Method To Prioritize Hazardousness of Locations For Black Spot Identification Using Analytic Hierarchy Process
Method To Prioritize Hazardousness of Locations For Black Spot Identification Using Analytic Hierarchy Process
Method To Prioritize Hazardousness of Locations For Black Spot Identification Using Analytic Hierarchy Process
E-mail: amit.goel@pilani.bits-pilani.ac.in
ABSTRACT
A vehicular traffic accident analysis process should start with identifying the cause(s) of the
accident. The multitude of factors responsible for causing an accident, acting solitary or in
combination, include road, driver, vehicle, and environment characteristics. As such, the accident
analysis process assumes complexity in deciding the perceived domain, and relative importance
of the influencing factors. This paper presents a method to simplify this complexity by use of
analytic hierarchy process (AHP). Once this is done, AHP is further applied to identify the most
hazardous ones, called the black spots. The methodology is demonstrated by taking a case study
of a National Highway in India (NH-52, Sikar-Reengus), whereby, the identification of accident
black spots, and further investigation is done so as to be able to suggest improvements.
Geographic information system (GIS) is used for spatial visualization of the accident data. Four
accident black spots are identified. The major reasons contributing to the accidents are found to
be poor visibility, over speeding, drunk driving, etc. Suitable mitigation and preventive measures
are suggested for improvement at the spots.
INTRODUCTION
Road deaths and injuries highly affect the lives of people worldwide. Still it is one of the
neglected global problems and requires collective efforts for mitigation and prevention.
According to the records of World Health Organization (WHO), about 1.25 million people die
each year due to road traffic crashes. About half of these deaths are among vulnerable road users
i.e., pedestrians, cyclists, and riders of motorized two-wheelers and their passengers. Another 20-
50 million people gets injured with many sufferings from disability as a result of the injury.
Considerable economic losses are also caused to victims and affected families in the form of
rehabilitation and treatment costs, also reduced productivity due to those killed or disabled.
Approximately 90% of the road traffic deaths occur in low and middle-income countries even
though they account for 54% of the world’s vehicles (WHO 2015). India has faced serious
impact on road safety with rapid expansion of its road network and increased growth rate of
motorized vehicles. Transport Research Wing (TRW) of Ministry of Road Transport and
Highways (MoRTH), India reported 501,423 total number of road accidents (increased by 2.5%
in one year) and 146,133 fatalities (defined as died within 30 days of crash) in the year 2015 due
to road accidents. Vulnerable road users are found to be among the largest proportion of
fatalities, associated with collisions with buses, trucks and cars (Mohan et al. 2016).
Among the Indian states, Rajasthan accounts for eighth highest (24,072) in total number of
crashes and fifth highest (10,510) in road fatalities. About 1.7% of national highways and 3.9%
of state highway road network accounts for 52.4% of the reported road crashes (MoRTH 2015a).
In general, the key risk factors that cause likelihood of accident occurrence and increased
© ASCE
severity include, increase in average speed of vehicle, drinking and driving, distracted driving,
and driving without wearing helmets and seat-belts. Improved legislation, enforcement, safer
roads and vehicles have achieved success in reducing the number of road deaths in high-income
countries.
Data reveal that fifty-eight black spots are located in Rajasthan in the year 2016 and twenty
of them are found on NH-8. Guideline used for these black spot identification is that either five
road accidents took place during the last three calendar years or ten fatalities took place during
Downloaded from ascelibrary.org by York University/Scott Library on 12/15/18. Copyright ASCE. For personal use only; all rights reserved.
the last three calendar years, in a stretch of National Highway of about 500 m in length (MoRTH
2015b). However, this guideline has no statistical or analytical basis. So it may not correctly
represent the situation. In this study, a method is presented to identify black spot considering
several parameters.
Analysis of accident data can be based on three basic requirements. Since accident locations
are distributed spatially over a large scale, hence there is a need of spatial visualization of data.
Parameters that influence the occurrence of an accident need to be identified and studied,
whether they have been formally recorded or not. Once the parameters have been identified there
is a requirement of certain decision making to assign relative importance/weights to them.
Decision making is also required to rank the accident locations according to their hazardousness.
LITERATURE REVIEW
Occurrence of accident is governed by many parameters. These need to be identified and
assigned relative importance, which involves decision-making. Road geometrics (pavement and
shoulder width, roughness, sight distance, gradient, cross fall, super elevation, extra-widening,
etc.) driver and road user characteristics (physical, mental and psychological factors), vehicular
characteristics (vehicle dimensions, turning radius, weight, axle configuration, braking and
acceleration characteristics, etc.), environmental factors (rain, fog, snow, etc.) are some of the
factors to be considered for the study. Some researchers gave more importance to factors related
to road geometrics such as surface, shoulder, drainage, lighting, marking, signs and signals and
developed a safety hazardous index (Agarwal et al. 2013). Some others considered driver and
road-user characteristics as important factors in their study (Najib et al. 2012; Islam and
Kanitpong 2008). Environmental factors were also considered by few others (Sordyl 2015).
Some have applied statistical approach to associate certain degree of reliability to the estimates
(Rao et al. 2005).
Due to limited availability of budget, it is advisable to prioritize the hazardous locations so
that they can be treated preferably. Thus, it is required to prioritize the locations by ranking them
using suitable decision-making tool. These ranked locations can be further analyzed in details by
using comprehensive accident studies, to arrive at the location specific problems and suitable
remedial measures.
One such decision making method is Analytic Hierarchy Process (AHP). It was first
introduced by Thomas L. Saaty (Saaty 1980). The problem is presented in the form of hierarchy
and suitable weights are assigned to each element of the hierarchy. Systematic and quantitative
comparison of relative importance of the criteria is performed. It is based on the principle that
knowledge and judgement of people is valuable while making decision. Thus, decision makers
are involved and they are asked to judge the relative importance of each criterion and rate the
preference using a scale called Saaty’s scale, from 1 to 9 to ultimately rank the alternatives. The
process involves four major evaluation stages (Saaty 1990; Vargas 1990):
© ASCE
1. Structure the problem into hierarchy by breaking down the problem into simpler
constituent decision elements.
2. Generation of pairwise comparison matrices based on Saaty’s scale.
3. Estimation of weights using eigenvalues and normalizing them to obtain consistency
among measures.
4. Calculate the overall ranking.
Some researchers prioritized methods to identify accident black spots, such as frequency of
Downloaded from ascelibrary.org by York University/Scott Library on 12/15/18. Copyright ASCE. For personal use only; all rights reserved.
accidents method, accident rate method, accident intensity method and accident rate-intensity
method. Traffic volume, accident intensity, comprehensiveness, precision in results, effect of
distance and previous experiences were chosen as criteria from literature (Rouhi and Behzadi
2016). Some other prioritized policies to achieve targets of reduced traffic accidents, reduced
pollution, reduced carelessness and minimize congestion (Hajeeh 2012). Thus, this method has
been applied in wide variety of decision situations arising from various sectors.
Graphic data representation methods have also been found useful to draw conclusions
regarding accident black spots (Meshram and Goliya 2013; Nikhil and Kulkarni 2013; Patel et al.
2013; Shruthi et al. 2013). For the locations where the accident data is not available or is limited,
public participation approach is found useful for identifying black spot locations. Its other benefit
is that such studies help in generating public awareness towards the road safety (Kowtanapanich
et al. 2006).
The present study aims at identifying the black spots on Sikar-Reengus section of NH-52,
and detailed analysis of those spots to identify the reasons/contributing factors for the accidents.
GIS has been used to map accident and geometric data due to its spatial analysis capabilities and
various thematic maps are generated. Many researchers have found the use of GIS to represent
and visualize the accident data (Apparao et al. 2013; Deepthi and Ganeshkumar 2010; Isen et al.
2013). AHP has been applied to rank various accidents spots and identify the black spots.
Geometric and traffic studies are conducted specifically at these spots to aid the detailed
analysis. The various contributing factors for the accidents are analyzed and identified, with
recommendations and suggestions for improvement.
DATA ANALYSIS
Police records and National Highway Authority of India (NHAI) provided the accident data
required for this study. The data is for approximately four years (December 2012 to October
2016), containing 309 number of accidents. It include the particulars of accident location, nature
of accident, classification of accident based on severity, reason for accident, geometric features,
weather condition at the time of accident, and other details (IRC 2012a). The geometric features
of the road are identified after visiting the site and the video-graphic data recorded for the entire
stretch of Sikar-Reengus section. The maps obtained from NHAI of the entire road section also
helped to identify the road features. Table 1 and Table 2 present the geometric design standards
(IRC 1980) and flexible pavement structure with bituminous overlay (IRC 1997), respectively
for the road section under study.
The prominent geometric features of Sikar-Reengus section of NH-52 at accident locations
can be seen from Figure 1. The road stretch is mostly straight with slight curve, sharp curve and
steep incline at few locations. These changes in road features may be some of the factors
contributing to the accidents. Stopping sight distance requirements for the design speed of 100
km/h is calculated to be 181.98 m. However, as the road section is straight at the black spot
locations, the required sight distance is assumed to be already existent. At many locations, there
© ASCE
are access roads joining National Highway, forming a T-junctions. The presence of these
junctions resulted in the merging of traffic originating from the surrounding villages, mostly at
high speeds due to the absence of any traffic calming and monitoring facility. This also involves
the wrong side maneuvers which hinders the traffic flow of National Highway and has been a
factor contributing occurrence of fatal accidents, mainly in the form of head-on collisions. The
carelessness in drunk-driving countermeasures is also equally responsible for high number of
accidents.
Downloaded from ascelibrary.org by York University/Scott Library on 12/15/18. Copyright ASCE. For personal use only; all rights reserved.
© ASCE
Figure 2. Total number of accidents during day & night time (hours) at accident locations,
Sikar-Reengus, NH-52.
© ASCE
Figure 3. Total number of fatal and non-fatal accidents at accident locations, Sikar-
Reengus, NH-52.
© ASCE
The developed matrices obtained using the judgement of decision makers are then analyzed
using Expert Choice software and weights for criteria and alternatives are then obtained. Figure 5
and 6 show the obtained weights for criteria and alternatives, respectively. Presence of
intersection is the criterion allotted with the highest weight, followed by the ratio of fatal number
of accidents to the total number of accidents; ratio of grievous number of accidents to the total
number of accidents; ratio of minor number of accidents to the total number of accidents;
presence of curve; and straight road. This means that the presence of an intersection at any
location is the most severe parameter governing the hazardousness at that location, primarily due
to increased number of conflict points.
© ASCE
Ranoli
The land use around the road section in Ranoli is such that it has substantial residential and
commercial activities spread around the four-lane divided National Highway. Due to this,
movement of pedestrians and local vehicles around the highway is quite high. A high number of
head-on and side-swipe collisions, as well as overturning of vehicles have been reported. The
main contributing factor of these accidents is identified as overspeeding according to the police
records. As there is an access road directly connected to the National Highway at this location, it
is observed that wrong side traffic maneuvers originating/culminating from this access road is
quite high. A high number of cross movement of pedestrians is observed through the gaps
between the crash barriers and the iron barricades, which may be cited as a major contributing
factor for accidents.
Trilokpura
The horizontal profile of the road at this section has presence of horizontal alignment, which
affects the available sight distance. Overspeeding, particularly at curves, increases the risk of
accidents due to the reduced sight distance. The junction of access road connecting to the
National Highway is not provided with any traffic calming measure. Due to this high number of
wrong maneuvers originating from access road are observed. This eventually leads to accidents
in the form of head-on or right-turn collisions, sometimes causing overturning of vehicles.
Goriya
The land use along Goriya road section consists of the presence of a fuel station, a bus-stop
along the highway, and proximity to a railway station. This leads to increased number of road
© ASCE
users, pedestrian movements, as well as wrong side vehicle movements. Available sight distance
is reduced due to the presence of horizontal alignment and overspeeding in such cases increases
the risk of occurrence of accident.
Palsana
Palsana has two-lane road section in-between the longitudinal profile of the four-lane divided
highway. This creates bottleneck situation and increases the risk of accidents. The road has
Downloaded from ascelibrary.org by York University/Scott Library on 12/15/18. Copyright ASCE. For personal use only; all rights reserved.
gentle incline, and iron barricades are not present along the incline. Quite high number of
collisions is reported, occurring mainly due to illegal wrong turning of vehicles from access road
connecting to the National Highway. This eventually leads to accidents in the form of head-on-
collisions, or right-turn collisions, sometimes causing overturning of vehicles.
CONCLUSION
The AHP based method for quantifying the location hazardousness has been applied to rank
various accidents spots, and hence identify the black spots on Sikar-Reengus section of NH-52.
A further detailed analysis has been done at these identified spots to ascertain the reasons/
contributing factors for the accidents, as well as to relate them to the parameters chosen in the
AHP method for verification purpose. From the method verification, it can be concluded that the
hazardousness of a (i) straight road stretch without intense land use depends more on the fault of
driver or motor vehicles, (mostly overspeeding), (ii) straight road stretch with intense land use
depends more on the cross movements of road users, (iii) road stretch with horizontal curve
© ASCE
depends more on visibility at curves, and (iv) road stretch with junction depends on high number
of conflict points and lack of traffic calming/traffic monitoring measures. The proposed AHP
based method is found to be effective in quantifying the hazardousness of accident locations, and
identification of black spots, on the NH section.
REFERENCES
Agarwal, P. K., Patil, P. K., and Mehar, R. (2013). "A Methodology for Ranking Road Safety
Downloaded from ascelibrary.org by York University/Scott Library on 12/15/18. Copyright ASCE. For personal use only; all rights reserved.
Hazardous Locations using Analytical Hierarchy Process." Procedia-Soc. Behav. Sci., 104,
1030–1037.
Apparao, G., Mallikarjunareddy, P., and SSSV Raju, G. (2013). ''Identification of Accident
Black Spots for National Highway using GIS.'' Int. J. Sci. Technol. Res., 2(2), 154–157.
Deepthi Jayan, K., and Ganeshkumar, B. (2010). ''Identification of Accident Hot Spots : A GIS
Based Implementation for Kannur District , Kerala.'' Int. J. Geomatics Geosci., 1(1), 51–59.
Hajeeh, M. A. (2012). ''Traffic Accidents in Kuwait: A Decision Making Analysis.'' Int. J. Appl.
Math. Informatics, 6(2), 84–91.
IRC (Indian Road Congress). (1980). Geometric Design Standards for Rural (Non-Urban)
Highways, IRC: 73-1980, Indian Road Congress, New Delhi, India.
IRC (Indian Road Congress). (1996). Highway Safety Code, IRC: SP:44-1996, Indian Road
Congress, New Delhi, India.
IRC (Indian Road Congress). (1997). Guidelines for Strengthening of Flexible Road Pavements
using Benkelman Beam Deflection Technique, IRC: 81-1997, Indian Road Congress, New
Delhi, India.
IRC (Indian Road Congress). (2012a). Road Accident Recording Forms A-1 and A-4, IRC:53-
2012, Indian Road Congress, New Delhi, India.
IRC (Indian Road Congress). (2012b). Code for Practice For Road Signs, IRC: 67-2012, Indian
Road Congress, New Delhi, India.
Isen, L., A., Shibu, and M. S., Saran. (2013). ''Identification and Analysis of Accident Black
Spots using Geographic Information System.'' Int. Conf. Energy Environ. (ICEE 2013). 131–
139.
Islam, M. B., and Kanitpong, K. (2008). ''Identification of Factors in Road Accidents through in-
depth Accident Analysis.'' IATSS Res., 32(2), 58–67.
Kowtanapanich, W., Tanaboriboon, Y., and Chadbunchachai, W. (2006). ''Applying Public
Participation Approach to Black Spot Identification Process– A Case Study in Thailand''.
IATSS Res., 30(1), 73–85.
Meshram, K., and Goliya, H. S. (2013). ''Accident Analysis on National Highway-3 between
Indore to Dhamnod.'' Int. J. Innov. Eng. Manag., 2(7), 57–59.
Mohan, D. (2009). ''Road Accidents in India.'' IATSS Res. 33(1), 75–79.
Mohan, D., Tiwari, G., and Mukherjee, S. (2016). ''Urban Traffic Safety Assessment: A Case
Study of Six Indian Cities.'' IATSS Res., 39(2), 95–101.
MoRTH (Ministry of Road Transport and Highways). (1988). Motor Vehicles Act, 1988,
Ministry of Road Transport and Highways, Government of India, New Delhi.
MoRTH (Ministry of Road Transport and Highways). (2015a). Road Accidents in India-2015,
Transport Research Wing, Ministry of Road Transport and Highways, Government of India,
New Delhi.
MoRTH (Ministry of Road Transport and Highways). (2015b). Protocols for Identification and
Rectification of Road Accident Black Spots on National Highways, Ministry of Road
© ASCE
55.
Ponnaluri, R. V. (2012). ''Road Traffic Crashes and Risk Groups in India: Analysis,
interpretations, and prevention strategies. IATSS Res. 35(2), 104–110.
Rao, B. S., Madhu, E., Jalihal, S., and Sreddy, T. (2005). ''Accident Study on National Highway-
5 between Anakapalli To Visakhapatnam.'' East. Asia Soc. Transp. Stud., 1973–1988.
Rouhi, A., and Behzadi, G. A. (2016). ''Prioritizing Four Methods and Presenting a Combination
Method for Determining Black Spots using Analytical Hierarchy Process (AHP)''. J. Sci.
Today’s World, 5(4), 105–109.
Saaty, T. L. (1980). The Analytic Hierarchy Process, Priority Setting. Resource Allocation,
McGraw-Hill, New York International Book Company.
Saaty, T. L. (1990). How to make a decision: The Analytic Hierarchy Process. Eur. J. Oper.
Res., 48(1), 9–26.
Shruthi, P. Venkatesh, V. T., Viswakanth, B., Ramesh, C., Sujatha, P. L., and Dominic, I.R.
(2013). Analysis of Fatal Road Traffic Accidents in a Metropolitan City of South India. J.
Indian Acad. Forensic Med., 35(4), 317–320.
Sordyl, J. (2015). Application of the AHP method to analyze the significance of the factors
affecting road traffic safety. Transp. Probl., 10(2), 57–68.
Vargas, L. G. (1990). An overview of the analytic hierarchy process and its applications. Eur. J.
Oper. Res., 48(1), 2–8.
W.H.O. (World Health Organization). (2015). Global Status Report on Road Safety 2015. World
Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland.
© ASCE