Accurate depiction of existing traffic states is essential to devise effective real-time traffic management strategies using Intelligent Transportation Systems (ITS). Existing applications of Dynamic Traffic Assignment (DTA) methods are mainly based on either the prediction from macroscopic traffic flow models or measurements from the sensors and do not take advantage of traffic state estimation techniques, which produce estimate of the traffic states with less uncertainty than the prediction or measurement alone. On the other hand, research studies highlighting estimation of real-time traffic state are focused only on traffic state estimation and have not utilized the estimated traffic state for DTA applications. This research introduces a framework which integrates real-time traffic state estimate with applications of DTA to optimize network performance during uncertain traffic conditions through traveller information system. The estimate of real-time traffic states is obtained by combining the prediction of traffic density using Cell Transmission Model (CTM) and the measurements from the traffic sensors in Extended Kalman Filter (EKF) recursive algorithm. The estimated traffic state is used for predicting travel times on available routes in a traffic network and the predicted travel times are communicated to the commuters by a variable message sign (VMS). In numerical experiments, the proposed estimation and information framework is applied to optimize network performance during traffic incident on a two route network. The proposed framework significantly improved the network performance and commuters’ travel time when compared with no-information scenario during the incident. The application of the formulated methodology is extended to model day-to-day dynamics of traffic flow and route choice with time-varying traffic demand. The day-to-day network performance is improved by providing accurate and reliable traveller information. The implementation of the proposed framework through numerical experiments shows a significant improvement in daily travel times and stability in day-to-day performance of the network when compared with no-information scenario. The use of model based real-time traffic state estimation in DTA models allows modelling and estimating behaviour parameters in DTA models which improves the accuracy of the modelling process. In this research, a framework is proposed to model commuters’ level of trust in the information provided which defines the weight given to the information by commuters while they update their perception about expected travel time. A methodology is formulated to model and estimate logit parameter for perception variation among commuters for expected travel time based on measurements from traffic sensors and estimated traffic state. The application of the proposed framework to a test network shows that the model accurately estimated the value of logit parameter when started with a different initial value of the parameter.

The coordination of street works activities in urban networks has been highlighted by the Government as one of the most important aspects of street works practice, benefiting street authorities, undertakers and road users alike (Department for Transport, 2012c). The present research aims to develop an optimisation model for minimising the overall costs and disruptions incurred by all stakeholders as a result of implementing a number of street works schemes in an urban traffic network. The output of the optimisation model consists of optimum values for the underlying decision variables of the model such as start time of each street works scheme, type of traffic management strategy for each link, sequence of link closures and the level of resources allocated to undertake each scheme. The following two distinct objective functions, which are subject to minimisation by the optimisation model, have been developed: A primary objective function which captures the monetised effects of street works schemes such as cost of delays to road users, and cost of undertaking street works schemes. A secondary objective function (developed as a fuzzy inference system) to capture the non-monetised disruptive effects of street works schemes. The fuzzy variables of this inference system correspond to the level of ‘accessibility degradation’ of the network links, ‘connectivity degradation’ of the origin-destinations of the network, and ‘time sensitivity’ of the disruptive events (i.e. street works schemes). Next the street works optimisation problem was mathematically formulated as a bi-level optimisation programming problem, where the higher level problem is associated with minimising the aforementioned objective functions, and the lower level problem deals with predicting traffic flows, and thus the amount of delays incurred by the road users. Subsequently this study developed a genetic algorithm solution method to solve the resulting non-convex and NP-hard optimisation problem with integer or mixed type variables. Finally the performance of the optimisation algorithm was verified by a number of experimental tests on a small hypothetical network for three street works schemes.

The university academic timetable is the framework which defines the rhythm of the term-time student activities that occur on campus. This thesis explores how the design of the academic timetable affects student trip-making behaviour to and from campus and is motivated by concerns around the environmental, social and economic sustainability of the campus-based university. The thesis investigates the current understanding of student trip-making behaviour and shows that whilst it is informally generally accepted that students may plan their trips to campus around the demands placed on their time by the academic timetable, this appointment based approach is not generally recognised in student trip models. The thesis demonstrates that it is the timetable which is the main driver of student travel demand, that changes to the timetable can influence trip-making behaviour, and that a policy of timetable compression, combined with a greater use of online resources could be employed to reduce student trips to/from campus and student presence on it, thereby making the university more environmentally sustainable. However, students with compressed timetables appear to be less engaged with their studies, and exhibit a greater degree of variation in terms of their attainment level compared with students whose timetables force them to be on campus on an almost full-time basis. Students appear to prefer timetables that limit the time they need to spend on campus, and the thesis suggests that addressing this mismatch between what students currently appear to want, and what seems to offer them the best potential academic outcome represents a major future challenge to the long term academic sustainability of the campus based university.

Recently, researchers in the field of Network Science have begun to study how the structural properties of road traffic networks affect their performance characteristics. An understanding of how different structures of network infrastructure and travel demand combine to yield different performance characteristics would be useful because it could help identify how existing road traffic networks could be used more effectively or how structural features, which yield desirable performance characteristics, could be built into the construction of new road traffic networks. Thus far, however, these studies have been restricted to numerical experiments with synthetic networks that do not provide plausible representations of real road traffic networks. Furthermore, these studies have used a disparate range of parameter settings for supply and demand structure, making it difficult to generalise their findings, and have provided no explanations for their conclusions. To address these deficiencies, this thesis proposes an investigative framework for studying the effects of structure on the performance characteristics of road traffic networks. This framework comprises an experimental part, which describes how to design and conduct numerical experiments so as to provide useful insights into how performance varies with respect to specific aspects of network structure; and an analytical part, which focuses on developing explanations for patterns uncovered numerically. This thesis then demonstrates the application of this framework to an investigation of how two performance indicators; the average link Volume-to-Capacity ratio and the Price of Anarchy, vary with respect to four aspects of road traffic network structure. As part of this investigation, a simple model of road network generation is presented that produces spectrums of plausible, synthetic road traffic network ensembles, which vary with respect to specific aspects of structure. Focussing on the variation of the Price of Anarchy with travel demand, this thesis then establishes theory that explains several features of the variation shown numerically.

Transport planning relies extensively on forecasts of traveller behaviour over horizons of 20 years and more. Implicit in such forecasts is the assumption that travellers tastes, as represented by the behavioural model parameters, are constant over time. This assumption is referred to as the temporal transferability of the models. This thesis presents four main contributions in this area. First, a comprehensive review of the transferability literature in the context of the temporal transferability of mode-destination models. This review demonstrated that there is little evidence about the transferability of mode-destination models over typical forecasting horizons, and further that most evidence is from models of commuter mode choice. Second, further empirical evidence on the temporal transferability of mode destination models using data from Toronto and Sydney for transfer periods of up to 20 years in duration. The transferability of commuter and non-commuter travel has been compared, and models of non-commute travel were found to be less temporally transferable. Improving model specification through fixed socio-economic parameters was found to improve model transferability, and the travel time and socio-economic parameters were found to be more transferability than the cost parameters and the model constants. Third, and most novel, what is believed to be the first empirical evidence on the impact of taking account of heterogeneity in cost and in-vehicle time sensitivity on the temporal transferability of mode-destination models. This analysis demonstrated that while accounting for taste heterogeneity led to a better fit to the base data, there was no evidence that these models were more transferable than models without random heterogeneity. This may be due to the taste heterogeneity specification over-fitting the base data. Fourth, practical recommendations are presented for model developers on how to maximise the transferability of mode-destination models used for assessing policy.

Prolonged exposure to elevated levels of ozone (O3) has been proven to adversely affect human health, agricultural crops and building materials. Therefore, ground level O3 is currently regarded as one of the most harmful air pollutants, and identified as a priority pollutant in almost all national and international air quality legislations. Historically, O3 has been regarded as a rural issue because relatively high levels of O3-depleting species like NO from traffic and other combustion sources have helped to reduce its impact in urban area. However, more recently, and most likely as the result of increasingly more aggressive and effective management of NOx, urban O3 levels have begun to rise much more rapidly than ‘background O3’ in rural areas. As a result O3 pollution is becoming an increasing important urban issue. Therefore, the main aim of this PhD project is to characterise the temporal - spatial variability of ground level ozone. As in other studies, rural time-series of O3 and meteorological data and spatial parameters are considered as part of the analysis. However, here the analysis is extended using recently collected urban data, most notably co-located O3 and traffic related pollutant time-series and road traffic characteristics (from double loop counters), to provide more insight into the mechanisms driving traffic-related O3 trends. Traditionally, many researchers have used linear, mean-centred (and parametric) statistical methods to ozone. However, here, in light of the non-linear association between ozone and its potential covariates, the focus has been on non-parametric methods. In particular two methods have been extensively used, namely Quantile Regression and Generalised Additive Modelling. Such approaches provide a much more comprehensive description of O3 trends because they are better able to represent the distinctly different behaviour of O3 at concentration extremes. Statistical analysis shows that on average ozone concentrations are about 26 % higher in rural areas than in urban areas. The urban decrement varies at different quantiles of ozone concentrations and ranges from 10.5 μg/m3 (25%) to 21.56 μg/m3 (30%) at quantile 0.1 and 0.99, respectively. The results of quantile regression model show that up to 90 % of ozone variations between rural and urban sites can be explained with the help of road traffic characteristics. It is shown that the strength and nature (positive or negative) of the association between ozone and its covariates change at different levels of ozone. The model result shows that ozone levels increase towards north and decrease towards east in the UK. Ozone concentration appears to be negatively correlated with the distance from the coast within a range of 0 to 60 km, a trend associated with relatively less dry deposition of ozone molecules on water surfaces. Furthermore, there is a positive association between the altitude of monitoring site and ozone level, most likely due to local topographical effects. Ozone temporal trends show significant variability at different statistical metrics (e.g., mean, median, maximum and selected quantiles). Urban and rural trends have different rates and indicate that urban decrement has been decreasing over the period. Ozone trends during 1993 to 2011 and 2004 to 2011 showed different patterns, i.e., average ozone trends at both urban and rural sites are positive for the former and negative for the latter case. NOx trends have been stabilised during the last 8 years in urban areas and could have caused the ozone trends to change from positive to negative.

The implementation of ITS to increase the efficiency of saturated highways has become increasingly prevalent. It is a high level objective for many international governments and operators that highways should be managed in a way that is both sustainable i.e. environmental, social and economically sound and supportive of a Low-Carbon-Energy Future. Some clarity is therefore needed to understand how Intelligent Transport Systems perform within the constraints of that objective. This thesis describes the development of performance criteria that reflect the contributions of Information Communication Technology (ICT) emissions, vehicle emissions and the embedded carbon within the physical transport infrastructure that typically comprises three types of Intelligent Transport System. Active Traffic Management, Intelligent Speed Adaptation and the Automated Highway System are a collection of systems designed to transform the road network into a highly efficient and congestion free transport solution and all possess varying levels of uncertainty in terms of sustainability performance. The performance criteria form part of a new framework methodology ‘EnvFUSION’ (Environmental Fusion for ITS) outlined here. An attributional LCA and c-LCA (consequential lifecycle assessment) are both undertaken which forms part of a data fusion process using data from various sources. The models forecast improvements for the three ITS technologies in-line with social acceptability, economic profitability and major carbon reduction scenarios up to 2050 on one of the UK's most congested highways. Analytical Hierarchy Process and Dempster-Shafer theory are used to weight criteria which form part of an Intelligent Transport Sustainability Index. Overall performance is then synthesized. Results indicate that there will be a substantial increase in socio-economic and emissions benefits, provided that the policies are in place and targets are reached which would otherwise delay their realisation. To conclude, an integrated strategic performance management framework is proposed which performs socio-technical comparisons of four key performance areas between ITS schemes in order to identify energy and emission hotspots.

This thesis is concerned with addressing the cost analysis challenges in vertically separated railways. Both the challenges in infrastructure management and passenger railway operations are considered. A hedonic cost function is applied to better incorporate measures of heterogeneity of output into passenger train operation cost analysis. This allows for a richer understanding of the cost structure of the industry, through explicitly making scale economies a function of output heterogeneity, which in turn allows for tenders to be specified in a cost minimising manner. Three example tender re-mappings are considered for Britain. It is estimated that two out of three actually increase costs, reflecting that the model implies that for very heterogeneous TOCs, returns to density can not be exploited. In addition, the thesis details methodological work in developing analytical frameworks to exploit a multi layer panel dataset comprising observations on regions of many individual infrastructure managers. As well as providing extra observations to estimate cost frontiers, the data structure permits estimation of a dual-level inefficiency model which separates sub-company persistent inefficiency from subcompany varying inefficiency. This decomposition gives an indication as to whether inefficiency predominantly varies within firm or between firms. The example shows statistically significant inefficiency variation at both levels, and importantly, failure to take into account the dual-level nature of inefficiency is shown to under predict inefficiency. The thesis also develops new techniques to quantify uncertainty in inefficiency predictions from stochastic frontier models. This has application across the applied efficiency analysis discipline and not just in railways. Overall, this thesis finds that robust cost and efficiency analysis can only be conducted through explicit allowance for heterogeneity in output (both observed and unobserved), ensuring sufficient data quantity and that data relates to the organisational level to which decisions are made and that, in any analysis, the impact of uncertainty is quantified.

There are finite limits to resources, both extractable raw materials and planetary life support resources. Because of this, it is possible that there will be a severe and long lasting reduction in the fuel available for motorised transport which could manifest itself suddenly as a fuel shock. This thesis is concerned with the conceptual design, methodological development and application of a new spatially explicit transport policy indicator which estimates: Who could get to work tomorrow by walking and cycling if there was a fuel shock today? This thesis estimates the potential that walking and cycling have to increase resilience to fuel shocks in the period immediately after the fuel shock. A conceptual model of resilience to fuel shocks by individuals was devised. A novel hybrid static spatial microsimulation technique was developed. It was used to generate a population of individuals with the appropriate attributes to estimate for large populations the capacity to make journeys using only walking and cycling. This modelling process is generic and can be used to generate indicator results wherever suitable data exist. Using a simple scenario of a fuel shock which occurs today, current data could be used to estimate the indicator. A case study using the census data covering England, the Health Survey For England and other data sets was produced. Validation of the modelling process informs the analysis of the results. The results demonstrate the ability of the indicator to show variation between areas, in both a base case and when specific policy measures are applied. The base case indicator estimated that nationally in England only 44% (±4.85%) of individuals have capacity to commute to work by walking and cycling following a fuel shock. A local analysis of Leeds identified the spatial patterns of attributes which influence the indicator, allowing greater understanding of the geographical influences on capacity to travel by active modes. A policy package increasing bicycle availability, health and fitness and ensuring the ability of children to travel to school without needing adult escort was found to have a significant effect in 99% of English Output Areas. The indicator calculation methodology has produced significant improvements in the estimation of capacity to travel by active modes. Assuming everyone can cycle 8km (a common assumption in transport planning) overestimates capacity of the population to commute by active modes. The indicator identified a mean difference of 26% across all OAs. By considering constraints the indicator estimates of mean maximum distance travel distance by active modes differ by 73% compared to methods which ignore constraints. The indicator produced is policy relevant; The indicator can be judged as a good indicator when assessed against criteria for good indicators established by other workers. The modelling process is generic and can be applied to other scenarios. The results were presented at different extents and resolutions; making a useful and flexible spatially explicit indicator tool.

The communication gap between local authorities and freight operators has long been acknowledged as a research problem. Such a gap is problematic because of a lack of data and understanding about freight movements in most local authorities, which leads to the design of sub-optimal policies for freight. Hence policies and or interventions designed to address particular problems may in fact result in unforeseen new issues with resulting negative environmental impacts. Although these problems have been well identified in academic research relatively few attempts have been made to understand them and think of new ways to approach addressing them. This thesis makes a distinctive contribution through new empirical research and the development of new methodological tools to help overcome the communication gap. This thesis describes interviews with twenty-two urban freight stakeholders, namely local authority and freight operator representatives. In addition, it draws on decision analysis theory and techniques, and utilises elements of the Nominal Group Technique, Positional Analysis and focus groups to create an interactive, facilitated workshop method to foster improved freight stakeholder engagement. Three facilitated workshops were held in different locations around England to understand the effectiveness of this method. The interviews and initial stages of the workshop show a divergence of opinion between the two groups of stakeholders on what the most effective and desirable policy solutions are. However, the workshop format also provided a means to debate differences, identify solutions that had support across the board, and to put new solutions on the table for one or other group. The evaluation exercises from the workshop showed support for this type of method as a means of engagement as it provided a focus around potential implementation. Comparison with existing means of engagement such as Freight Quality Partnerships suggests this to be a more dynamic and purposeful form of engagement, at least in the short term. In addition, since these solutions are the result of joint participation between local authority and private sector freight stakeholders it would seem that they would be likely to receive fewer objections if carried forward to the consultation stage of policy implementation.