Transportation contributes to the economic, industrial and social development of any country. In India, transportation by road is a widely utilized mode that serves maximum service to all its commuters. As per World Health Organization (WHO), the major causes of traffic accidents are driving at higher speeds than at specified limit, driver under the influence of alcohol and not wearing helmets or seat belts. The frequency of road accidents in India is the highest when compared with the world. National Crime Records Bureau (NCRB) report reveals that every year more than 135,000 traffic related deaths occurs in our country.

Hyderabad, is the 6^th largest metropolitan city in India, located in the northern part of Deccan plateau and in Southern India on the banks of River Musi extending an area of 625 square kilometers. There are over 3.5 million vehicles operating in the city, of which 58% are two-wheelers, 23% cars and 11% three-wheelers. The remaining 8% include buses, goods vehicles and taxis. Hyderabad road network system is not homogenous in traffic system and road safety. A few locations in the city have an abnormally high number of accidents which are illustrated as crash concentrated hotspots or black spots. These Black spots are responsible in declination of road service and its quality. To provide a safe driving environment in city for its commuter's, the road sections for black-spot shall be identified and provide proper treatment for reducing the cause of road accidents.

Black-spot is a place where traffic accidents have been historically concentrated. At these locations, accidents occur for various reasons of which the major factors contributing for the cause of road accidents are,

• Road environment,
• Condition of vehicle,
• Concentration and physical state of road users.

Improvement in road safety is expressed by providing alternatives for road geometry and environmental characteristics at highly accident concentrated locations. More specifically, the tasks involved in targeting and treating road sections of high number of accidents.

The present study identifies accident black-spots in the selected locations based on the accident data collected for the past three years and to suggest improvement measures. The objectives of the study include,

• Identification of accident black-spots through accident severity method and frequency method.
• Prioritizing the black spot locations.
• To conduct audit studies for providing measures in improvements at black-spot locations.

A step wise procedure is adopted for identification of blackspot locations and performing audit studies.

• Accident data was collected from a secondary source and the road geometry was also collected on selected locations of Hyderabad city. Later was compiled in MS Excel spread sheets.
• The top ten black spots were identified using the accident severity method and the accident frequency method. Accordingly it was prioritized and ranked for improvements.
• Audit studies were conducted on highly ranked blackspot locations and identified for the cause of occurrence.

Majority of studies by various authors were aimed on Road Safety management. A few have contributed on identification of black spots for the improvement of road safety through improving geometrical parameters, street furniture etc. Mahal et al. [10] defined black- spots as those locations whose accident frequency is significantly higher than expected at prescribed level of significance. Hauer [5] had provided ranking by accident rate and accident frequency method. Frequency, severity and accident rate were used to identify the black-spot locations. Deelesh and Rajiv [2] have determined accident-prone locations which are commonly termed as accident black spots using a Geographic Information System (GIS). Fajaruddin [3] et al. had developed an accident prediction model by using Multiple Linear Regression. Krishnamurthy [7] et al. had focused on accident frequency and severity index methods to identify accident black-spots and ranked them by frequency and severity values. Shenjun and Becky [11] had focused on hot zone identification and dangerous road locations in context with pedestrians. Aliasghar et al. [1] had explained on identification and prioritization of Accident-Prone Sections (APSs) which was based on efficiency concept to emphasize accidents. Huayun [6] has discussed on utilization of geo-coding and VRS-GPS positioning technology for recording traffic accidents. Liyamollsen et al. [9] had discussed on identification of accident locations using Geographic Information System (GIS). Evaluation was carried through Weighted Severity Index (WSI) method and correspondingly ranked accident locations. Gourav [4] et al. has made an attempt for ranking the black-spot using accident severity value. Leelakajonjit et al. [8] has attempted to study the road safety management system and proposed a better method to identify black spots.

3.1 Critical Appraisal

Most of the studies conducted were on identification of black-spot using accident frequency method, accident severity method, accident rate method, rate quality method, etc., for homogeneous and heterogeneous traffic conditions. But limited studies were carried out for our city of Hyderabad. In this context, an attempt is made in identifying black-spot locations. Audit studies were also conducted on highly ranked black-spot locations. An accident model is also developed for improving the existing situation.

3.2 Black Spot Identification Methods

There are several methods for identification of black spot location for a given roadway facility. The data pertaining to the roadway accident, geometric, traffic and environmental conditions decides which method to be employed for identification of these black-spot. A few methods used for identification of accident black-spot are Accident Frequency Method, Accident Rate method, Accident Severity Method, Accident Density Method, Frequency Rate Method, Safety Indices Method, Severity- Rate Method, Rate-Quality-Control Method, and Empirical Bayes Method. In this study black-spot has been identified using accident frequency method and accident severity method.

3.2.1 Accident Severity Method

In Accident Severity Method [12], injury severity is taken into consideration. Number of fatalities, number of injured people and number of property damage are only the cases in accident to be taken into consideration. Separate weights for fatal, injuries and property damage of severity are provided. These weights are provided based on the importance of injury and severity.

The severity value for road section j is

(1)

Severity = number of fatalities (f) × 9 + number of injured persons (b) × 3 + number of damaged vehicles (d) × 1.

The relative severity value is then,

(2)

The above equation is severity per accident. The average value is estimated as,

(3)

and the variance σ² is estimated with

Black-spot from the severity point of view is if Q_j > Q_c.

where the critical value

A_j is number of accidents on section j during certain time period.

Q_j is the relative severity value on section j.

S_j severity value for road section j.

Kα is a constant chosen for the significance test and is determined from a normal distribution for a given certain significance level α.

α=1% gives k_α = 2.576,

α = 5 % gives k_α = 1.645, and

α = 10 % gives k_α = 1.282.

3.2.2 Accident Frequency Method

Accident Frequency Method [12] considers a number of accidents at a particular place for ranking the black-spots. Three year accident data was used for determining yearly frequencies and average frequency on selected locations of Hyderabad city. These black-spot are ranked in the descending order of observed frequencies. A road section is considered as black-spot if accident frequency A_j > A_c ,

where

(4)

A_c is the critical value for accident frequency.

L_j is the length of the road section. Here, L_j is assumed to be 1 km.

F_ave is the average accident frequency for all road sections.

Accident data for the last three years (2012-2014) were collected as shown in Figure 1. Year wise data was obtained from the traffic commissioner office of police department and is shown in Figure 2. In the present study, the black-spot locations were selected locations whose fatalities are high. The main study stretches includes Banjara Hills, Jubliee Hills, Punjagutta, Sanjeev Reddy Nagar, Gopalapuram, Begumpet, Nampally, LungerHouze, Bowenpally, Bollaram, Secundrabad (Osmania University location) and is indicated in Figure 3. Data was also collected for accident time, date of occurrence, type of vehicles involved, location of the accident, details of injury or severity of accidents, road geometry etc., Figure 4 shows accused vehicles in the year 2014 and Figure 5 explains the accident victim's proportion, it is also observed from the Figure 5 that 2-wheelers and pedestrians are mostly involved in fatalities in the city.

Figure 1. Accident Rate

Figure 2. Study Areas of Hyderabad

Figure 3. Accused Vehicles Count

Figure 4. Accident Victim's Proportion

Figure 5. Comparison of Total Number of Accidents from Existing Data with the Model Values

Ranking of accident spots helps to coordinate the improvement programs and to assess the place that shall be provided for immediate attention. Accident frequency method, the accident frequencies of all locations for three year period is calculated and black-spot are listed in descending order. In accident severity method, severity index is used for black-spot identification and is provided as descending order for severity. The accident frequency and severity values of the three consecutive years are shown in Table 1 and Table 2. Accident severity value for Bollaram location calculated as,

Table 1. Accident Frequency Values of three consecutive years in Hyderabad location

Table 2. Accident Severity Values of three consecutive years in Hyderabad location

Accident frequency value for Banjara hills location:

Common black spot were considered from severity and frequency method in order to perform audit studies. These studies shall explain the reasons on exiting situation for high accident rate and thus shall improve the existing situation and are shown in Table 3.

Table 3. Accident frequency and severity values for the study areas

5.1 Audit Study

Site reconnaissance studies as spot speed surveys and geometric surveys were conducted at top ten accident black-spot locations identified from severity and frequency method as shown in Table 3. Spot speed surveys are carried out to understand the speed of different vehicles travelling at higher speeds than that of design speed. Bowenpally and Bollaram locations spot speeds are measured on both the direction for all categories vehicles.

5.1.1 Identification of Accident causing Factors

(i) Bollaram :

There are few accidents observed but with more severity. The road design is dangerous as there are two sharp curves one at Koteswara temple – the 4^th training battalion and the second at St. Anns School which are having the high accident rate record. These curves specially cause accidents by heavy vehicles. Majority of accidents occur during night time due to heavy vehicles passing from this area. As part of detailed analysis of spot speed studies carried out at various sections in this locations and was observed that more heavy vehicles are running with speeds more than design speed specified.

Suggested Improvements:

• Required extra widening on the curve.
• Speed must be restricted to 30 kmph on the entire stretch.
• Provide or upgrade high visibility during night time.

(ii) Bowenpally

The accidents occur with more frequency but with less severity. More traffic accidents are observed during peak hours. No signals during the early morning and late nights are observed. Vehicular speeds are found to be higher. From the interview survey, it was observed that drawbacks like pot holes, pipe line damages on the road will always distract the flowing traffic. The major accident spots in this area include Chintal road, the Mallareddy function hall road, and junction at Muslim grave yard.

Suggested Improvements:

• Placing proper sign boards on this stretch.
• Guiding the road users through sign boards with reflecting colors especially during night times when the visibility is low.
• Speed regulating measures shall be monitored strictly to make sure that three wheelers and four wheelers passing through the stretch run with desired speed.
• Traffic rules should be strictly enforced at minor intersection in this location.
• Alternative left turn routes shall be allowed for two wheelers. This is available at Bowenpally intersection and must be implemented to reduce collisions of vehicles.
• The road should be widened and construction works or any repair works on the roads must be planned to carry out during the off peak hours.

(iii) Begumpet

The section had heavy traffic flow during the peak hours. The curve located at the Yellama Temple – Balkampet Junction is well designed. But, major accidents occur during night time due to rash driving, over speed, drunk and drive, and due to the negligence of road users. Pedestrians and two wheelers are mostly affected. Majority of accidents occurs at Balamrai check-post section and the paigah plaza point.

Suggested Improvements:

• Providing separate pedestrian crossing at Balamrai and Paigah Plaza which would significantly reduce pedestrian collisions.
• Speed regulating measures are to be monitored strictly during off-peak hours.
• Continuous footpath is essential as this stretch is covered with commercial establishment and school zone (Hyderabad Public School).

(iv) Punjagutta

More accidents occur due to the vehicular collision approaching from opposite direction. A very large number of shopping malls and commercial buildings are located in this area. Street parking is not designed properly which reduces the roadway width. Accidents' occurring on this road section is at Keertilal Jewellery Complex, and on Rajbhavan Road.

Suggested Improvements:

• On street parking, restrictions must be implemented at all locations of commercial establishments on this section during peak-hour traffic.
• Separate parking zones shall be designed for accommodating commuters of shopping complexes, buildings, hospitals etc.
• On these roads, the usage and provision of foot paths shall be made mandatory.
• Speed regulating measures shall be monitored strictly during night times.

(v) Nampally

Heavy traffic is observed towards Nampally Junction. More shops are located on the road side just besides the junction. Foot paths were available at a few of the locations and are provided haphazardly. Parking on the streets is the major cause for the traffic accidents. From the interview survey, it was found that major number of accidents take place because of rash driving, signal jumping, and wrong side drives. More frequently accidents occur at Shivaji Hotel, STN Road signal, besides Niloufur Hospital.

Suggested Improvements :

• Speed regulating measures are to be monitored strictly.
• Parking restrictions must be implemented on either side i.e. at Shivaji Hotel and beside Niloufur hospital.
• Separate parking zones should be designed for at commercial establishment of Namaplly Junction.
• Foot paths and medians should be made mandatory for usage of pedestrians only instead of street vendors.

“An Accident Model” is generally an algorithm pitting a dependent variable against several independent variables. The dependent variable in the accident prediction model is the number of accidents, while the independent variables are quantitative variables as vehicular speed, cause of the accident, time of occurrence of accident, or driver wearing seat belt or helmet. The errors of this model were calculated and tested for t-test. The results show that, the coefficient of determination lies between 0 and 1 and therefore the model holds good fit and having the t- values within the limits. The commonly used model prediction of accident at black spot locations will be in the form of multi linear regression equation as,

where Y is the Number of accidents at the black spot location, X₁: Vehicular speed, X₂ : Driver wearing seat belt or helmet, X₃ : Cause of road accident, X₄ :

Time of occurrence of accident. The model is obtained as, Y=-124.638+17.23X₁ +24.16X₂ +36.25X₃ - 5.868X₄.

This model is also compared with the existing variables and provides a close fit as shown in Figure 6. Slight improvements on all the model variables will comparatively provide reduction in the rate of accidents at all the black spot locations considerably.

Figure 6. Comparison of Total Number of Accidents from Existing Data with the Model Values

7. Summary and conclusions

The analysis of accident black-spots helps in identifying the stretches where accidents are more. In the present study, black-spots are identified and audit studies were conducted to find out the deficiencies of road related factors (traffic related). Based on the study, the following suggestions are recommended.

• As both Accident Severity Method and Accident Frequency Method are used in the identification of black-spot locations, this study is more reliable for accident black-spot treatment programs.
• Banjara hills and Bollaram are found to be the top black spot areas with respect to Accident Frequency Method and Severity Method respectively.
• From all the spots analyzed, two wheelers and pedestrians are found to be more effected.
• In all the places studied, the safety problem rose with the sign boards. Providing proper sign boards to guide the road users help to reduce almost 70% of the accidents occurring in every location.
• The model provides certain variables where the improvements in these values will help in reduction of accidents at black-spot locations.

 

Overall this study appears to be effective in reducing number of causalities if suggested improvements were implemented.

Appendix

Photo 1. Irregular traffic flow at Banjara Hills space at Nampally

Photo 2. Vehicles parked in NO Parking Area

Photo 3. Signal Jump at Jubille Hills

Photo 4. Road Repair Works at Bowenpally

References

[1]. AlisgharSadegi, EasmaeelAyati, and Mohammadali Pirayesh Naghab, (2013), “Identification and prioritization of Hazardos Road Locations by Segmentation and Data Envelopment Analysis Approach”. Safety and Security in Traffic Preliminary Communication, Vol. 25, pp. 127-136.

[2]. Deelesh Mandloi, and Rajiv Gupta (2003), “Evaluation of accident black spots on roads by using Geographical Information Systems (GIS)”, Map India Conference.

[3]. Fajaruddin Mustakim, Ismail Yusof, Ismail Rahman, Abdul Aziz Abdul Samad, and Nor Esnizah Binti Mohd Salleh, (2008), “Black spot Study and Accident Prediction Model Using Multiple LinerRegression”. First International Conference In Developing Countries (ICCIDC-I), Advancing and Integrating Construction Education, Research & Practice.

[4]. GouravGoel, and S.N. Sachdeva (2014), “Identification of Accident Prone Locations Using Accident Severity Value on a Selected Stretch of NH-1”. International Journal of Engineering Research and Applications; National Conference on Advances in Engineering and Technology.

[5]. Hauer, E, (1997). Observational Before-After Studies in Road Safety. Pergamum, Tarrytown.

[6]. Huayun Chen, (2012), “Black Spot Determination of Traffic Accident Locations and Its Spatial Association Characteristic Analysis Based on GIS”. Journal of Geographic Information System, Vol. 4, pp. 608-617.

[7]. Krishnamurthy K, M.V.L.R. Anjaneyulu, and R. Rakesh, (2011), “Black spot identification, analysis and improvement measures on selected national highway stretches in Kerala, India”, TRB Annual Meeting.

[8]. Leelakajonjit A, Brannolte U, Kanitpong K, and Iamtrakul P, (2014), “Analysis of Road Accident Hazardous Locations in Bangkhen Police Station, Thailand”. Acta Technical Jaurinensis, Vol. 7, No.1, pp. 33-45.

[9]. LiyamolIsen, Shibu. A, and Saran M. S, (2013), “Evaluation and treatment of accident black spots using Geographic Information System”. International Journal of Innovative Research in Science, Engineering and Technology, Vol. 2, No. 8.

[10]. Mahal, D Hakkert, A.S., and Mahalel, D. (1978). “Estimating the number of accidents at intersections from a knowledge of the traffic flow on the approaches”. Accident Analysis and Prevention. Vol.10, pp. 69–79.

[11]. Shenjun YAO, Becky P.Y. LOO (2012), “Identification of Hazardous Road Locations for Pedestrians”. International Symposium on Safety Science and Technology, Vol. 45, No.12, pp. 815 – 823.

[12]. The black spot manual of Swe-Road Traffic Safety Project (2001). ANKARA Traffic Safety Consultancy Services.