Background: A third of a billion children travel to school every day in India, yet little is known about this journey. Increasing motorisation in India is likely to have implications for road safety of children. This thesis develops methods to measure distance to school, transport modes, and risk of road traffic injury, on journeys to school in Hyderabad. Methods: Following a systematic review, a self-completion questionnaire was developed to estimate the distance and modes of travel to school in India. Its validity and reliability was assessed using the kappa statistic. A cross-sectional survey using a two-stage stratified cluster sampling design was conducted in government funded, government aided, and private schools in Hyderabad. The relationship between modes of travel and distance to school was analysed using logistic regression, adjusting for confounders. The prevalence of road traffic injury in the previous 12 months during school journeys was estimated, and the impacts of alternative transport scenarios on road injury was modelled. Results: The questionnaire provided reliable information on the usual mode of travel to school, and road injury. Distance to school measured by asking for the nearest landmark to a child’s home was found to be a valid measure of distance compared to a method based on in-depth interviews with children. Forty five schools including 5,842 children aged 11-14 years participated in the survey, with a response rate of 99%. Most children in Hyderabad walked or cycled to school. Others travelled by motorised 2-wheelers, auto-rickshaw, school bus, public transport bus, and car. Greater distance to school was strongly associated with the use of motorised transport. A sixth of all children reported a road injury during school journeys, which was strongly associated with travel mode and distance to school. The overall risk of road injury was 25/100,000 child km per year. Relative to school bus occupants, bicyclists, pedestrians and motorcycle passengers were more likely to be injured, for the same distance travelled. The model showed that road injuries can be prevented under transportation scenarios that restrict distance and motorised vehicles near schools. Conclusions: The questionnaire reliably measured mode of travel to school and estimated distances to school in Hyderabad. Most children walked or cycled to school and if these levels are to be maintained, there is an urgent need to ensure that walking and cycling may be done safely.

Legislation in Europe requires children younger than 12 or shorter than 150cm to use a child restraint system (CRS) when travelling in motor vehicles. These have decreased the number of children fatalities in car accidents, yet certain body regions such as the abdomen are still prone to severe trauma. One of the mechanisms by which injuries to this body region occur is called "submarining". It involves the seat-belt slipping over the pelvis onto the abdomen during rapid deceleration, therefore afflicting the abdomen with injurious loads. The motion characterising submarining with CRS is as of yet poorly defined, and although CRS are assessed for protection level, there is currently no established identification criteria for submarining. As part of the Enabling Protection for Older Children project (EPOCh), standard frontal impact sled tests of 10 CRS (6 high back booster seats; 4 booster cushions) with 10 year old anthropomorphic testing devices (ATDs) were analysed qualitatively, transversally and longitudinally for submarining detection. The methods used included video analysis, descriptive and inferential statistics, principal component analysis, time series analysis, as well as multiple linear regression and logistic regression, applied on both ATD trajectories and ATD instrumentation recordings (dynamic data). From the videos, trajectories and dynamic data, submarining motion is shown to embody an exaggerated slouching movement regardless of the CRS type. The observations and quantitative results confirm that exaggerated forward knee displacement is characteristic of submarining, as well as a very strong criterion for the latter's detection. No other individual trajectory or dynamic variable distinguish clear submarining behaviour, however the regression analyses on dynamic variables establish the association of the pelvis, lumbar and chest as representative of the knee displacement, and demonstrate the combination's capacity to distinguish submarining cases. These findings establish the complexity of the movement involved in submarining and the potential of using current ATD instrumentation for its assessment with CRS. This opens a path for an integral approach to ATD movement in CRS appraisal and suggests considering pelvis, lumbar and chest motion control for submarining prevention.

Transport safety has a direct impact on people’s lives despite considerable improvements in recent decades. By treating transport modes independently and not taking full account of modal options available, policymakers have overlooked potentially important and low cost contributions to overall passenger safety. This thesis investigates the extent to which Cross Modal Switching (CMS), encouraging users to transfer to safer modes of transport, can be deployed as an instrument of transport safety policy. Research was conducted to establish the safety differences between modes on specific journeys, taking account of composite risks including all transport modes used. Primary research used surveys and qualitative interviews to target three different groups to understand their views on transport safety, willingness to switch modes, reasons that would cause them to switch and modal perceptions on risk and travel behaviour. The feasibility of promoting CMS was assessed by measuring substitutability between modes and calculating cross-elasticities using data from the empirical surveys conducted and previously published work. Cost benefit calculations were made using monetised risk and the cost of fares subsidies to assess the net safety benefits for three selected journeys. This analysis shows that there is a marginal justification for CMS as a tool within an overall integrated transport policy that considers safety in all modes simultaneously. This must recognise that the absolute safety benefits are not very large relative to other benefits, although the relative size depends on the manner in which changes of consumer surplus are treated in the CBA. CMS can be demonstrated to be cost effective in low risk modes, relative to larger infrastructure investments only yielding marginal safety improvements. Further research, using a larger sample of journey net benefit calculations, is thus required to validate the case robustly for CMS, identifying beneficial opportunities for modal switching on specific routes and target modes.

With the ever-growing traffic density, the number of road accidents is anticipated to further increase. Finding solutions to reduce road accidents and to improve traffic safety has become a top-priority for many government agencies and automobile manufactures alike. It has become imperative to the development of Advance Driver Assistance Systems (ADAS) which is able to continuously monitor, not just the surrounding environment and vehicle state, but also driver behaviours. Dangerous driver behaviour including distraction and fatigue, has long been recognized as the main contributing factor in traffic accidents. This thesis mainly presents contributing research on vision based driver distraction and fatigue analysis and pedestrian gait identification, which can be summarised in four parts as follows. First, the driver distraction activities including operating the shift lever, talking on a cell phone, eating, and smoking, are explored to be recognised under the framework of human action recognition. Computer vision technologies including motion history image and the pyramid histogram of oriented gradients, are applied to extracting discriminate feature for recognition. Moreover, A hierarchal classification system which considers different sets of features at different levels, is designed to improve the performance than conventional "flat" classification. Second, to solve the effectiveness problem in poor illuminations and realistic road conditions and to improve the performance, a posture based driver distraction recognition system is extended, which applies convolutional neural network (CNN) to automatically learn and predict pre-defined driving postures. The main idea is to monitor driver arm patterns with discriminative information extracted to predict distracting driver postures. Third, supposing to analysis driver fatigue and distraction through driver's eye, mouth and ear, a commercial deep learning facial landmark locating toolbox (Face++ Research Toolkit) is evaluated in localizing the region of driver's eye, mouth and ear and is demonstrated robust performance under the effect of illumination variation and occlusion in real driving condition. Then, semantic features for recognising different statuses of eye, mouth and ear on image patches, are learned via CNNs, which requires minimal domain knowledge of the problem. Finally, works on pedestrian subject identification using convolutional neural networks( CNNs) and multi-task learning model(MTL), is presented additionally. Gait identification is strongly motivated by the demands of security that require automatically identifying person at a distance. This be particularly relevant with respect to police/detective vehicle that is tracking criminal.

A review of maritime accidents conducted in 2006 confirms that human error is the main contributing factor in maritime accidents. This study illustrates that major maritime accidents are not caused by technical problems but by failure of the crew to respond to the situation appropriately. Non-technical skills (NTS) encompass both interpersonal and cognitive skills such as situation awareness, teamwork, decision-making, leadership, managerial skills, communication and language skills, etc. In a crisis situation good NTS allow the deck officers to recognise a problem quickly and manage the situation and team safely and effectively. As a result, the evaluation and grading of deck officers’ NTS is necessary to assure safety at sea. This research aims to identify the links between maritime accidents and deck officers’ NTS and identifies significant criteria and their contributions to the deck officers’ NTS by using the Formal Safety Assessment concept. Taxonomy of deck officers’ non-technical skills was developed by conducting interviews with experts and collecting NTS weighting data for calculating each NTS weight by the AHP (Analytical Hierarchy Process) method. Based on the taxonomy of the deck officers NTS behavioural markers were developed for the assessment of their NTS in the bridge simulator. A set of bridge simulator crisis scenarios was developed to assess deck officers’ NTS.Two sets of Chief Mate volunteer students’ NTS performance was assessed in the bridge simulator. One set of students are those who have not obtained NTS training i.e. HELM (Human Element Leadership and Management) and the other set of students are those who have obtained the HELM training as part of their main course of study. All groups’ NTS performances are calculated by the ER (Evidential Reasoning) Algorithm and are compared to see if there are any improvements in the NTS performance with the HELM training. After comparison it was found that NTS performance of the groups with HELM training was only 0.8% better than the NTS performance of the groups without HELM training. HELM course effectiveness is evaluated and suggestions are given for further improvements to the course. Cost benefit analysis for improving deck officers’ NTS was carried out by Bayesian Network and Decision Tree Model.

Implementation intentions are IF-THEN plans that facilitate the performance of intended behaviours by linking critical situations in which to behave with goal-directed responses, or strategies, that ensure successful behavioural performance. This research investigated the effect of implementation intentions on drivers’ speeding behaviour. Chapter 1 discusses the negative impact of speeding on road safety, the economy, the environment and health and wellbeing. Chapter 2 explores the influence of motivation (e.g., goal intentions) and habit on speeding behaviour and concludes that a substantial proportion of drivers speed despite holding generally positive intentions to avoid speeding. Chapter 3 introduces the concept of implementation intentions and reviews research suggesting that they could be useful for reducing speeding. Chapter 4 presents study 1, in which the effect of implementation intentions on self-reported speeding was tested while addressing several limitations with the existing evidence-base. Implementation intentions were effective at reducing speeding and moderated the past-subsequent behaviour and goal intention-subsequent behaviour relationships, in line with the idea that implementation intentions can weaken habits, thereby allowing drivers to behave in accordance with their goal intentions. Chapter 5 presents study 2, in which a driving simulator was used to test the effect of implementation intentions’ on objectively measured behaviour and to test the extent to which the effects of implementation intentions generalise from the situations specified in the IF component of the plan, to unspecified situations. Behaviour-change occurred in specified situations and, also, contextually similar unspecified situations. Chapter 6 presents study 3, which focused on the THEN component of implementation intentions. The most effective type of goal-directed response for reducing speeding was explored. No effect of implementation intentions was observed. Potential reasons are discussed. Chapter 7 presents the implications for road safety and future research.

The primary objectives of the research programme described in this thesis ( and commissioned by the United Kingdom Civil Aviation Authority) were to evaluate the effectiveness of safety cards for communicating safety information to passengers and to assess the influence of possible alternative styles of safety briefing on passenger attention. These objectives were investigated in two separate simulated pre-warned ditching situations on board a stationary aircraft. Participants' knowledge of the less complicated aspects of safety information, such as the location of oxygen masks or how to inflate the lifejacket, was generally high but comprehension of more complex procedures, such as the correct method of donning the lifejacket, was more limited. A visual demonstration was shown to significantly increase the likelihood that participants would know the correct method of donning the lifejacket, operating the oxygen mask and adopting the brace position correctly. A number of human factors problems, most frequently due to lack of specific information, were also identified as adversely affecting participants' ability to carry out safety procedures. Passengers' motivation to pay attention to safety information would be likely to be enhanced if briefings were more meaningful to them, for example, personalised and less repetitive, or if they were given the opportunity to practise safety procedures in a low stress situation. Participants' opinions indicated that emphasis on the importance of passengers knowing how to operate i terns of safety equipment would not put the majority of passengers off flying and would be likely to encourage attention to safety briefings and cards. The evaluation of the safety cards and briefings indicated that no one card or briefing was more effective overall. Some general principles for the design of safety cards have been suggested, such as the consultation with non-aviation personnel when designing safety cards and the inclusion of brief statements to explain complex procedures. The lack of major differences between the safety briefings may be attributable to the conventional manner in which they were presented suggesting that a different type of presentation style, for example one which personalises or makes the information more meaningful to passengers, would be more likely to encourage attention. A secondary objective of the thesis was to explore the influence of perceived locus of control on passenger comprehension of safety information and on their perceptions of aircraft accident survivability. This was investigated in a survey of airline passengers in addition to the two simulated ditchings. Analysis of a specifically designed locus of control questionnaire indicated that locus of control had only a slight influence on participants' knowledge of safety information and on their perceptions of aircraft accident survivability. The dominant cultural image of aircraft accidents as being non-survivable would appear to obscure any possible influence of locus of control as a predictor of passenger emergency behaviour. Locus of control would appear to be more effectively utilised in situations where cultural definitions are less dominant and perceptions of chance and skill influences are more equally balanced. Air travel was considered to be the safest form of transport but aircraft accidents were perceived to be less survivable than accidents involving other forms of transport. Participants also tended to underestimate their chances of survival in a range of potential aircraft accident situations. In order to improve the accuracy of passengers' perceptions of aircraft accident survivability a more realistic image of aircraft safety is needed. The stability of perceptions of locus of control and the survivability of aircraft accidents observed in the participant groups indicates that attitudes and beliefs are firmly entrenched in the general population and that major efforts will be required to influence attitudes and beliefs. Any attempt to change attitudes towards aircraft accident survivability would need to begin with improving the media image as the media are the most influential information source in forming public opinion regarding accident survivability.

In-car distractions have always been present in one form or another. There is no doubt that passenger conversations impaired drivers historically, and continue to do so; however, the new incarnations of low-cost microprocessor controlled nomadic and integrated equipment, compound upon, not replace those historic distractors. The consequence is that these highly distracting secondary tasks pervade and compete for drivers' finite resources, and thereby contribute to collisions that result in deaths and serious injuries for society to injure. To better understand the impact of distractions on the most impressionable driving cohort, an even gender-split of young (18-24) Northern Ireland residing drivers, were asked to participate with a laboratory based experiment, which was also complimented with an on-line survey. The Laboratory experiment was primarily based on the ISO 26022:2010 lane change test (LeT) and the prototype ISO 17488 detection response task (DRT); and data was collected for driver impairment (car control and selective attention) and driver workload (objective: Reali Rate Fluctuation, and subjective: NASA TLX). This study found that complex visual distractors impaired selective attention and car control, but the latter impairment was disproportionately extreme; whereas, complex auditory distractions also had a significant degree of selective attention impairment, but had a limited effect on car control; furthermore, the compounded interaction between distraction types was observed for the texting task, which required both physical and complex visual inputs, resulting in an impairment that was exaggerated by the influence of the complex visual distraction type. Therefore, the degree and manifestation of driver impairment, was directly relative to the distraction type and severity of the secondary task performed whilst driving, and the compounded impairment caused with secondary tasks with more than one sensory distraction type, was predominantly defined by the most influential component.

Eco-driving has the potential to reduce fuel consumption and therefore emissions considerably. Previous research suggests that drivers already possess a certain level of eco-driving capability, which they do not practise in their everyday lives. The studies reported in this thesis are based on a mental models approach, which enabled an in-depth exploration of eco-driving knowledge and skills and broadened the understanding of the underlying cognitive mechanisms. This thesis describes two driving simulator experiments aiming to measure, activate and ultimately interrupt eco-driving mental models in a variety of scenarios, relevant for safe and eco-driving. The studies used simple driving task instructions, text message interventions as well as a workload task. Changes in the participants’ behaviour and thoughts were analysed with a combination of quantitative and qualitative methods. The results suggest that drivers have mental models of eco-driving on several levels, ranging from knowledge and strategies to tactics in specific situations to automated behaviour. However, in the first experiment they did not use them when they were instructed to ‘Drive normally’. In the second study text message primes and advice provided over two weeks were not able to replicate the effect of experimental instructions given directly before driving. Behavioural changes following these instructions were abandoned when performing a workload task, and not resumed afterwards. Future research needs to consider alternative methods to prompt drivers to use their existing eco-driving knowledge and skills. Studies with a larger number of participants, and in real-world settings can then validate findings. It is suggested to expand the mental models approach into other fields such as sustainable transport in general.

No description available.