Road Accident Prediction for Nashik City

Transcription

1 Road Accident Prediction for Nashik City V.S. Aher 1, K.H. Ahirrao 2, M.R. Gaikar 3, S.S. Surve 4, R.R. Kshatriya 5 1 to 4 UG Student, Department of Civil Engineering, Matoshri College of Engineering and Research Centre, Nashik, Maharashtra, India. 5 Assistant Professor, Department of Civil Engineering, Matoshri College of Engineering and Research Centre, Nashik, Maharashtra, India. Abstract In today s time due to galloping urbanization and development ridership has increased. This has subsequently led to perilous situations for drivers as well as for pedestrians as the number of accidents over the past years has significantly increased. The study focuses on developing a regression model to predict the number of accidents along the selected stretch of road in Nashik city of India. Best fit trends for various parameters have been found to arrive at the best model from the analysis using MS Excel 2007 and Origin Software. Keywords: Road Accidents, Accident Prediction, Road Geometrics, Traffic Volume 1. Introduction There has been rapid increase in urbanization which has led to better lifestyle of people. But these advancements has created burden on roads due to increase in vehicle ownerships due to which problems related to traffic have escalated to an alarming rate. The major issue with this increase is the increase in road fatalities which causes individual losses and economical losses, Hence, there is a basic requirement for assessing these accidents through identification of the various causes that are responsible for their occurrence, analyze their dependency on those causes and also to recommend various remedial measures to alleviate losses due to these accidents. According to road accident statistics of 2011 by Ministry of Road and Highway Transport research wing, In Maharashtra the number of accidents has increased by 12.7 percentages in the year of 2015 and the Maharashtra state road transportation corporation (MSRTC) has identified 158 accident prone spots on major road passing through the Pune region. According to the survey 2010 accident prone spot in the state of these, the highest (392) are in the Mumbai Region followed by Nashik (286). Among the major Roads passing through the Pune Region, The Pune Aurangabad Highway has been identified as having the highest numbers of accident prone spot (180). World Health Organization indicates that India has the highest number of traffic accident fatalities in the world. Various factors such as increase in number of vehicles, faults in geometric design, attitude of drivers, age of drivers, type of vehicle et cetera are responsible for increase in number of accidents. Therefore, to assuage the damages caused by accidents it is important of predict the number of accidents before its occurrence. Following statistics have been given by Transport Research Wing of Ministry of Road Transport and Highways. In ,86,476 road accident deaths- highest in the world-25.2 per cent (1, 22,589) were fatal accidents. Major contributor to traffic deaths in India has been attributed to mistakes of drivers and also drunk and drive, which is responsible for 70% of road fatalities. India accounts for about 10 percentage of road accident fatalities worldwide. An estimated 12,75,000 persons are grievously injured on the road every year. 75 V.S. Aher, K.H. Ahirrao, M.R. Gaikar, S.S. Surve, R.R. Kshatriya

2 An accident is defined as as an event which comprises of damage to the property and/or injury to the road users, which are recorded first by the police and/or emergency services. Prediction of accidents is generally done by using statistical methods. India currently has only one mechanism of collecting road accident data and that is from the police accident investigations. The police data is analyzed by two government organizations to publish two annual reports - The National Crime Record Bureau s Accidental Deaths and Suicides in India, and the Transport Research Wing, Ministry of Road Transport and Highways Road Accidents in India report. These reports provide an understanding of fatal road traffic accidents at a national level and information such as type of road users (age, sex, and vehicle type), incidence rates for states and cities, road type, etc. 2. Study Area and route details The present study is being done in Nashik city in the state of Maharashtra, India. Nashik is one of the busiest cities of Maharashtra which has made it more prone to accidents. According to a survey, Nashik accounts for highest number of traffic violation in Maharashtra. According to the survey 2010 accident prone spot in the state of these, the Mumbai Region followed by Nashik (286), among the major Roads passing through the Pune Region, The Pune Aurangabad Highway has been identified as having the highest numbers of accident prone spot (180). Image:1 Road Map Of Nashik City 76 V.S. Aher, K.H. Ahirrao, M.R. Gaikar, S.S. Surve, R.R. Kshatriya

3 In 2014, approximately 20,252 accidents were recorded in Nashik city. Also, in 2014 there has been an increase of 20 percentages in number of fatal accidents. All the areas were studied and based on number of accidents, availability of high speed road and feasibility of location, the national highway of Mumbai-Agra passes via city Nashik. It is also well connected to city Pune with highway NH-50. Nashik is actually a main road connection of main state highways. Nashik is also connected to Surat, Mumbai, and Aurangabad, Pune, Dhule, Ahmednagar and all other important cities of India. The NH-3, national highway is being changed into multi-lane highway and this multi-lane street has about six flyovers that will go through city of Nashik. The flyovers will begin from main garvare point and will reach at temple of Hanuman at Panchvati and Nashik- Pune highway NH-50. Apart from these, other major cities like Aurangabad (200 km) are connected via state highway which is also four lane highway. Nashik is easily accessible by road from Gujarat state in Western India, has been selected for present study. The entire stretch has been divided into sections, each being bound between two consecutive major intersections so that homogeneity can be maintained and analysis can be made easier. On an average 40% trips are made by two-wheelers, 43% trips are made by four-wheelers, 9 % by three-wheelers and the rest trips comprises of those made by buses, trucks and LCVs. The map shows entire stretch on Google Maps. 3.Methodology The major activity in such research work was to identify databases of data collection, geometrical design, traffic control, traffic volume and traffic accident data for at-grade (level) intersections outside urban areas (rural intersections), that were suitable for testing the methodology and the development(re establishment) of statistical models for traffic accident prediction. The intent of this paper is not to replicate the detailed discussions of the methodological alternatives provided in those papers, but instead to focus on discussing the methodological evolution, the current methodological frontier and remaining methodological issues (the interested reader is referred to those papers for a review of previously used methodological approaches). 4. Data Collection and Analysis Vehicle Population: The data related to the vehicle population are collected from the Nashik Regional Traffic Office (RTO) of the Nashik. Mr. Pawar Sir, gives the data of vehicle population of the 2015, of the Total Nashik district vehicles with category wise. Accident Data: All the required data which have been used to give a structure to this paper are being collected from the Head office of Nashik traffic police, Sharanpur Road Nashik. It is the only source of Collection of the data related to road traffic accident in this area. All accidental data collected from the police from the traffic department. Generally there is a very high level of data availability (exceeding 90%) for accident types, accident severity and additional outside accident influences, both for motorways / freeways and for two-lane two-way rural roads. It is notable that the availability rate of prevailing accident cause data is somewhat lower. The information recorded consists of numerous types of data, such as crash information which includes number of accidents, accident severity and vehicle type, field information such as major/minor road width, lane marking, control type, geometry and location, and miscellaneous information such as weather, time of accident, and Average Annual Daily Traffic (AADT). Since most of the above information is recorded by Police and later digitized in raw form, it is required to reduce the above information customizing it for specific analysis. 77 V.S. Aher, K.H. Ahirrao, M.R. Gaikar, S.S. Surve, R.R. Kshatriya

4 No Of Accidents International Journal of Engineering Technology, Management and Applied Sciences Table:1 Accidents Cause Due To Location Sr No Location Near School or College Near or Inside a village NA 3 Near a Factory Industrial Area NA 4 Near a Religious Place Near a recreation place/cinema NA 6 In Bazar Near Office complex Near Hospital Residential area Open Area Near Bus Stop Near Petrol Pump At Pedestrian Crossing Affected by Encroachments Narrow Bridge or Culverts Other NA Total ACCIDENTS CAUSE DUE TO LOCATION Graph:1 Area Wise Distribution of Accidents 78 V.S. Aher, K.H. Ahirrao, M.R. Gaikar, S.S. Surve, R.R. Kshatriya

5 Sr. No Year Table:2 Total Number of Accidents According to Year Fatal Grievous Inj. No. of Accidents Minor Inj. Non Inj. No. of Person Total Killed Injured No Of Accidents According To Year No Of Accidents Total Number of Accidents According to Year 79 V.S. Aher, K.H. Ahirrao, M.R. Gaikar, S.S. Surve, R.R. Kshatriya

6 Table :3 Nature Of Accidents SR.NO. NATURE OF % ACCIDENTS 1 Overturning Head on collision Rear & collision Collision brush/side swipe Right angled collision Skidding Right turn collision Hit & run Others Hit from Back Hit fixed objects Hit pedestrian TOTAL % 0.22% 0.90% Nature Of Accidents 0.43% Overturning 10.54% Head on collision Rear & collision Collision brush/side swipe Right angled collision Skidding 13.49% Right turn collision 28.38% Hit & run Others 4.37% 5.23% 5.88% 8.60% 9.15% Hit from Back Hit fixed objects Hit pedestrian Graph :3 Nature Of Accidents 80 V.S. Aher, K.H. Ahirrao, M.R. Gaikar, S.S. Surve, R.R. Kshatriya

7 Details of Drivers (A) Total Driver's Age Table :4 Accidents Due To Age of Persons Number of Accidents Total % 1. Less than 18 years and Above Unknown Age Total (B) Total Educational Quali. of Driver 1. Less than 8th Standard Standard Standard 10 and Above Qualification not known Total (C)Total Type of Licence 0 1. Regular Licence Learner's Licence Without Licence Total (A) +(B) +(C) TOTAL Accidents Due To Age Of Person 28% 37% (A) Total Driver's Age (B) Total Educational Quali. of Driver (C)Total Type of Licence 35% Graph :4 Accidents Due To Age of Persons 81 V.S. Aher, K.H. Ahirrao, M.R. Gaikar, S.S. Surve, R.R. Kshatriya

8 Table: 5 Accidents Due To Time of Day Time Year Total % to 9.00 hrs (Day) to hrs (Day) to hrs (Day) to hrs (Day) to hrs (Night) to hrs (Night) to 3.00 hrs (Night) to 6.00 hrs (Night) TOTAL ACCIDENTS DUE TO TIME OF DAY to 9.00 hrs (Day) to hrs (Day) to hrs (Day) to hrs (Day) to hrs (Night) to hrs (Night) to 3.00 hrs (Night) to 6.00 hrs (Night) Graph: 5 Accidents Due To Time of Day 5. Conclusion From the accident data 2005 To 2016, it is observed that out of total accidents that have occurred in Nashik city 24.86% of vehicles involved were four-wheelers and 99% of vehicles involved were two-wheelers and vehicles were commercial vehicle like buses (18.19%) and trucks& Lorries (11.36%) and 50% of accidents due to commercial vehicles were fatal. From Table 5, it is observed that during morning peak hours 9:00 hrs to 12:00 hrs (Day) maximum number of accidents has occurred which indicates that this is because of under-performance of carriageway and not due to high speed. This situation remains the same till 6 p.m. in the evening. This observation supports the final expression obtained to predict number of accidents which shows that carriageway width is a major parameter influencing number of accidents. 82 V.S. Aher, K.H. Ahirrao, M.R. Gaikar, S.S. Surve, R.R. Kshatriya

9 Also, it can be observed from Table 7 that after evening peak hours, i.e. after 18:00 hrs. To 21:00 hrs (Night) accidents are more which can be attributed to high speeds of vehicles as the traffic volume reduces after 21:00 hrs. which also supports final expression for predicting number of accidents. References 1. Fatima Pereira Silva, Jorge Almeida Santos, Anderia Meireles, CongresoPanamericano de, Transporte, y Logistica (panam 2014) Road Accident: Driver Behaviour, Learning and Driving Task (adapted by Carvalhasis,2002),Procedia Social and Behavioral Sciences 162, (2014), De Ona Juana, Garach Laurab, Ph.D. TRYSE Research Group. Department of Civil Engineering, University of Granada, has published the paper Accidents prediction model based on Speed Reductions on Spanish Two lanes Rural Highways Procedia Social and Behavioral Sciences 53 (2012), Punit Goyala, Sanskruti Joshi, Hemanth Kamplimath C, Dhruv Prajapati D, By Accident Prediction Modelling Of Major Road Of Ahmedabad Abstract number 268, 12 th TPMDC IIT Bombay December 2016 under graduate student and assistant professor of civil engineering Dept. Nirma University Ahmedabad, Gujrat, India. 4. Kazimierz Jamroz, Wojciech Kustra, Marcin Budzynski a Gdansk University of Technology, 12 Narutowicza strs, , Poland. 6 th Transport research Aerene April 18-21, 2016, Pedestrian protecton, speed enforcement and road network structure the key action for implementing poland s vision zero in 18 th April V.S. Aher, K.H. Ahirrao, M.R. Gaikar, S.S. Surve, R.R. Kshatriya