Download presentation
Presentation is loading. Please wait.
Published byJoanna Lane Modified over 9 years ago
1
A Case Study on Traffic Violations in the City of Colombo Udara Perera Sandun Silva Oshada Senaweera Yogeswaran Akhilan Amani Subawickrama
2
Introduction Driving is very important for working, social life, entertainment, economic, recreational and other reasons Number of registered vehicles in Sri Lanka have risen from 3.1 million in 2007 to 5.6 million in 2014 During last two decade approximately 25,000 km road were added to the national grid. Violation in traffic laws are very common in Sri Lanka Traffic law violations are a contributing factor to the majority of road accidents that occur in Sri Lanka
3
Objectives of the study Identify the most frequent traffic law violations in Colombo Examining the factors that influence traffic law violations Identify the relationship between traffic law violations and other factors Build a suitable model to predict the probability of doing a traffic law violation
4
Data Collection Response variable Violation type Predictor variables Location, vehicle type, gender of the driver, age of the vehicle, time, number of passengers Target population Motor vehicles using the roads in Colombo area Sampling technique Stratified sampling based on the type of location
5
Data Collection (ctd)
6
Data collection method Observational study
7
Analysis and Interpretation Univariate Analysis
8
Analysis and Interpretation (ctd)
10
Relationship Analysis Relationship between violation and road type H 0 : There is no relationship between violation and road type H 1 : There is a relationship between violation and road type P value = 0.007 (<0.05) Reject H 0 at 5% significance level. Violation depends on the road type
11
Analysis and Interpretation (ctd) Relationship between violation and vehicle type H 0 : There is no relationship between violation and vehicle type H 1 : There is a relationship between violation and vehicle type P value = 0.048 (<0.05) Reject H 0 at 5% significance level. Violation depends on the vehicle type
12
Analysis and Interpretation (ctd) Relationship between violation and age of the vehicle H 0 : There is no relationship between violation and age of the vehicle H 1 : There is a relationship between violation and age of the vehicle P value = 0.622 (>0.05) Do not reject H 0 at 5% significance level. Violation is independent of the age of the vehicle
13
Analysis and Interpretation (ctd) Relationship between violation and gender of the driver H 0 : There is no relationship between violation and gender of the driver H 1 : There is a relationship between violation and gender of the driver P value = 0.047 (<0.05) Reject H 0 at 5% significance level. Violation depends on the gender of the driver
14
Analysis and Interpretation (ctd) Relationship between violation and time H 0 : There is no relationship between violation and time H 1 : There is a relationship between violation and time P value = 0.022 (<0.05) Reject H 0 at 5% significance level. Violation depends on the time
15
Analysis and Interpretation (ctd) Relationship between violation and number of passengers H 0 : There is no relationship between violation and number of passengers H 1 : There is a relationship between violation and number of passengers P value = 0.347 (>0.05) Do not reject H 0 at 5% significance level. Violation is independent of the number of passengers
16
Model Fitting Binary logistic regression model Logit (Pi) = -0.653+0.386 Time(1) – 0.857 Location(1) + 0.051 Location(2) – 0.127 Location(3) – 0.174 Location(4) Here, Pi = Probability of violating a traffic lawTime(1)= Peak hours Location(1)= One-wayLocation(2)= Two-way Location(3)= T-junctionLocation(4)= Cross junction Eg:- Consider a vehicle at a one way road in peak hours Logit (Pi) = -0.653+ 0.386(1)-0.857(1)+0.051(0)-0.127(0)-0.174()) Log (Pi/1-Pi) = -1.126 Pi/1-Pi = exp(-1.126) Pi = 0.245
17
Thank You
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.