ERSE Unit II

Accident data collection and Analysis

Causes of Road Accidents

Road accident is most unwanted thing to happen to a road user, though they happen quite
often. The most unfortunate thing is that we don't learn from our mistakes on road. Most of
the road users are quite well aware of the general rules and safety measures while using
roads but it is only the laxity on part of road users, which cause accidents and crashes. Main
cause of accidents and crashes are due to human errors. We are elaborating some of the
common behaviour of humans which results in accident.

1.Over Speeding
2.Drunken Driving
3.Distractions to Driver
4.Red Light Jumping
5.Avoiding Safety Gears like Seat belts and Helmets
6.Non-adherence to lane driving and overtaking in a wrong manner

Various national and international researches have found these as most common behavior
of Road drivers, which leads to accidents.

Over Speeding:

Most of the fatal accidents occur due to over speeding. It is a natural psyche of humans to
excel. If given a chance man is sure to achieve infinity in speed. But when we are sharing
the road with other users we will always remain behind some or other vehicle. Increase in
speed multiplies the risk of accident and severity of injury during accident. Faster vehicles
are more prone to accident than the slower one and the severity of accident will also be
more in case of faster the severity of accident will also be more in case of faster vehicles.
Higher the speed, greater the risk. At high speed the vehicle needs greater distance to stop
i.e. braking distance. A slower vehicle comes to halt immediately while faster one takes
long way to stop and also skids a long distance due to law of notion. A vehicle moving on
high speed will have greater impact during the crash and hence will cause more injuries.
The ability to judge the forthcoming events also gets reduced while driving at faster speed
which causes error in judgment and finally a crash.

1
Drunken Driving:

Consumption of alcohol to celebrate any occasion is common. But when mixed with
driving it turns celebration into a misfortune. Alcohol reduces concentration. It decreases
reaction time of a human body. Limbs take more to react to the instructions of brain. It
hampers vision due to dizziness. Alcohol dampens fear and incite humans to take risks. All
these factors while driving cause accidents and many a times it proves fatal. For every
increase of 0.05 blood alcohol concentration, the risk of accident doubles. Apart from
alcohol many drugs, medicines also affect the skills and concentration necessary for
driving. First of all, we recommend not to consume alcohol. But if you feel your
merrymaking is not complete without booze, do not drive under the influence of alcohol.
Ask a teetotaler friend to drop you home.

Distraction to Driver:

Though distraction while driving could be minor but it can cause major accidents.

Distractions could be outside or inside the vehicle. The major distraction now a days is
talking on mobile phone while driving. Act of talking on phone occupies major portion of
brain and the smaller part handles the driving skills. This division of brain hampers reaction
time and ability of judgement. This becomes one of the reasons of crashes. One should not
attend to telephone calls while driving. If the call is urgent one should pull out beside the
road and attend the call. Some of the distractions on road are:

1.Adjusting mirrors while driving

2.Stereo/Radio in vehicle
3.Animals on the road
4.Banners and billboards.

The driver should not be distracted due to these things and reduce speed to remain safe
during diversions and other kind of outside distractions.

Red Light jumping:

It is a common sight at road intersections that vehicles cross without caring for the light.
The main motive behind Red light jumping is saving time. The common conception is that
stopping at red signal is wastage of time and fuel. Studies have shown that traffic signals
followed properly by all drivers saves time and commuters reach destination safely and
timely. A red light jumper not only jeopardizes his life but also the safety of other road
users. This act by one driver incites other driver to attempt it and finally causes chaos at
2
crossing. This chaos at intersection is the main cause of traffic jams. Eventually everybody
gets late to their destinations. It has also been seen that the red light jumper crosses the
intersection with greater speed to avoid crash and challan but it hampers his ability to judge
the ongoing traffic and quite often crashes.

Avoiding Safety Gears like seat belts and helmets:

Use of seat belt in four-wheeler is now mandatory and not wearing seat belt invites penalty,
same in the case of helmets for two wheeler drivers. Wearing seat belts and helmet has been
brought under law after proven studies that these two things reduce the severity of injury
during accidents. Wearing seat belts and helmets doubles the chances of survival in a
serious accident. Safety Gears keep you intact and safe in case of accidents. Two wheeler
deaths have been drastically reduced after use of helmet has been made mandatory. One
should use safety gears of prescribed standard and tie them properly for optimum safety.

Detrimental effects of traffic on environment

1. Safety 2. Noise 3. Land Consumption 4. Air Pollution 5. Degrading the Aesthetics

How different factors of Roads contribute in Accidents:

Drivers: Over-speeding, rash driving, violation of rules, failure to understand signs,

fatigue, alcohol.
Pedestrian: Carelessness, illiteracy, crossing at wrong places moving on carriageway,
Jaywalkers.
Passengers: Projecting their body outside vehicle, by talking to drivers, alighting and
boarding vehicle from wrong side travelling on footboards, catching a running bus etc.
Vehicles: Failure of brakes or steering, tyre burst, insufficient headlights, overloading,
projecting loads.
Road Conditions: Potholes, damaged road, eroded road merging of rural roads with
highways, diversions, illegal speed breakers.
Weather conditions: Fog, snow, heavy rainfall, wind storms, hail storms.

Preventive measures for accidents:

1. Education and awareness about road safety

3
 2. Strict Enforcement of Law
3. Engineering:
(a) Vehicle design (b) Road infrastructure

Direct Consequences of Accidents:

1 Fatality (Death) 2. Injury 3. Property Damage

Statistical Data Analysis:

 Accident Rate/km (R): Accident hazard is expressed as a number of

accidents of all types per Km of each highway and street classification.

R=A/L.

Where, A= Total no.of accidents in one year.

L= Length of control section in kms.

 Accident involvement rate: It is expressed as number of drivers of vehicles

with certain characteristics who are involved in accident per 100million
vehicle kilometers of travel.

R= (N/V) X 100000000

Where N= Total no.of drivers of vehicles involved in accidents during the period
of investigation.

V= Vehicle-kms of travel on road section during the period of

investigation.

 Death rate based on population: Traffic hazard to life in a community is

expressed as the no. of traffic fatalities for 100000 population. This reflects

4
 the accident exposure for entire area.

R= (B/P) X 100000.

Where, B= Total no.of accident deaths in one year.

P= Population of the area

Methods to Identify and prioritize Hazardous Locations and elements

The three approaches to identification of hazardous road locations:

(1) Traditional reactive accident-based approach, resulting in identification of

accident black spots

(2) State-of-the-art empirical Bayes method using accident prediction model,

which identifies critical locations, i.e. both real and potential black spots

(3) Proactive “preliminary” road safety inspection, identifying the risk factors,
which may potentially increase accident occurrence and severity.

Crash Analysis

5
Determine possible causes of crashes

6
DIAGNOSIS OF ROAD CRASH PROBLEMS

 Six steps in the diagnosis phase

 Study detailed crash reports
 Data sorting to determine groups of accident types and the locations
at which they occur
 Data amplification by detailed on-site investigation (perhaps
including conflict studies)
 Detailed analysis of all data
 Identification of dominant factors and/or Road features
 Determine nature of the crash problem


Crash Reduction Capabilities and Countermeasures

Black Spots

• At certain sites, the level of risk of road accidents is higher than the general
level of risk in surrounding areas.

• Crashes tend to be concentrated at these relatively high-risk locations. These

7
 locations with an abnormally high number of crashes are generally
described as black spots.

• 120 black spots are identified across our city and suburb of Hyderabad,
where 26 were located in GHMC limits.

• Areas included Yousufguda checkpost, chilkalguda crossroads, RTC cross

roads, NTR Garden Road etc. in the year 2019 for the month of July as per
Telangana Today, Times Now report.

• Rectifying works were made at these identified spots with a cost of 1.50
crore by the municipal corporation of Hyderaabd.

Accident Evaluation and Black Spot Investigation

• The accident data collection involves extensive investigation which

involves the following procedure:
8
Reporting: It involves basic data collection in form of two methods:

• Motorist accident report - It is filed by the involved motorist involved in all

accidents fatal or injurious.

• Police accident report - It is filed by the attendant police officer for all
accidents at which an officer is present. This generally includes fatal
accidents or mostly accidents involving serious injury required emergency
or hospital treatment or which have incurred heavy property damage

At Scene-Investigation: It involves obtaining information at scene such as

measurement of skid marks, examination of damage of vehicles, photograph of
final position of vehicles, examination of condition and functioning of traffic
control devices and other road equipment.

Technical Preparation: This data collection step is needed for organization and
interpretation of the study made. In this step measurement of grades, sight distance,
preparing drawing of after accident situation, determination of critical and design
speed for curves is done.

Professional Reconstruction: In this step effort is made to determine from

whatever data is available how the accident occurs from the available data. This
involves accident reconstruction which has been discussed under Section No.7 in
details. It is professionally referred as determining “behavioral” or “mediate”
causes of accident.

Cause Analysis: It is the effort made to determine why the accident occurred from
the data available and the analysis of accident reconstruction studies

Empirical Bayes Method

9
 • The Empirical Bayes method is used to estimate the expected long-term
crash experience, which is a weighted average of the observed crashes at
the site of interest and the predicted crashes from an SPF

• The empirical Bayes method assumptions are (1) a Poisson distribution for
the accidents, and (2) a Gamma distribution for the distribution of the
averages in the population of systems.

• With these two assumptions, the number of systems with k accidents must
obey the negative binomial distribution. The expected number of accidents,
a‟k, in the after period on a system that had k accidents in the before period
is

a'k =(k+1)*N‟(k + 1)/N‟

Accident Reconstruction

The purpose of reconstruction is to identify factors that are critical in a road

accident like pre-impact direction and velocities of colliding vehicles

To answer the questions “how” and “why” an accident happened. There are
numerous factors to be considered, all of which add up to an understanding of the
events that took place

10
Application of computer analysis of accident data

• TEAAS-Traffic Engineering Accident Analysis System: a Software used to

analyze the accident data ( North Carolina)

• FARS (Fatality Analysis Reporting System) and GES (General Estimates

System) UK

• Road Accident Data Management System (RADMS) India

 Data collected from accident sites covers several aspects and is subjective
to the person collecting the data.

 Identifying the cause of road accidents is the aim behind accident data
collection and reconstruction of the event with the main aim being reduction
damages caused by traffic accidents.

 Because of exponential growth in population leading to increased number

of vehicles on the road and consequently increasing accidents, the volume
of data from accidents has reached explosive proportions.

11
  In order to manage this humongous data and analyse it to make sense to
policy planners, data mining technologies are used 'WEKA' is a popular data
mining program that can handle huge sets of data efficiently

 The results of data mining will help organizations such as transportation,

to explore the accident data recorded by the police information

system, discover patterns to predict future behaviors and

effective decisions to be taken to reduce accidents.

 Road accidents are predicted through machine learning algorithms and

advanced techniques for analyzing information, such as convolutional
neural networks and long short-term memory networks, among other deep
learning architectures.
 Data sources for the road accident forecast is made.
 A classification is proposed according to its origin and characteristics, such
as open data, measurement technologies, onboard equipment and social
media data.
 Road accident forecasting and Traffic accident prediction are driven
by traffic engineering, data analysis and machine learning 

The main areas of interest of models obtained from computer analysis of

accident data are
 detection of problematic areas for circulation
 real time detection of traffic incidents
 road accident forecasting and
12
 prediction of the severity of the consequences suffered by involved in a
road accident

 Therefore, the study of road accident prediction is a field of relevant and

current scientific knowledge, open to innovation in the research of
algorithms and data analysis techniques that respond to the challenge
of

generating a more secure mobility environment, which considers

the pecularities of each country or region, i.e., traffic composition, weather
conditions, roads conditions, and demography.

 Data about accidents can be gathered by installing equipment on

vehicles, for example satellite positional systems (GPS, GLONASS,
Galileo), cameras and sensors, in order to gather data like acceleration,
unexpected braking events, sudden lane changes and information about the
driver behavior and status like drowsiness and level of stress

 Another emerging data source suitable for proposing models of road

accident prediction is social media

 Government data like police bodies, traffic police and road concessionaires
can be characterized as historical, since it contains data spanning several
decades, and can be considered as reliable, because it is supported by the
custody process of the entities responsible for the data.

13
  Open data can be defined, as the data that is produced and funded with
public money, that is made available and accessible without restriction to
the public .

 Road traffic information is usually one of the most available data.

 Measurement technologies include all kind of equipment that is part of the

road infrastructure, such as radar, cameras, or equipment embedded on the
road itself.

 By using analytic methods, researchers seek to characterize the

information and variables of the road accident, in order to discover hidden
patterns, profile behaviors, generate rules and inferences.
These patterns are useful to
 profile drivers or drivers‟ behavior on the road
 limit unsafe areas for driving
 generate classification rules related to road accident data
 perform selection of variables to be fetched in real-time model of accidents
and
 Select relevant variables to be used to train other methods, such as artificial
neural networks and deep learning algorithms.
 Clustering is a method of partitioning and grouping objects into groups
(clusters), so that objects grouped in each cluster share common
characteristics, while looking for them to be clearly different from
other objects grouped in other clusters.

14
 Common characteristics can be interpreted as the level of correlation of
objects according to the characteristics on which clustering techniques are
applied.

 Unlike classification methods, clustering does not require that the data be
previously marked with any particular category in order to distinguish
different groups within the data.
 The absence of these previous categories or classes indicates that
the objective of clustering is to find an underlying structure in the
information and achieve a more compact representation of it instead of
discriminating future data into categories.

 The main advantages of clustering algorithms are that they do not require
prior data processing, work well with large data sets, and their results can
be interpreted graphically.

 On the other hand, clustering algorithms are sensitive to the possibility of

finding a local maximum instead of a global maximum on their optimization
functions.

 Clustering algorithms use a distance function to calculate the similarity in

characteristics when they work with continuous elements and a measure of
similarity for data with qualitative elements.

 Among the techniques based on similarity functions we can include K-

15
 nearest neighbor and K-means clustering

 Cluster techniques whose similarity function is based on distribution

probabilities, their operation is based on the premise that each cluster has
an

underlying probability of distribution from which the data elements are

generated. An example of this type of algorithm is latent class clustering
(LCC)
 For data sets with attributes both qualitative and quantitative, clustering
techniques such as two-step clustering
 Batch clustering, in combination with fuzzy C-means and real time
clustering is used to study abrupt braking events in real time
 Batch clustering results, correlations were obtained that indicate
potentially dangerous places for driving, according to the time of day.
 K-means clustering and association rules model in order to determinate the
variables that influence the event of road accidents, obtaining a 6-cluster
model, which was used as an input to a rules association model.
 It was found by computer analysis of accident data that accident severity,
type of road, lighting present in the road and the type of surrounding
area were important factors in any accident
 Real-traffic data is used in order to predict the number of accidents on any
road or intersection and to identify risk factors using clustering to group
roads and finding risk patterns.
 The quantity of clusters was evaluated and selected using the Bayesian
information criterion (BIC)
 A decision tree builds classification models in the form of trees or

16
 dendrogram, each node represents one of the input variables, and each node
has several branches equal to the number of possible values of said input
variable.
 Decision trees are useful tools in pattern classification applications.
 Decision tree method of analysis is exploratory and not inferential.

 arners and classifiers do not require prior data processing and work well
with large data sets and rule learners and classifiers can be interpreted
graphically; however, their results are not as accurate
 Road Accident Data Management System (RADMS) is a Geographic
Information System (GIS) based software that is funded by world bank used
for collecting, comparing and analyzing road accident data.Currently, it is
being used by the government of Tamil Nadu.
 RADMS is a comprehensive traffic-management system which helps to
study and analyse traffic accidents in a scientific manner.
The various components of RADMS are:
 Creation of GIS database
 Web based access and data flow
 Report generation and plotting results on maps
 Analysis and identification of black-spots for police and
transport departments to take-up necessary measures 
 RADMS generates the following twelve types of reports for analysis
and suggestion of remedial measures 
 Driver report
 Vehicle report
 Road report
 Yearly report

17
  Enforcement
 Collision type
 Time period report
 Alcohol usage report

 Person report
 Landmark report
 Weather report
 General report
 RADaR is a robust road crash database in order to reduce road accidents.
RADaR is Road Accident Data Recorder
 RADaR is an end-to-end solution for road accident data recording and
reporting as it helps identify the factors contributing to road accidents
 RADaR is designed as a n application for android tablet with connectivity
to web-based database server.
 It used GPS/GPRS to record exact accident location in global coordinate
system and transmits data to web-based central server
 It also provides a facility to take photographs of the accident scene and
upload it to the network
 It features a pictorial menu-driven recording of road layout of crash site and
collision diagram plotted on layout for scientific investigation
 RADaR can draw data for vehicle registration and driver license
information from national databases
 The pilot studies for RADaR was carried out in New Delhi (India) and Addis
Ababa (Ethiopia)
 AI machine-learning method is used to create decision trees distinguishing

18
 the characteristics of accidents
 In order to identify factors causing accidents, Data Mining (DM) techniques
such as Decision Trees (DTs) that are used as they allow certain decision
rules to be extracted. These rules could be used in future road safety
campaigns thereby enabling managers to implement priority actions.

 Artificial Neural Network (ANN) models are used for the analysis and
prediction of accidents. In this technique, the number of vehicles, accidents,
and population are selected and used as model parameters. The sigmoid and
linear functions are used as activation functions with the feed forward-back
propagation algorithm.
 The ANN model has demonstrated to be better than statistical methods in
use.
 Since the data collected from accident sites is huge, it falls under the domain
of 'BIG DATA'.
 Traffic on highways is monitored and lots of data is processed daily to
predict probability of accidents based on highway conditions like road
surface, light on highway, turns etc.
 Accident prediction is based on different queries and in order to process this
big data, Hadoop has been used.
 Execution time is very less on Hadoop as compared to other sequential
techniques.

19