Search

Reported cyclist casualties are disproportionally high relative to their modal share. This is a well-documented problem and yet we currently do not have local or national models that can estimate exposure (i.e. intensity of travel by bike). There is therefore is limited capacity among practitioners and professionals to estimate normalised risk at a disaggregate level and there is little evidence available for network level risk factors associated with cyclists' safety. Simultaneously an increasingly popular vulnerable road user policy development is the safety in numbers effect. The main theory behind safety in numbers is simply that more cyclists, or pedestrian activity, reduces the overall risk of having an accident. This research investigates whether there is a safety in numbers effect in Scotland; examines if there are wider spatial, demographic and policy differences affecting cyclists; and develops a novel modelling method to estimate cyclist exposure based on open data and open software. A comparison of traditional road safety macro-level global regression models, with local meso-level geographically weighted regression models to investigate safety in numbers was used to explore the nature of the safety in numbers effect in Scotland. The comparison of the global and local model forms yielded four main results. First, local models' account of spatial dependence provide a better statistical fit than the traditional global models. Second, both the global and the local models confirm that there is a safety in numbers effect in Scotland but that the effect is less than reported in the literature and referenced in Scottish policy documents and guidance. Third, the local models confirm that safety in numbers is not static and that the effect varies spatially, depends on local infrastructure factors and the intensity of cycling activity. Finally, a safety in numbers effect can co-exist with hazard in scarcity, weaker safety in numbers effects were found among women and between injury severity levels. Edinburgh was identified as an urban area with high potential for a safety in numbers effect within Scotland because, unlike across the most of Scotland, cycling doubled between 2001 and the 2011 census and is likely to double again by 2021 given current trends. The results found that there is a safety in numbers effect in Edinburgh for slight casualties but that there is little to no effect for killed or serious injuries (KSIs). The strength of the effect (i.e. less cycling risk) is associated with higher concentrations of some types of cyclist infrastructure but not others. Unprotected on-road cycle lanes, advanced stop lines and bus lanes were not positively associated with improved cyclist safety, however quiet routes, offroad cycle lanes and segregated facilities were found to be safer. Therefore, despite higher cycling activity, Edinburgh does not yet benefit substantially from a safety in numbers effect. This confirms that cycling numbers alone do not produce safety in numbers; and effective and ineffective cycling infrastructure was also identified. A further finding and benefit of using spatial modelling is the visualisation of safety in numbers in a local context to identify where it does or does not manifest and this also facilitates evaluation of facilities and other policy interactions or factors. Furthermore, the safety in numbers effect can be used as a Safety Performance Function to assess road safety which is a superior metric than rate-based measures. This has not previously been demonstrated in the literature and therefore this research contributes and adds to the understanding of the safety in numbers effect and demonstrates the need to develop cycling flow models to provide evidence based research.

The current debate on transport policy in the UK is focused on the need for a sustainable transport system. Buses play a vital role in achieving this, as they are the most frequently used and most accessible mode of public transport. However, the literature shows that the delivery of sustainable transport policies is not producing the desired outcomes (Hull, 2009) and the application of such policies in real situations remains inconsistent. This is evident across the UK where there has been a decrease in bus patronage and bus mileage. To address this gap, the aim of this research is to identify why bus policies are not imple-mented successfully at a local level and to provide recommendations for implementation and decision making that will aid policy makers, local authority staff, regional transport partnerships, bus operating companies and other practitioners working within the field of transport. A mixed methodology was chosen for this research and is divided into three key stages to address the research problem. The first methodology included an online ques-tionnaire and 143 questionnaires were sent to all public transport officers in Great Britain. 80 surveys were returned giving a response rate of 56%. The second methodology in-cluded telephone interviews conducted with 10 of those public transport officers who responded to the questionnaire in order to elicit a deeper understanding of the results, which could not be achieved from the questionnaire results alone. Finally, the third meth-odology included four case studies on specific bus schemes within Great Britain. These case studies were the Quality Contract Scheme in Tyne and Wear, Fastlink Scheme in Glasgow, Bus Priority Scheme in Solihull and Smart Ticketing Scheme in Dundee. While the questionnaires and telephone interviews provide an overview of bus policy imple-mentation across Great Britain, the multiple case studies were required to investigate the topic in depth, thus identifying the greatest barriers to bus policy implementation. Analysis of the three sets of data is based on the application of a new decision support frame-work developed in this research. The findings in this thesis reveal that local authorities in Great Britain are under-performing in the implementation of bus policy due to the barriers they face. The greatest barriers to implementation include the lack of a policy document; the characteristics of the organisation; availability of resources; intra-organisation support and communication; economic, social and political environments; and opposition, conflict, and ambiguities. Overall, this research has identified several concerns with bus policy implementation. The most obvious concern is the unclear link between policy objectives and measures and the setting and monitoring of performance targets. Meanwhile, the deregulation of the bus sector in the UK means that, in some cases, a lack of control over the implementation of certain measures places limits on policy implementation and results in the frequent im-plementation of policy measures that are achievable rather than those that are necessary to the achievement of policy objectives. The findings from this research also help policy-makers and transport planners to predict what makes implementation successful and to address problems and issues through improved policies and regulations, as well as to an-ticipate and plan for likely barriers. Moreover, addressing these barriers can help tackle the decline in bus mileage and bus usage across Great Britain.

Driver behaviour has become an important aspect of transport research and over the years a considerable number of car following models have been developed. However, many of these models do not accurately simulate actual driving behaviour due to a lack of suitable qualitative and quantitative data. Moreover, the inclusion of socioeconomic variables in the existing models to ascertain the effect on car following behaviour is lacking. This research underlines the need to further investigate driving behaviour and car following models and to develop techniques to provide a better understanding of driver-vehicle interactions during car following. It investigates data collection techniques and develop better techniques to enhance and improve the collection of microscopic driver behaviour and traffic flow data. This study developed a novel data collection technique which involved instrumenting a private vehicle with front and rear advanced radar sensors, both forward and rear facing video-audio recorders connected to GPS based time series speed and distance measurement devices, an in-vehicle computer logging vehicle speed and a CAN monitoring interface user program to provide real time monitoring and display of data. This system has been utilised to collect a more enhanced and reliable microscopic driver behaviour data in three consecutive vehicles movements which represents an improvement from previously used systems. Three different versions of the GHR car following model were produced for: car following car, truck following car and car following truck. Further analysis of the GHR model showed that in the case of car following car, car drivers responses to the lead car are more obviously stronger than in the case of truck following a car. A distance-based car following model and distance-based two-leader car following model that predict the safe following distance of following vehicles were developed to provide a better understanding of driver behaviour. An extension of these models to include gender, corridor (road) type and vehicle occupancy showed evidence of statistical significance of these variables on driver behaviour. A bus following model that predicts the “following distance” also has been calibrated to describe the interactions between a bus and a car within urban-rural driving conditions. In addition, data analysis showed that drivers were inconsistent with their driving behaviour and that there was variability in driving behaviour across the drivers observed in keeping a safe or desired following distance. This study provides a platform for a number of future research agendas including data collection techniques for collection of driver behaviour data; evaluation of different ITS technologies; impact assessment of ACC on driver safety and improvement of traffic microscopic simulation tools in order to strengthen their ability to simulate realistic transport problems for efficient and effective transportation systems.

Climate change is a significant issue in today's society as countries work towards decarbonising the economic sectors that contribute to significant greenhouse gas emissions. The electric vehicle (EV) is proposed as a solution to reduce the level of emissions in the transport sector. However, if an EV is powered by an electrical fossil fuelled source, their penetration into the UK market will have minimal mitigating effects, as emissions will simply shift from the transport sector to the energy production sector. Limited research has evaluated the loss of propulsion energy as a result of operating on-board climate control systems, and has focused more on traction energy. Unlike conventional fossil fuelled vehicles, EVs do not produce waste heat to warm the interior space of the vehicle. The present research found that up to 30% of a vehicle's total energy consumed per trip is allocated to heating requirements, thus the present research developed a temperature predicting numerical algorithm to compute indoor cabin temperatures. The vehicle was exposed to ambient climate conditions with an auxiliary heating or cooling system to evaluate this thermal model. The numerical algorithm could predict the temperature of a cabin space under solar space heating conditions with 62% more accuracy than previously developed models when comparing the Root Mean Square Error performance indicator. The presently developed temperature prediction algorithm may be applied to a route planning application, thus indicating the electrical energy required by the vehicle's battery for users to increase or decrease the desired temperature level. Additionally, this study investigated the ability of a renewable energy resource to decarbonise the vehicle's built-in climate control system. Integrating solar panels on the roof and bonnet of an EV to power an auxiliary climate control system reduced the electrical loading required to reach the occupant's thermal comfort. By installing an auxiliary heating system to increase cabin temperature by 2 or 5°C, the present research found that energy consumption of the built-in climate control system was reduced by 22% or 57%, respectively. This illuminates the potential an auxiliary climate control system has in improving the thermal performance of EVs.

Vegetation growing on railway trackbeds and embankments present potential problems. The presence of vegetation threatens the safety of personnel inspecting the railway infrastructure. In addition vegetation growth clogs the ballast and results in inadequate track drainage which in turn could lead to the collapse of the railway embankment. Assessing vegetation within the realm of railway maintenance is mainly carried out manually by making visual inspections along the track. This is done either on-site or by watching videos recorded by maintenance vehicles mainly operated by the national railway administrative body. A need for the automated detection and characterisation of vegetation on railways (a subset of vegetation control/management) has been identified in collaboration with local railway maintenance subcontractors and Trafikverket, the Swedish Transport Administration (STA). The latter is responsible for long-term planning of the transport system for all types of traffic, aswell as for the building, operation and maintenance of public roads and railways. The purpose of this research project was to investigate how vegetation can be measured and quantified by human raters and how machine vision can automate the same process. Data were acquired at railway trackbeds and embankments during field measurement experiments. All field data (such as images) in this thesis work was acquired on operational, lightly trafficked railway tracks, mostly trafficked by goods trains. Data were also generated by letting (human) raters conduct visual estimates of plant cover and/or count the number of plants, either on-site or in-house by making visual estimates of the images acquired from the field experiments. Later, the degree of reliability of (human) raters' visual estimates were investigated and compared against machine vision algorithms. The overall results of the investigations involving human raters showed inconsistency in their estimates, and are therefore unreliable. As a result of the exploration of machine vision, computational methods and algorithms enabling automatic detection and characterisation of vegetation along railways were developed. The results achieved in the current work have shown that the use of image data for detecting vegetation is indeed possible and that such results could form the base for decisions regarding vegetation control. The performance of the machine vision algorithmwhich quantifies the vegetation cover was able to process 98% of the image data. Investigations of classifying plants from images were conducted in in order to recognise the specie. The classification rate accuracy was 95%. Objective measurements such as the ones proposed in thesis offers easy access to the measurements to all the involved parties and makes the subcontracting process easier i.e., both the subcontractors and the national railway administration are given the same reference framework concerning vegetation before signing a contract, which can then be crosschecked post maintenance. A very important issue which comes with an increasing ability to recognise species is the maintenance of biological diversity. Biological diversity along the trackbeds and embankments can be mapped, and maintained, through better and robust mo nitoring procedures. Continuously monitoring the state of vegetation along railways is highly recommended in order to identify a need for maintenance actions, and in addition to keep track of biodiversity. The computational methods or algorithms developed formthe foundation of an automatic inspection system capable of objectively supporting manual inspections, or replacing manual inspections.

The thesis reports the outcome of three main investigations. The first investigated the technical issues involved in how commuter's departure time and mode choice might be affected by road user charging. The second investigated the public's preferences for allocating revenues from a road user charging scheme. Both were based on surveys in Edinburgh, where plans for a particular scheme had been developed1. The third assessed how different charging levels affected both the demand for travel by car and the revenue generated, raising questions as to how the balance between differing objectives might be struck. The surveys undertaken used stated preference techniques to investigate departure time and mode choice changes for a sample of 211 employees working in the city centre of Edinburgh in June 2002.Departure time to work was found to be influenced not only by the work schedule (hitherto the only component of scheduling flexibility) but also by other social factors. These included whether or not regular activities were carried out before or on the way to work; and whether or not a respondent had dependent children where all adults in the household were working. The highest levels of scheduling flexibility were observed for individuals who had flexibility in both their work schedules and non-work activities. The calibrated models support the relevance of aspects of flexibility to the departure time choice. Mode choice was modelled using data from stated preference experiments. It was shown that many car drivers were prepared to switch to the bus when faced with paying a toll. Moreover, a small percentage of bus users were prepared to switch from the bus to the car, perhaps seeing an opportunistic reduction in traffic levels, which would make the car a viable alternative. Respondent's preferences for the allocation of revenues from a road user charging scheme were jointly modelled with choice of mode of travel using the nested logit model. The justification for this is that the allocation of the revenues might affect other choice decisions. The overall goodness of fit of the model was reasonable, although not all of the model coefficients were statistically significant. However, the approach to jointly modelling mode choice and revenue allocation offers an interesting concept, which is worthy of further research. The effect of different levels of charge on traffic levels and revenue raised was investigated based on the three levels of charge, which were offered to all respondents. The findings suggest that there would be a slight increase in traffic at the lowest level of charge (in this case £2), while traffic reduces significantly at each of the next two levels of charge (£3.50 and £5). On the other hand, the revenue from the scheme increases with increasing levels of charging (up to £3.50), and then decreases with the decrease in demand. Therefore, based on this pattern, alternative optimum levels of charge can be identified to achieve different objectives. For example, in this case, if the objective is to maximise revenues, then the optimum level of charging appears to be about £3.50 - somewhat above the level of charge of £2 proposed in the Edinburgh scheme. If the objective is to reduce traffic levels, that would certainly require a higher charge than £2, but what the optimum level of charging might be cannot be determined for such an objective, since the quantity demanded still appears to be falling even when the charge is £5.The main principle behind the theory of road user charging is that motorists should be charged for using the roads, where the extent of the charge reflects the costs that they impose on others and on the environment (thus helping to reduce traffic congestion). However, the experience from other studies suggests that in almost all cases the revenues generated are as relevant. These two objectives are conflicting; there is a need for further research and development of methodologies for the investigation of optimum specifications of schemes that reach a satisfactory compromise between the two objectives, taking into account the fact that raising revenue helps to provide funds that (under a hypothecated scheme) can be ploughed back into transport, and hence help reduce the demand for road use. Moreover, there is a need for an explicit recognition of the relevance of the revenues for the success of the scheme in terms of technicality and public acceptability.