ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
UNIT-II Notes
Accident Data Collection and Analysis
Importance of Accident Data Collection
Helps identify patterns and trends in road crashes.
Supports policy-making, road safety strategies, and infrastructure
improvements.
Assists in legal investigations and insurance claims.
Provides basis for road user education and awareness programs.
Causes of Road Accidents (Common Factors)
➤ Human Factors
Over-speeding
Drunk driving or substance abuse
Distracted driving (e.g., mobile use)
Fatigue or drowsy driving
Disobeying traffic rules
Lack of driving experience or poor judgment
➤ Vehicle Factors
Mechanical failures (e.g., brake failure, tire burst)
Poor vehicle maintenance
Defective lights or indicators
➤ Road and Environmental Factors
Poor road conditions (potholes, lack of signage)
Sharp curves or blind spots
Bad weather (rain, fog, snow)
Inadequate lighting at night
Poor traffic management or lack of signals
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
Recording of Crash Data
➤ What Data is Collected
Date, time, and location of the accident
Type of crash (e.g., head-on, rear-end, side-impact)
Number of vehicles and people involved
Injuries or fatalities
Vehicle details (type, registration, condition)
Road conditions at the time of the crash
Weather conditions
Witness statements, photos, and sketches
Driver information (license, sobriety, actions)
➤ Data Collection Sources
Police reports
On-site crash investigations
Hospital and emergency services records
CCTV or dashcam footage
Witness accounts
Vehicle black box (event data recorder)
Crash Data Analysis Techniques
Statistical analysis (mean, trends, frequency)
Geospatial analysis using GIS tools (accident-prone zones)
Root cause analysis (human, mechanical, environmental)
Regression and predictive modelling
Heat maps and cluster analysis for high-risk areas
Before-and-after studies to evaluate interventions
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
Importance of Statistical Methods in Accident Analysis
Helps identify patterns and trends in road accidents.
Supports decision-making for road safety improvements.
Assists in risk assessment and accident prediction.
Provides a quantitative basis for policy formulation.
Enhances resource allocation for road safety measures.
Common Statistical Methods Used in Accident Analysis
🔹 Descriptive Statistics
Mean, median, mode – To understand average accident rates.
Standard deviation & variance – To measure data dispersion.
Frequency distribution – To analyze accident occurrences by type,
location, or time.
Cross-tabulation – To compare accident factors (e.g., vehicle type vs.
accident severity).
🔹 Regression Analysis
Linear regression – Examines the relationship between accident
frequency and influencing factors (e.g., speed vs. crash severity).
Logistic regression – Used to predict the likelihood of an accident
under specific conditions.
Poisson regression – Used for analyzing accident count data over a
time period or location.
🔹 Time Series Analysis
Identifies trends and seasonal variations in accident occurrences.
Used for forecasting future accident rates.
Methods include moving averages, ARIMA models.
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
🔹 Spatial Analysis (GIS-Based Methods)
Heat maps – Identifies accident hotspots.
Cluster analysis – Detects areas with high accident densities.
Kernel Density Estimation (KDE) – Estimates accident risk zones.
🔹 Hypothesis Testing
Chi-square test – Checks relationships between categorical variables
(e.g., weather conditions vs. accident severity).
t-tests and ANOVA – Compare accident data across different groups
(e.g., urban vs. rural roads).
🔹 Survival Analysis
Analyzes time until an accident occurs under given conditions.
Useful for studying driver behaviour, road safety improvements.
🔹 Machine Learning & Predictive Modelling (Advanced Methods)
Decision trees & random forests – Predict accident-prone areas.
Neural networks – Detect complex accident patterns.
Bayesian networks – Estimate accident risks based on multiple
factors.
Applications of Statistical Analysis in Accident Data
Identifying high-risk zones and accident-prone locations.
Evaluating the effectiveness of road safety measures.
Predicting accident severity and frequency.
Supporting traffic management and law enforcement.
Improving vehicle safety regulations
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
What is a Black Spot?
A black spot is a location on a road where accidents occur
frequently.
These spots are identified based on accident history, often involving
fatal or serious injuries.
Typically defined as a stretch of about 500 meters where more than
a specified number of accidents (e.g., 10 per year) have occurred.
Objectives of Black Spot Identification
Detect accident-prone areas for targeted improvements.
Reduce fatalities and injuries by implementing safety measures.
Optimize resource allocation for road maintenance and redesign.
Support data-driven decision-making for traffic safety policies.
Criteria for Identifying Black Spots (Common Standards)
Number of accidents over a time period (e.g., 3 or more fatal crashes
in 3 years).
Crash severity – injuries or fatalities carry more weight.
Crash rate per vehicle-km or per unit of traffic volume.
Frequency index or accident density over road segments.
Data Required for Black Spot Identification
Crash data (type, severity, time, date, location).
Traffic volume data (AADT – Average Annual Daily Traffic).
Road geometry details (curves, intersections, gradients).
Environmental conditions (weather, visibility, lighting).
Vehicle and driver information.
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
Methods of Black Spot Identification
🔹 Accident Frequency Method
Spots with highest number of accidents over a period are selected.
🔹 Accident Rate Method
Considers number of accidents relative to traffic volume.
🔹 Weighted Severity Index (WSI)
Assigns weights to accidents based on severity (e.g., fatal > injury >
damage-only).
🔹 Empirical Bayes Method
Combines actual crash data with expected values from similar sites
to reduce bias.
🔹 GIS and Spatial Mapping
Heat maps and cluster analysis to visualize accident concentrations.
Black Spot Investigation Process
1. Preliminary Data Review
Analyze crash records and patterns.
2. Site Inspection
Physical survey of road geometry, signage, visibility, and traffic
conditions.
3. Cause Identification
Determine contributing factors (human, vehicle, road,
environmental).
4. Stakeholder Consultation
Input from police, engineers, local authorities, and public.
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
5. Propose Remedial Measures
Engineering (e.g., widening roads, adding signs), enforcement,
or educational.
6. Implementation and Monitoring
Apply solutions and evaluate their effectiveness over time.
Examples of Remedial Measures
Installation of traffic signals or speed breakers.
Improved signage and road markings.
Better lighting and visibility enhancements.
Intersection redesign or road realignment.
Pedestrian crossings or footbridges.
What is Accident Reconstruction?
A scientific process of analyzing and recreating the events leading to,
during, and after a road accident.
Aims to determine the cause, contributing factors, and dynamics of
the crash.
Commonly used in legal cases, insurance claims, and road safety
analysis.
Objectives of Accident Reconstruction
Determine vehicle speeds, point of impact, and movement paths.
Analyze driver actions/reactions (e.g., braking, steering).
Identify violations of traffic rules or mechanical failures.
Assess contributing factors (road, weather, vehicle, human).
Reconstruct the sequence of events leading to the collision.
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
Data Required for Reconstruction
Scene evidence: skid marks, debris, vehicle rest positions.
Vehicle damage and deformation.
Tire marks, gouge marks, and fluid spills.
Measurements of the accident scene (distances, angles).
Weather and lighting conditions at the time of the crash.
Eyewitness accounts, CCTV/dashcam footage, photos.
Vehicle specs: weight, brake system, tire type, etc.
Techniques Used in Accident Reconstruction
🔹 Skid Mark Analysis
Used to estimate vehicle speed before braking.
Depends on coefficient of friction, skid length, and road surface.
🔹 Momentum and Energy Analysis
Conservation of momentum helps determine pre-collision speeds.
Kinetic energy used to analyze damage and impact forces.
🔹 Time-Distance Analysis
Evaluates reaction time, braking time, and stopping distance.
🔹 Photogrammetry
Use of photos and videos to extract measurements and recreate the
scene.
🔹 Simulation Software
Tools like PC-Crash, HVE, or Virtual CRASH simulate crash events.
Helps visualize collisions in 3D and test various scenarios.
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
🔹 Event Data Recorders (EDR) / "Black Box"
Records vehicle speed, braking, steering, and airbag deployment at
the time of crash.
Steps in Accident Reconstruction Process
1. Scene Documentation
Photograph and map the site, measure evidence.
2. Evidence Collection
Vehicle inspection, witness interviews, data retrieval.
3. Data Analysis
Apply physics, math, and software tools.
4. Modelling and Simulation
Recreate accident scenario digitally or mathematically.
5. Conclusion and Reporting
Prepare reports, charts, diagrams; present findings in court if
needed.
Applications of Accident Reconstruction
Legal proceedings and expert witness testimony.
Insurance dispute resolution.
Driver training and public awareness.
Design improvements in roads and vehicles.
Enforcement of traffic laws and understanding crash patterns.
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
Importance of Computer-Based Analysis
Enhances accuracy, speed, and efficiency in analyzing accident
data.
Helps in managing large datasets effectively.
Facilitates data visualization, pattern detection, and predictive
modeling.
Supports automated reporting and decision-making for road safety
policies.
Applications of Computer Analysis in Accident Data
🔹 Database Management
Use of relational databases (e.g., MySQL, MS Access) to store and
manage accident records.
Enables easy retrieval, sorting, filtering, and updating of crash
data.
🔹 Statistical Analysis Software
Tools like SPSS, R, Python, Excel used for:
Descriptive statistics (mean, median, trends)
Regression analysis (linear, logistic, Poisson)
Hypothesis testing (t-test, chi-square)
Time-series forecasting
🔹 Geospatial Analysis (GIS)
Identifies accident hotspots and black spots.
Maps spatial patterns of crashes.
Tools: ArcGIS, QGIS, Google Earth.
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.
ESSENTIALS OF ROAD SAFETY ENGINEERING (OE 805 CE)
🔹 Accident Simulation and Reconstruction Software
Recreates crash scenarios to determine causes and effects.
Tools include PC-Crash, Virtual CRASH, HVE (Human-Vehicle-
Environment).
Useful in legal investigations and vehicle safety analysis.
🔹 Data Visualization Tools
Create graphs, charts, heat maps, and dashboards.
Tools: Tableau, Power BI, Python (Matplotlib, Seaborn), Excel.
Makes data more understandable for stakeholders and policymakers.
🔹 Machine Learning & AI Applications
Predictive modelling of accident risks.
Pattern recognition for recurring crash conditions.
Used to develop intelligent transportation systems (ITS).
Tools: Python, RapidMiner.
🔹 Automated Reporting Systems
Generate periodic accident reports and safety audits.
Integrates data from police, hospitals, insurance, etc.
🔹 Decision Support Systems (DSS)
Help authorities in planning interventions and resource allocation.
Combine crash data with road, traffic, and environmental data.
Benefits of Computerized Accident Analysis
Increases data accuracy and reduces manual errors.
Facilitates real-time data analysis and monitoring.
Enables multi-factor analysis for comprehensive insights.
Aids in formulating data-driven safety measures.
Supports research, policy, and urban planning.
Prepared by Ashveen Kumar Puppala, Matrusri Engineering College, Hyderabad.