1)

Undivided (0)

Shoulder (S)

No (1)

Yes (0)

Month (M)Others (1)

April (0)

Addition of variablesForward selection

•Literatures•Preliminary analysis•Data availability

Variables Total Mean Std Units

Accident 1220 1.22 1.89 Number

Fatality 61 0.06 0.33 Person

Injury 578 0.54 1.68 Person

PD 15596 15.66 50.05Thousand baht

Variables (per month)Dependent Independent

AADT and lane is highly correlated, so lane has been excluded

Objective 2

Data from 2001- 2005 was used in model development

Page 18: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Variables AccidentFatalit

yInjury PD

AADT(*1000)

HVLength

(km)Lane A I C R

Dependent Variables

Accident 1                      

Fatality 0.217 1                    

Injury 0.65 0.251 1                  

PD 0.386 0.189 0.201 1                

Independent Variables

AADT 0.488 0.132 0.258 0.295 1              

HV 0.083 0.006 0.071 0.118 0.335 1            

Length0.149 0.035 0.109

-0.005

-0.148 -0.07 1          

Lane 0.461 0.117 0.238 0.363 0.899 0.379 -0.294 1        

A 0.536 0.145 0.487 0.075 0.276 -0.05 0.54 0.229 1      

I0.019 0.028 -0.048 0.014 -0.126

-0.172

0.587-

0.2470.367 1    

C0.141 -0.012 0.083

-0.128

0.113-

0.0840.249

-0.067

0.357 0.724 1  

R-0.025 -0.078 -0.077 0.002 0.052

-0.004

-0.023 0.053-

0.002-

0.019-0.02 1

Pearson Correlation

Page 19: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Variables Accident Fatality Injury PDAADT(*1000)

HV(%)

Length(km)

A' I' C' R

Dependent Variables

Accident 1                    

Fatality 0.217 1                  

Injury 0.650 0.251 1                

PD 0.386 0.189 0.201 1              

Independent Variables

AADT 0.487 0.132 0.258 0.295 1            

HV 0.090 0.008 0.074 0.122 0.341 1          

Length 0.149 0.035 0.109 -0.005 -0.148 -0.066 1        

A' 0.382 0.123 0.375 0.043 0.387 0.222 -0.243 1      

I' -0.196 -0.031 -0.131 -0.097 -0.183 -0.276 -0.402 0.100 1    

C' -0.074 -0.023 -0.051 -0.156 -0.158 -0.442 -0.229 -0.168 0.699 1  

R -0.025 -0.078 -0.077 0.002 0.052 -0.005 -0.023 0.021 -0.002 -0.024 1

Pearson Correlation

Page 20: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Methodology Methodology cont.cont.

Forward selection of variables

Model development•GLM- Poisson regression•GLM- NB regressionE(λ) = exp∑βjXij λ = accident per monthβj = parameter coefficientXi = explanatory variable

Is included variable significant? And is the

goodness of fit better?

Identification of hazardous location

•Accident Data 2006•(Visual validation)

Selection of model (Poisson or NB)

If yes•Continue to includeIf not

•Exclude the variable

Any explanatory variables

remaining?

Yes

No

Page 21: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Generalized Linear Generalized Linear Model ? Model ?

Risk model

Empirical Bayes

Accident Modeling

Descriptive model

Accident consequences Model

Predictive Predictive modelmodel

MultivariateMultivariate

Fuzzy Logic

Artificial Neural Network

Linear Model

GLM

Normal dis of accident with constant mean & variance

Only Poisson /Negative Binomial regression model- poisson trial

Accident- Normally follows poisson trial rather than binomial trial

Objective 2

Page 22: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

E(λ) =μ= ∑βiXi Linear model

η= ∑βiXi Generalized Linear Model

λ (i, t) = e∑βjXij

Link function used gives non negative value which comply with nature of accident.

Parameter is estimated by max likelihood method unlike OLS method.

Generalized Linear Model Generalized Linear Model cont. cont.

Objective 2

Link Function

Page 23: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Discuss about the goodness of fit

Significance of parameters

Estimation of parameters β

Maximum log likelihood method. SPSS (16)

95% confident interval

= standard deviationW = Wald value

Tests Formula Criteria PurposeLog likelihood (LR) test.

P value >0.05 To select Step

Deviance

Less the value better is the selected step

model

AIC “ “BIC “ “Total explained variation (R2

D)

Greater the value better is the model

To select either Poisson or NB

Objective 2

Significance & Goodness Significance & Goodness of testsof tests

Page 24: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Variables/goodness of

fit

Steps

1 2 3 4 5 6 7 8 9 10 11

Constant0.203

(0.000)- 0.656 (0.000)

- 1.13 (0.000)

-1.426 (0.000)

-1.019 (0.000)

-1.147 (0.000)

-1.132 (0.000)

-851 (0.000)

-739 (0.000)

-1.235 (0.000)

-1.189 (0.000)

AADT0.018

(0.000)0.019

(0.000)0.016

(0.000)0.017

(0.000)0.017

(0.000)0.016

(0.000)0.016

(0.000)0.016

(0.000)0.015

(0.000)0.015

(0.000)

Length0.063

(0.000)0.078

(0.000)0.079

(0.000)0.08

(0.000)0.101

(0.000)0.101

(0.000)0.097 (0.00)

0.112 (0.00)

0.112 (0.000)

Access0.118

(0.000)0.125

(0.000)0.126

(0.000)0.101(0.0

00)0.101

(0.000)0.101

(0.000)0.103

(0.000)0.103

(0.000)

HV-0.021 (0.000)

-0.025 (0.000)

-0.024 (0.000)

-0.024 (0.000)

-0.025 (0.000)

-0.017 (0.004)

-0.017 (0.004)

Median0.233

(0.083)

Shoulder0.639

(0.000)0.639

(0.000)0.601

(0.000)0.769

(0.000)0.769

(0.000)

Month-0.348 (0.000)

-0.348 (0.000)

-0.348 (0.000)

-0.338 (0.000)

Intersection-0.055 (0.380)

Curve0.298

(0.020)0.294

(0.020)

Rain 0.0 (0.059)

Deviance 2281 1719 1610 1491 1474 1471 1415 1392 1391 1382 1378Pearson-Chi 2886 1866 1724 1561 1531 1529 1533 1490 1490 1481 1479

LL -1796 -1515 -1460 -1401 -1392 -1391 -1363 -1351 -1351 -1346 -1344AIC 3593 3034 2927 2809 2795 2794 2738 2716 2717 2708 2707BIC 3598 3044 2942 2829 2819 2823 2767 2750 2756 2748 2751

LR ratio0

(0.000)562

(0.000)670

(0.000)790

(0.000)806

(0.000)810

(0.000)865

(0.000)889

(0.000)890

(0.000)899

(0.000)903

(0.000)R2 D 0.25 0.29 0.35 0.35 0.35 0.38 0.39 0.39 0.39 0.39

Detail forward selection procedure for Accident model (Poisson)

Page 25: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Variables/Goodness Tests

Accident Fatality Injury Property Damage (PD)*1000 baht

Poisson(1)

Negative Binomial

(2)

Poisson(3)

Negative Binomial

(4)

Poisson(5)

Negative Binomial

(6)

Poisson(7)

Negative Binomial(8)

Selected Step 10 10 11 11 11 11 10 10

Constant-1.235 (0.000)

-1.162 (0.027)-2.855 (0.000)

- 2.576 (0.000)

-2.153 (0.00)

-1.911 (0.000)

-0.449 (0.00) 2.100 (0.000)

AADT (1000) 0.015 (0.000) 0.014 (0.000) 0.016 (0.000) 0.017 (0.00) 0.015 (0.000) 0.013 (0.00) 0.018 (0.00) 0.031 (0.00)length (km) 0.112 (0.00) 0.121 (0.000) 0.102 (0.002) 0.093 (0.003) 0.148 (0.000) 0.155 (0.00) 0.087 (0.00) -0.056 (0.055)Access (per km) 0.103 (0.000) 0.115 (0.000) 0.097 (0.017) 0.104 (029) 0.210 (0.000) 0.224 (0.00) -0.115 (0.00) -0.21 (0.000)

HV (%)-0.017 (0.004)

- 0.017 (0.053) 

  

  -0.009 (0.00)  

Median   

-1.054 (0.030)

- 1.217 (0.008)

-0.506 (0.003)

-0.654 (0.004)

1.783 (0.00) 

Shoulder 0.769 (0.000) 0.884 (0.00) 0.963 (0.030) 0.791 (0.042) 0.561 (0.001)0.767

(0.0017)1.567 (0.00) 1.086 (0.00)

Month-0.348 (0.000)

- 0.552 (0.001)

-0.916 (0.000)

- 0.906 (0.011)

-0.714 (0.000)

-0.814 (0.00) -0.206 (0.00)  

Intersection (per km)             0.589 (0.00) -0.342 (0.00)Curve (per km) 0.298 (0.020) 0.333 (0.013)     0.43 (0.005) 0.381 (0.037) -1.427 (0.00) -0.33 (0.004)

Rain (mm)   

-0.003 (0.012)

- 0.005 (0.016)

-0.002 (0.000)

-0.003 (0.00)   

Deviance 1382 739 329 267 1459 841 36730 3182Scaled Deviance 1382 739 329 267 1459 841 36730 3182Pearson Chi-Square 1481 751 1567 1480 3494 2459 96300 9800Scaled Pearson 1481 751 1567 1480 3494 2459 96300 9800LL -1346 -1332 -211 -200 -980 -793 -19300 -3092AIC 2708 2680 441 417 1978 1604 38620 6198BIC 2748 2719 485 456 2023 1648 38670 6232

LR ratio

899 (0.000)

386(0.000)

72(0.000)

66(0.000)

726(0.000)

410(0.000)

23509(0.000)

1350(0.000)

R2D 0.39 0.34 0.18 0.19 0.33 0.32 0.39 0.29

Page 26: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Prediction Prediction ModelsModels

Accident

Fatality

Injury

Property Damage

Objective 2

Unit: per month

Page 27: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Multiplier Multiplier FactorsFactors

Objective 2

Annual Average Daily Traffic Length

The factor is computed for its changes in magnitude of each predicting variables while considering all the other variables to be constant

Percent of heavy vehicle

No of Access per km

Page 28: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Objective 2

Median

Shoulder

Multiplier factors Multiplier factors cont.cont.

Intersection

Intersection

Page 29: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Objective 2

No of Curve per km

Rain fall

Multiplier factors Multiplier factors cont.cont.

Page 30: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Methodology Methodology cont.cont.

Forward selection of variables

Model development•GLM- Poisson regression•GLM- NB regressionE(λ) = exp∑βjXij λ = accident per monthβj = parameter coefficientXi = explanatory variable

Is included variable significant? And is the

goodness of fit better?

Identification of hazardous Identification of hazardous locationlocation

•Accident Data Accident Data 20062006•(Visual validation)(Visual validation)

Selection of model (Poisson or NB)

If yes•Continue to includeIf not

•Exclude the variable

Any explanatory variables

remaining?

Yes

No

Page 31: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Comparative study : Actual vs Model Comparative study : Actual vs Model prediction prediction

Predicting Variables

Mean Standard Deviation Critical Frequency

Actual Model Actual Model Actual Model

Accident 0.42 1.65 0.76 1.20 1.18 2.85

Fatality 0.01 0.11 0.07 0.12 0.07 0.23

Injury 0.33 0.94 0.74 1.32 1.06 2.26

PD 1.83 22.79 7.20 23.13 9.03 45.92

Predicting Variables

Mean Standard Deviation Critical Rate

Actual Model Actual Model Actual Model

Accident 3.51 13.63 6.48 10.21 9.99 23.84

Fatality 0.07 0.94 0.54 1.01 0.62 1.95

Injury 2.78 7.86 6.24 11.33 9.02 19.19

PD 14.77 183.38 57.08 182.15 71.85 365.52

Visual validation

Total road section of 26.98 km

Road section divided into constant length of 2km, with few less then 2 km.

Page 32: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Control

Section

Chainage

AADTLengt

h(km)

A’ HV MD S I’ C’From To

201.1.1

26+420

27+700

117.187 1.28

12.5 36.1 1 No 0 0

201.1.2

27+700

29+700

117.187 2 12 36.1 1 No 1 0.5

201.2.1

29+700

31+700

117.187 2 5 36.1 1 Yes 1 1

201.2.2

31+700

33+700

117.187 2 4 36.1 1 Yes 0.5 0.5

201.2.3

33+700

35+700

117.187 2 3.5 36.1 1 Yes 0.5 1

201.2.4

35+700

37+700

117.187 2 2.5 36.1 1 Yes 1 0.5

201.2.5

37+700

39+700

117.187 2 3 36.1 1 Yes 0.5 0.5

201.2.6

39+700

41+700

117.187 2 3 36.1 1 Yes 0 0

202.1.1

41+700

43+830

126.068 2.13 3.28

25.14 1 No

1.406 1

202.1.2

43+830

45+830

126.068 2 2.5

25.14 1 No 1 0.5

202.1.3

45+830

47+830

126.068 2 2

25.14 1 No 2 0.5

202.1.4

47+830

49+830

126.068 2 3

25.14 1 No 1 0.5

202.1.5

49+830

51+830

126.068 2 2

25.14 1 No 0.5 0.5

202.2.1

51+830

53+830

126.068 1.57 1.28

25.14 1 Yes

1.276

0.319

Control Section

Month

Chainage

AADTLength

(km)

Hazardous location

From ToFrequency Rate

201.1.1 Jan- Dec 26+420 27+700 117.2 1.28 Yes Yes

201.1.2 Jan- Dec 27+700 29+700 117.2 2 Yes Yes

202.1.1 April 41+700 43+830 126.1 2.13 Yes YesObjective 3

Hazardous Locations for Hazardous Locations for accidentaccident

Page 33: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Hazardous Locations for Hazardous Locations for accidentaccident

Objective 3

NakhonpathNakhonpathomom

Jan- DecJan- Dec

AprilApril

Page 34: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Accident trend is highly dependent on the exposure factors (MVK). 76% of accidents - exceeding of speed limit. Light vehicles have comparatively greater influence to the accidents than the HV. April has higher trend of accident and its severity than in rest of the months.

S.N

Variables

Total Explained variation (%)

Poisson Negative Binomial

1 Accident

3939 34

2 Fatality 18 19193 Injury 3333 324 Propert

y Damage

3939 29

AADT (1,2,3,4)Length (1,2,3,4)Access per km (1,2,3,4)HV % (1,4)Median (2,3)Shoulder (1,2,3,4Month (1,2,4)Intersection per km (4)Curve per km (1,3,4)Rainfall (2,3)

Model Model DevelopmentDevelopment

Characteristic of Characteristic of AccidentsAccidents

ConclusionsConclusions

Significant Significant variablesvariables

Page 35: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

•Total explanatory variation is not surprising as data excludes detail station of traffic count, detail geometric data like lane width, shoulder width and the human behaviors. Comparable with to Caliendo et al. (2007).

•The variables on the different severity of accident comply with the preliminary analysis. i.e. methodology implemented for the model formulation is appropriate one.

Conclusions Conclusions cont.cont.

Identification of hazardous Identification of hazardous locationlocationUsing the accident prediction models as the tool for the identification of hazardous section, the road sections with high traffic volume, high number of curves per km and absence of shoulder were found to be hazardous.

Page 36: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

RecommendatiRecommendationsons

From preliminary analysis & the models accident is prominent in April. Hence, more instant safety measures would be taken to reduce the numbers of accidents during this period which would safe both huge life and economic losses.

Accident is enhanced by the light vehicles as depicted in the result. Traffic management enforcing the rules and regulation would be implemented such as provision of separate lanes.

The developed accident prediction models would be integrated with the GIS tools and develop interface that would explicitly present the hazardous road sections.

Future Future ResearchesResearchesDevelop separate accident prediction models for intersection.

Develop model with inclusion of more detail geometric data like width of lane, width shoulder, speed limit etc.

Page 37: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering

Recommendations Recommendations cont.cont.

Separate accident prediction model, such as for vehicle to vehicle collision, vehicle turn over etc.

Real time crash prediction model would be developed for the link and intersection provided the data availability is real time.

The real time crash prediction would be integrated with the simulation package in the network for the traffic assignment with the safety factor with addition to the delay factors.

The real time traffic model would be integrated with GIS or Google earth to display the risk of particular section.

Page 38: Development of Accident Prediction Models for the Highways of Thailand Lalita Thakali Transportation Engineering