Despite the increasingly stringent type approval limit values for vehicle emissions no quantitative difference has been seen in roadside concentrations of NOX concentrations (Carslaw et al., 2011b). This thesis aims to improve the ability of measuring and modelling the NOX emissions of passenger cars in urban environments by taking an in depth look at the emissions of vehicles observed in real driving environments over a number of years in Aberdeen, Cambridge, Leeds and Sheffield using a Remote Sensing Device. The remote sensing device is tested under controlled conditions to ascertain its measurement accuracy for both pollutants and vehicle specific power. A mathematical distribution function for describing the emissions of a fleet is presented and shown to be a good description of over 90% of the vehicles and a superposition of two of these distribution functions was able to describe the distribution of the rest of the fleet’s NO emissions with a high degree of accuracy. The distribution functions derived for one city were used to create a predictive model to determine how the average emission of a passenger car fleet vehicle performance would evolve over time showing that by 2025 a ≈ 30% reduction in NOX could be expected if the fleet was allowed to evolve naturally. In addition to these results a number of real world problems were assessed using the new framework developed in this thesis. The emissions of taxis compared to privately owned vehicles was assessed with taxis being shown to emit ≈ 50% more NOX than their equivalent vehicles in the fleet. The Volkswagen Group scandal, #dieselgate, is discussed and the data that the remote sensing device has been used to assess the real driving emissions of VWG passenger cars fitted with the EA189 engine. The observations show that whilst VWG vehicles are exceeding the limit values in real driving environments, they are observed to have equivalent or lower emissions factors than other marques.

With an increasing divergence between test certified vehicle emission and on-road vehicle emission, the accurate appraisal of real-world vehicle emission has arguably never been more important. This research advances the development, calibration and validation of a coupled traffic micro-simulation and instantaneous emission model. A portable emission measurement system (PEMS) data set was used to quantify the real-world CO2 emission of a passenger car through an urban traffic network, in Leeds (UK), over the diurnal range of traffic flow conditions. Utilising these data, analysis was conducted to assess the accuracy of CO2 micro-scale emission estimates by four emissions models; the Emissions Factors Toolkit (EFT), the Handbook Emission Factors for Road Transport (HBEFA); the Motor Vehicle Emission Simulator (MOVES); and the Passenger car and Heavy duty Emission Model (PHEM). The results demonstrated the strength of power based models over average speed based methods. The study identified the influence that road grade can have on the micro-scale modelling of exhaust CO2 emission and developed a novel methodology to incorporate grade into the modelling. A coupled traffic simulation and instantaneous emission model was developed for the study area, using the simulation tool AIMSUN and the emission model PHEM. The model was separately calibrated and validated for five time periods. Parameters within the coupled model, such as the vehicle fleet and vehicle dynamics were customised for the specific study area and time period. The research demonstrated the ability of the AIMSUN-PHEM model to deliver an accurate appraisal of on-road CO2 emission for each time period and its capacity to generate a range of emission factors from those of a single vehicle to those of the entire fleet. This work confirms that a properly calibrated traffic simulation emission model can provide an effective method for conducting high resolution analysis of vehicle emission in a network.

Asthma is the most common chronic disease of childhood. In this thesis, I investigated whether there is an association between traffic-related air pollution (TRAP) and the development of childhood asthma, quantified the magnitude of this association and estimated its public health impact in Bradford, UK. For these purposes, I conducted a systematic review and a meta-analysis. I then developed a new vehicle emission model to estimate traffic NOx and compared it to the standard European model. Subsequently, I set up and validated two full-chain health impact assessment models; linking distinct traffic, emissions, atmospheric dispersion and health impact models. Each full-chain model was underlined by a different vehicle emission model, the new and the standard one, and as such I tested the sensitivity of final air quality and health impact estimates to the vehicle emission estimates. I estimated the childhood population exposure to NO2 and NOx at the smallest census tract level and quantified the annual number of asthma cases associated with these exposures, whilst disentangling the impacts of traffic-related NO2 and NOx, and also the impacts of traffic-related NO2 and NOx specifically from minor roads and cold starts. I compared the full-chain models’ estimates to estimates from commonly used land-use regression models which further provided exposure and health impact estimates for black carbon, PM2.5 and PM10. I quantified positive and statistically significant associations for black carbon, NO2, PM2.5, PM10 and risk of childhood asthma. The association with NOx was positive but not statistically significant. I showed that the new vehicle emission model, as compared to the standard model, resulted in different source apportionment and higher emissions at low average speeds. These differences, however, did not translate into meaningful differences in air quality or health impacts, partly due to limitations in the traffic data which underestimated congestion. The full-chain models estimated NO2 and NOx with satisfactory predictive power but resulted in lower exposures and health impacts as compared to land-use regression. I estimated that 15% to 38% of all asthma cases in Bradford may be attributable to air pollution. Up to 6% and 12% of all cases were specifically attributable to TRAP, with and without minor roads and cold starts, respectively, but this percentage was underestimated. Full-chain health impact modelling was demonstrated as a valuable but underutilized tool to estimate the burden of disease associated with TRAP and to test the impacts of specific policy scenarios with a temporal and/or spatial element. There is a further need to improve the feasibility, utility, resolution and validity of the supporting data and the full-chain modelling approach, especially by addressing its underestimation of TRAP, and consequently, the associated health impacts.

This research is about power. It is completely immersed in modern societies, reflections on power have not settled its definition. Regardless of centuries of considerations about power, there is still much to learn about its exercise. For decades, power has been analysed by scholars all over the world; however, transport planners have avoided issues concerning power. This dissertation focuses on the use of power during planning and implementation processes of two Bus Rapid Transit Systems; one in Quito, Ecuador, and another in Cambridge, England. Using a multimethod phronetic approach, this research examines how decisions were made for the implementation of BRT in the selected case studies. Phronetic methodology aims to explain social phenomena by piecing together large and small details that shape the context of events; in this case, existing planning documentation and the narratives of key stakeholders and decision-makers —such as former Mayors, Council Members, and Heads of Transport Departments— were used to understand the reasons behind the adoption of bus systems. Despite the contextual differences of geography, population, and political and administrative terms between Quito and Cambridge, there are key themes found in both case studies. Solving congestion and improving economic growth is a key motivation for the implementation of the systems in both cases. A complex network of actors is formed during the planning processes of BRT systems that shape the way decisions are made. In both cases, an isolated group of actors that lack the opportunity to exercise power was also identified. A sophisticated mixture of power mechanisms were discovered, which contained actors that had more experience, training and opportunities to exercise power. The findings of the analysis of power in Quito and Cambridge suggest that stakeholders with opportunities to exercise power —from the beginning— are also actors with more opportunities to influence the final uptake of the systems. Planners interested in participatory processes need to focus efforts to involve different communities as early as possible in the planning processes. The inclusion of communities does not guarantee that their needs and objectives will be incorporated into the planning process. A set of power mechanisms need to be developed by all members involved in the process. An early participation of these communities can help focus planning on solving people’s needs rather than the implementation of a specific scheme.

The effectiveness of police strategy in influencing motorists’ choice of speed must be questioned because speeding has remained a consistent factor in accidents to this day. In light of this persistence, the objective of this research is to develop more effective speed enforcement by investigating the effects of police presence, motorists’ training, and roadside publicity on motorists’ choice of speed. These factors are the most widespread interventions implemented by police all over the world, although only a few have investigated how effective these interventions, in fact, are either as a single or as a combined intervention. This study was conducted in Indonesian road and applied a factorial experiment design where police, training and publicity were operated as the intervention factors. Participants’ responses were recorded during driving throughout prearranged test routes. Further, a traffic survey and questionnaires for motorists and police officers were utilized to support the study’s results. A survey of motorists shows that they are aware of the consequences of speeding, although prefer softer approaches to handle it, while a survey of police forces shows a high level of satisfaction for existing measures, including the new proposed speed enforcement program. There are still many aspects yet undiscovered that correlate to public attitude and police officer job satisfaction to current speed enforcement methods. One important finding of this research is the discovery of a three-way interaction effect on dual carriageways, which means that the addition of training and publicity to police interventions increases the effectiveness of speed enforcement. However, on single carriageways, the only significant effect was produced by the police presence. The combination of three factors not only reduced mean speed by 14% and 10% on dual and single carriageways respectively but also increased compliance by 72% and 33% until the end of test route. The estimated fatal casualty reductions are 52% and 33% for single and dual carriageways. Surprisingly, the training as the single factor has increased travel speeds on dual carriageways, although the effect has changed drastically when combined with police and roadside publicity. This finding could only be justified by the fact that training increase motorists’ confidence. Thus, we need guidance to avoid the confidence bias. Also, different distances of halo effect were observed on both routes in relation to the combination of intervention applied. Finally, the result shows that there is potential for further development of speed enforcement programs by combining training and roadside publicity into police enforcement. In addition, this study also proposes a number of policies so that enforcement agencies can increase the effectiveness police enforcement.

Speed is at the core of the problem of road safety, and speed management is a tool for improving road safety. Speed limits that are credible encourage drivers to comply with them, with consequent benefits for road safety. Credible speed limits have been found to be affected by the features surrounding the road by previous research in the Netherlands. However, not a great deal of empirical work has been done evaluating what a credible speed limit is for a given road layout and roadside environment based on motorists’ perceptions. This thesis builds a research model to link road environment, speed limit credibility, risk perception and speed limit compliance as a whole. The research presented here aims to verify the model. To do this, three separate experiments are used. Experiment 1 investigates, using a questionnaire, whether the current national speed limit is credible for a variety of UK road environments, and what the difference is between the speed limit and the chosen, self-reported, driving speed. The survey results reveal that road layout and roadside environment affect the intrinsic perception of the appropriate speed and speed limit. Chosen speed limit and chosen speed are not identical in terms of compliance, but are correlated. Experiment 2 provides various measurements of credible speed limits, and examines how to set more credible speed limits in order to improve driver compliance. A picture questionnaire, a driving simulator in automated conditions, and manual driving in a simulator are used for measurement. The experiment investigates how the layout of single carriageway roads and roadside environmental factors affect speed limit credibility, subjective risk perception and compliance with speed limits. Five indicators: the most common choice of speed limit by drivers; the highest credible rating score value; indication of comfort with speed in automated driving; risk rating in the range from feeling safe to very safe; arousal indicated by skin conductance, are used to evaluate a credible speed limit for a given road layout, which is used to define a credible speed limit. The method used for setting credible speed limits can be applied to other types of roads. The study develops the relationship between speed limit credibility, risk perception, and compliance with the speed limit. Experiment 3 investigates how road warning signs affect perception (credibility, safety and necessity) and driving behaviour for a given road layout and roadside environment, using a questionnaire and driving simulator. The study finds that road warning signs affect driving speed, specifically by slowing down the driving speed and reducing the proportion of time spent driving in excess of the speed limit. Combining the results of the three experiments, the research confirms that speed limit credibility is useful for speed management. The findings indicate that there exists a credible speed limit for each specific type of road that would lead to better speed management. As the credibility of the speed limit increases, drivers become more compliant with it. In terms of practical implications for road design, the research provides advice to local highway authorities on matching credible speed limits to rural single carriageway infrastructure in order to provide safe conditions for all road users. Speed limit compliance can be reinforced by using the most effective combination of warning and speed signs.

Emotions and moods are an inevitable part of human life. Previous research suggests that positive and negative moods affect human performance in many aspects: decision making, perception, reasoning and memory. The influence of mood on drivers’ behaviour has been studied to a less extent and mainly with respect to negative emotions. The studies reported in this thesis are based on psychological theories regarding the differences in the effects of positive and negative moods on information processing and mind wandering. The thesis describes two studies: a desktop study and a driving simulator study, which measure drivers’ responses to the actions of other traffic, their observational patterns and driving behaviours in a variety of scenarios. The effects of neutral, happy, sad and angry moods were studied. The simulator study also investigated possible ways to disconnect drivers’ minds from mood induced mind wandering by using different types of cognitive load. The results suggest that mood valence and arousal have different effects on driving safety, with negative moods resulting in the most dangerous driving. In order to draw conclusions about the effect of mood, a combination of multiple measures (e.g. glance patterns, driving performance and drivers’ physiological measures) should be analysed. The results also suggest that some amount of cognitive load, applied while driving, can have a positive effect on drivers’ attention. Further research is needed to establish the amount and type of the cognitive load necessary to improve drivers’ ability to maintain their attention on the driving task. Studies with a larger number of participants and field studies are needed to validate the findings. It is suggested that the findings are used to improve in-car assistance systems able to both detect the harmful effects of a driver’s emotional state and re-direct their attention to the primary task of driving.

Least developed landlocked countries' freight transport has been a step slower than other countries in the world trading market. While many research studies and international organisations report on landlocked countries' situation and issues but few explain the causes of the problem and the relationships between the factors. Moreover, there have been few comprehensive studies in which several solutions and policy implementations have been investigated to determine the best result. This research was aimed at identifying general problems by structuring a traditional causal loop diagram to present the key problems and relationships based on the previous studies' resources. The research used qualitative data collection to increase understanding of stakeholders' experiences and transport policy solutions. Thus, Laos was chosen as a typical case study, where its trade was obstructed by it being landlocked. The primary factors found through qualitative data collection were represented on a model that identifies the key dynamic factors within specific groups. The most concerning policy was chosen to formalise a dynamic hypothesis and set up a new hypothesis for an alternative solution. The research's results show the causal loop diagram has presented a better view of the structure and overall connection among themes. The fieldwork information represents complex real-world factors and causes that have effectively suggested small details that might be more important than seeing factors. Where the stock and flow model gave a better understanding of how dynamic freight transport system behaviours changed through the policy, it confirmed that cooperation among internal sectors and cooperation between landlocked countries and their transit neighbours is a primary approach to developing least developed landlocked countries' freight operations.

Travelling at illegal and/or inappropriate speed continues to be the single biggest factor in road traffic crashes and fatalities in Nigeria. Existing evidence suggests that drivers, particularly those who work in companies with a strong safety culture exhibit different sets of speeding attitudes and behaviours in work and private driving. This research is based on the premise stated above, and the lack of speed-related research in Nigeria. Using Ajzen's Theory of Planned Behaviour (TPB: Ajzen, 1991), this thesis investigates the socio-cognitive determinants of speeding behaviour of drivers' in their work and personal vehicles. It also sought to test and evaluate the efficacy of two speed limit compliance interventions on driver behaviour and safety. Using a multi-method approach, four independent but related studies were carried out. Study 1, a qualitative study hinged on the TPB, elicited the salient beliefs drivers' held towards speeding. Study 2, a quantitative study inspired by the TPB was used to investigate differences in drivers' attitudes, and self-reported behaviour in their work and private vehicles. It also measured the effects of the interventions on the TPB constructs. Study 3, an experimental study, tested the efficacy of a smartphone-based advisory Intelligent Speed Assistance (ISA) application, and TPB-based Speed Awareness Course (SAC) on drivers' speed choice. Study 4, a prospective survey, examines the acceptability of ISA using the Unified Theory of Acceptance and Use of Technology (UTAUT: Venkatesh et al., 2003). The results provided greater understanding into a range of salient beliefs influencing the speeding behaviour of Nigerian drivers which are peculiar to the socio-cultural context. The findings show the TPB model explained up to 24% of the variance in Intention to comply with speed limits. As predicted, participants reported a higher Intention to comply with speed limits in their work than private vehicle. Drivers' attitude emerged as the most significant predictor and strongest correlate with Intentions to comply with the speed limit in both work and private vehicle. The TPB was also applied to evaluate changes in drivers' speeding cognition following experience with the ISA and the speed awareness course. There was no evidence of any substantial changes to any of the TPB constructs following short-term experience with the ISA, and speed awareness course. Also, investigation of the relationship between TPB variables and observed speeding behaviour suggests that higher levels of drivers' Intentions toward speed limit compliance and Strong Perceived Behavioural Control are correlated with lower levels of objectively measured speeding behaviour. Further, the dichotomous groups of low Intenders and high Intenders had significant differences in their observed speed, with the former more likely to engage in speed limit violations. Findings from the ISA and SAC intervention with regards to speed choice and safety revealed significant reductions in speed violation, reduced mean speeds, and speed variability. The findings have important theoretical and applied implications for the development of better speed limit compliance interventions to improve driving behaviour, and general road safety.

Considerable reductions in energy demand across society are necessary in order for the UK to achieve its decarbonisation targets by 2050. Significant attention has been given to challenging the carbon intensity of transport to help achieve these targets, with commuting of particular interest. Flexible working practices are promoted as desirable policies to intervene in these journeys; however cutting emissions and reducing demand has proved difficult. Limitations of these predominant individualistic perspectives illustrate why alternative mechanisms and perspectives are required to approach the sustainability challenge. Social practice theory can help in this endeavour, by decentralising the individual and instead placing the practices which constitute individual lives at the centre of analysis. Exploring and understanding transport as a system of practice draws attention to the ways in which practices bundle together in the organisation of everyday lives. It is through this understanding that alternative avenues for intervention arise; for example into the practice of flexible working, which engender the need for particular modes of mobility. To understand the influence of flexible working, analysis of 29 in-depth semi-structured interviews was conducted, which considered an understanding of work and its connections to other adjacent practices, particularly related to the household and the commute necessary. The results show that although technology has allowed work practices to change in many ways and to allow for greater spatial and temporal flexibility, it has not, in the sample investigated, led to a major shift in commute behaviours to lower carbon modes. The results also show that it is not the apparent flexibility of different types of work ('creative' vs 'non-creative') that is important, rather that the workplace, for many professions and for many people, seems to hold very strong social, material and professional attractions which influence what flexibilities exist and how they might be deployed or integrated.