ABSTRACT
An accident prediction model was developed for determining the accident potential of a transport vehicle while on transit. The model identifies the various factors responsible for vehicle crashes with the help of accident data obtained from the database of the Nigerian Federal Road Safety Corps (FRSC), the percentage contribution of each factor is calculated. These accident-cause factors were further grouped into three distinct classes: Human factors (HF), Mechanical factors (MF) and Environmental factors (EF). Analysis of the accident data showed that HF is the chief cause of most road accidents recorded, followed by MF and EF with probabilities of 0.846, 0.138 and 0.016 respectively. Also, driver age, travel distance and maintenance frequency of the vehicle were considered in the development of the model. The model gives an output ranging from 0-1. Values close to ‘0’ mean low accident probability while values close to ‘1’ signify high accident probability. Application and adherence to this model will significantly reduce the frequency of road accidents. Finally, transport companies and fleet operators are therefore encouraged to embrace and use this innovation for safer operations.
TABLE OF CONTENTS
Title Page i
Declaration ii
Dedication iii
Certification iv
Acknowledgements v
Table of Contents vi
List of Tables ix
List of Figures x
Nomenclatures xi
Abstract xii
CHAPTER 1: INTRODUCTION
1.1 Background of Study 1
1.2 Statement of Problem 3
1.3 Aim and objectives of Study 3
1.4 Scope of Study 4
1.5 Justification for the Study 4
CHAPTER 2: LITERATURE REVIEW
2.1 The Nigerian Roads and Ownership 5
2.2 Road Traffic Accident Trend 6
2.3 Significance of Road Accidents Prediction Models 7
2.4 Road Traffic Accident Factors 8
2.5 The Costs of Road Traffic Accidents 9
2.6 Daily Driven Distance and Age Factor on Road Accident 10
2.7 Accidents and Causes 12
2.7.1 The need to understand accidents 12
2.7.2 Accident causes 13
2.7.3 The need for accident causation models 14
2.7.4 Accident theories (models) 15
2.7.5 Laws of accident causation 18
2.8 Road Traffic Accident Prediction Models 19
2.8.1 Multiple linear regressions 20
2.8.2 Poisson regression 20
2.8.3 Negative binomial regression model (NB) 21
2.8.4 Poisson-lognormal regression model 22
2.8.5 Zero inflated poisson and negative binomial regression models 23
2.8.6 Logit and probit models 23
CHAPTER 3: MATERIALS AND METHODS
3.1 Model Concept and Assumptions 26
3.2 Data Collection and Presentation 26
3.3 Model Development Analysis 27
3.3.1 Predictive-factor variables 27
3.3.2 The Accident prediction model 28
3.4 Accident Predictive-Factor Combinations 30
3.4.1 Number of possible predictive-factor combinations 31
3.4.2 Case of conditioned combination 31
3.5 Development of the Accident Calculator Program 32
CHAPTER 4: RESULTS AND DISCUSSION
4.1 Road Traffic Accident-Cause Probabilities 33
4.2 Effects of the Accident-Cause Factors 34
4.3 Accident Predictive-Factor Combination Equations 39
4.4 Individual Predictive-factor Effects on ‘µ' 43
4.5 Model Validation and Performance 45
4.6 Accident Prediction Results 46
CHAPTER 5: CONCLUSION AND RECOMMENDATIONS
5.1 Conclusion 50
5.2 Contributions to Knowledge 51
5.3 Recommendations 52
References 53
Appendices 55
LIST OF TABLES
2.1 Structure of Nigerian roads and ownership 6
4.1 Road traffic accident-cause fraction (probability), pr 33
4.2 RTA-causes and probabilities by the three basic accident factors 34
4.3 Percentage accident contributions per year by basic accident factors 38
4.4 Analysis of the accident predictive factors 39
4.5 Accident predictive-factor combination 42
4.6 Accident probability of individual predictive-factors 43
4.7 Conditioned combination of major RTA cause-factors and accident probabilities 46
4.8 Accident probability (µi,j,k) of the various possible combinations 47
4.9 Model accident prediction calculator interface 48
LIST OF FIGURES
2.1 The DVE model 16
2.2 The domino theory 16
4.1 Clustered column chart of accident distribution from 2012 to 2016 35
4.2 Accident distribution by human factors, HF 36
4.3 Accident distribution by mechanical factors, MF 36
4.4 Accident distribution by environmental factors, EF 37
4.5 Accident probability of Ai 44
4.6 Accident probability of Bj 44
4.7 Accident probability of Ck 45
4.8 Percentage accident-occurrence probabilities 47
NOMENCLATURE
BFL Brake failure
BRD Bad road
DGD Dangerous driving
DAD Driving under alcohol/drug
DOT Dangerous overtaking
FTQ Fatigue
LOC Loss of control
MDV Mechanically deficient vehicle
NJR Night journey
OTH Others, unidentified causes
OVL Overloading
PWR Poor weather
OBS Road obstruction violation
RTA Road traffic accident
RTV Route violation
SLV Sign/light violation
SOS Sleeping on steering
SPV Speed violation
TBT Tyre burst
UPD Use of phone while driving
WOT Wrongful overtaking
Ҳ(.i,j,k) Fractional score of ith, jth and kth accident cause-factor, respectively
CHAPTER 1
INTRODUCTION
1.1 BACKGROUND OF STUDY
Road transport remains the chief universal means of transportation in Nigeria in comparison to air, rail, and water transport. Recent advances in technological development have made it possible for the evolution of different types and models of modern and aesthetic vehicles with greater comfort and maneuverability; in contrast to the pre-colonial means of transportation such as the use of animals (Gupta and Gupta, 2009). The influx of these vehicles and the expansion of fleet operation in Nigeria have crowded the Nigerian motoring environment, thereby making road traffic a major challenge to combat in the country (Oyedepo, 2010).
These road traffic challenges often result in road traffic accidents, most times, with its attendant carnages. In the recent time, road traffic accident tolls have been on the increase in Nigeria. Several factors are responsible for this; they range from carelessness of the drivers to the deplorable condition of our roads. The Nigerian roads have become 'killer-spree' with no protection for the users (Oyedepo, 2010). The travelers are often faced with the uncertainty of whether they would be able to reach their destinations and so become apprehensive of the journeys they make. This bothersome trend has great unfavorable effects on the nation's health system as well as her social and economic endeavors. The ease in movement of human and items, notwithstanding, so many families have been bereaved of their breadwinners and loved ones by the menace of road traffic accidents in Nigeria. As reported by the FRSC, not less than 88,520 road users lost their lives between 1991 and 2000 alone, most victims being between 20 and 40 years (FRSC and Balogun, 2006).
According to the FRSC Annual report 2016, between year 2012 and 2016 alone, about 57,894 road traffic accidents were recorded. Several factors as environmental, mechanical, human factors, etc. were responsible for the accidents (FRSC, 2016). Considering the precarious condition of the Nigerian roads, the poor maintenance culture of most transport vehicles as well as the unwholesome attitudes of most drivers, there is a dire need to treat road accident as a major issue that requires urgent attention in order to prevent untimely deaths; reduce health risks, social and economic impacts it poses on the Nigerian road users in particular and the society at large. Over 50 percent of the aggregate global road traffic deaths involve persons of ages 15 to 44; in their key productive years (Beirness, 2011). Furthermore, the disability load for this age group records about 60.0 percent of all disability-life years (Beirness, 2011). The consequences and costs of these losses are monumental. About 3/4 of the total deprived families who lost their loved ones in a traffic crash reported a decline in their livelihood, and about 61.0 percent reported that they had resorted to borrowing money to take care of expenses, consequent upon their loss (Beirness, 2011). World Bank report estimated that road traffic injuries constitute between 2 percent to 3 percent of the GDP of developing countries, or twice the total development aid given worldwide to developing countries (World Bank, 2015). Though transport agencies often try to identify the most hazardous road spots, and put enormous efforts into protective measures, the yearly traffic crashes toll has not hitherto been appreciably reduced (World Bank, 2015).
Road traffic accident prediction models are crucial implement in ensuring highway safety, considering its ability to determine both the crash frequency and the degree of severity of crashes (Abdulhafedh, 2017). Measures for useful interventions to trim down crash toll include design of safer road infrastructure and integration of road safety elements into land use and transport planning; upgrading of vehicle safety attributes; advancement of post-crash care for victims of road traffic crashes; and enhancement of driver behavior as well as raising public consciousness (Abdulhafedh, 2017). About 35,092 road traffic fatalities were documented in the US in 2015, an increase of 7.2% compared to the preceding year (Abdulhafedh, 2017). Considering these trends, this research attempts to develop a model that could predict the probability of a transport vehicle to have accident while on transit. This prediction model would be used by transport agencies/owners to minimize the risk factors that may contribute to road traffic accidents.
1.2 STATEMENT OF PROBLEM
Road traffic crashes are among the world's chief causes of deaths for people of the ages of one and twenty-nine. Each year, about 1.25 million persons are killed in auto vehicle accidents and about 50 million more are injured worldwide. Going by the existing trend, approximately two million persons are expected to be killed in motor vehicle accidents by 2030. Presently, road crashes occupy the ninth position of the most severe cause of death in the globe; if no new improved schemes are adopted for road safety, fatal road traffic accidents could probably rise to the third position by the year 2020. Traffic fatalities are on the increase especially in the developing countries. About 85.0 percent of yearly deaths and 90.0 percent of injuries occur in the developing countries due to road traffic accident (WHO, 2015).
Several accident models were developed before now, but all such models were post-accident models, and as such cannot forecast accident occurrence probabilities beforehand. The need to model a framework that can predict if a transport vehicle is likely to have accident on transit, therefore, becomes imperative. This model differs from the previous accident models in that it is a pre-accident model. The model therefore could predict accident possibility even before the transport vehicles leave its terminal. This could be achieved by way of combining several predictive factors and incorporating them into the model framework.
1.3 AIM AND OBJECTIVES OF STUDY
The aim of this research is to develop an accident prediction model for transport vehicles. The specific objectives include to:
I. Determine the effects of driver age, vehicle maintenance frequency and the distance of travel on the accident potential of vehicles in transit.
II. Develop an excel-based accident-prediction program interface for the developed accident prediction model.
1.4 SCOPE OF STUDY
The scope of this study covered the collection and collation of a secondary data on road accident causes from the relevant traffic management authorities/agencies in Nigeria, FRSC in particular, between 2012 and 2016. The data were analyzed to obtain a base parameter for the model development. The base parameters and the relevant combined accident predictive-factors were manipulated to arrive at the desired probabilistic and predictive model. The developed accident prediction model was validated on transport vehicles/operator parameters and the accident occurrence probability range of 0 to 1 was obtained. However, the study is limited only to terminal-based transport vehicles/operators; it excludes private and articulated vehicles/operators that are not terminal-based.
1.5 JUSTIFICATION FOR THE STUDY
The development of an accident prediction model for transport vehicles will enable transport vehicle/fleet operators to forecast the likelihood of their vehicles getting involved in accident while on transit. By combining the driver age, the maintenance frequency and the travel distance factors and integrating them into the model, the fleet operators could predict the likelihood of accidents involving their vehicles and drivers. If these accidents are predicted early enough, they would help to make the motoring public safer and reduce carnages and vehicle damages on the roads; hence reduce economic wastes, costs, casualties and fatalities.
Login To Comment