Posters presented at Transportation Research Board
Walter E. Washington Convention Center. Washington DC, USA
Co-Authors: Hanna Grzybowska, Kasun P Wijayaratna, Vinaya Dixit, S. Travis Waller
Abstract: Ramp metering is a traffic management technique aimed at controlling the flow of traffic entering motorways and freeways, thus minimising congestion on the main thoroughfare. This technique has been studied and implemented globally since the 1960s attracting debate over the deficiencies and merits of the system. To date, there has not been a consensus on an evaluation methodology concerning the implementation of ramp metering.
This paper attempts to extend previous methodologies for evaluating ramp metering algorithms. The study conducted a comprehensive review of past applications, feasibility studies and research studies of ramp metering solutions to understand the key aspects of methodologies applied to date. Two main patterns were found in the literature, firstly, evaluations of the systems are typically conducted using before-after studies or simulation based studies. Secondly, evaluations have primarily focussed on the costs and benefits related to the performance of the motorway without consideration of system wide impacts.
This research proposes a novel evaluation methodology facilitating the evaluation of the impacts of ramp metering system on the entire network. The evaluation methodology has then been applied to a case study in Sydney, Australia using microsimulation to compare two state-of-the-art ramp metering algorithms. Results and outcomes of the application are presented with the aim of understanding the key performance metrics and factors which affect the successful implementation of these ramp metering solutions in a network wide context. In addition to the performance measures, other factors which affect the successful implementation of ramp meters are explored through direct interviews with the road authorities that have implemented ramp metering within Australia and New Zealand.
Co-Authors: Taha Hossein Rashidi, Lauren Gardner, S. Travis Waller
Abstract: Obesity and other chronic diseases are becoming more prevalent in affluent countries such as Australia, and as such researchers are trying to understand and combat this trend. One related growing stream of research explores the role of the built environment and the transport system on an individual's weight. However, results from many of the studies conducted have been contradictory. One of the primary causes of these contradictions is due to the way the neighborhood area is defined, which directly affects how the built environment variables are calculated in GIS. The potential impacts on regression analysis resulting from different data aggregation methods are well documented in spatial studies, geography and regional planning fields, and it is primarily referred to as the Modifiable Aerial Unit Problem (MAUP). In this paper, the focus is on reducing the error caused by MAUP by introducing a new data aggregation method. Individual health and lifestyle data are obtained from the Household, Income and Labour Dynamics in Australia (HILDA) survey, and the relationship between the built environment and obesity is evaluated using a discrete choice model. The proposed aggregation method is evaluated across three spatial scales, and compared against a conventional data aggregation method (i.e. utilizing predefined administrative boundaries such as census tracts). The results reveal a stronger relationship between land use variables and obesity when the proposed aggregation method is implemented.
This paper is relevant primarily to researchers as it provides an improved aggregation method to deal with some of the privacy restrictions of surveys. It is also relevant to practitioners and policy makers by quantifying the association between specific built environment variables and obesity.
Co-Authors: Lauren Gardner, David Rey
Abstract: It is important to understand how epidemics spread to new regions via the global air traffic network in order to develop effective strategies for outbreak control. Various studies have focused on predicting epidemic spread via the complex air traffic network. However, there is a gap in the literature demanding real-time predictive models that exploit the heterogeneous nature of the air travel pattern to optimize decision making among a set of potential control strategies. A bi-level optimization model is proposed to solve the resource allocation problem for an on-going epidemic spreading via the air traffic network. The upper level objective is to optimize the distribution of limited resources for epidemic control while the lower level simulation model computes the risk posed to the network under possible scenarios. Results from a demonstration network highlight the advantages of this model. A case study is also conducted which evaluates the risk posed by Ebola to the United States via the domestic air traffic network. The results demonstrate the ability of the model to develop real-time strategies which account for the heterogeneous nature of the air traffic network and the complex dynamics of epidemic spread.
Co-Authors: Michael W Levin, Stephen D Boyles, S. Travis Waller.
Title: On System Optimal Dynamic Lane Reversal To Improve Traffic Efficiency For Autonomous Vehicles
Abstract: As a future of autonomous vehicles becomes more certain, transport network managers may seek ways to reinvent elements of the traffic network to improve efficiency. One possibility is dynamic lane reversal, in which the network operator makes use of AV communications and behavior to change the direction of flow on a road link at smaller time intervals than would be possible with human drivers. While much research is looking at the mechanical details of AVs, this work focuses on a planning application in which AVs are already present in order to motivate the need for future research. We examine a novel extension to an established system optimal dynamic traffic assignment model based on the cell transmission model. The model determines the optimal lane configuration at small space-time intervals. Results demonstrate the model on a single link and a grid network, exploring the dynamic demand scenarios which are most conducive to increasing system efficiency with dynamic lane reversal
Co-Authors: Neeraj Saxena, Sai Chand, Nima Amini, Hanna Grzybowska, S. Travis Waller.
Title: On System Optimal Dynamic Lane Reversal To Improve Traffic Efficiency For Autonomous Vehicles
Abstract: Dynamic traffic assignment has received an increasing amount of attention in recent years, with numerous examples of practical implementations. This work adds to the existing body of literature by describing the ongoing experience of building the first large-scale simulation-based DTA model in Australia. We provide a summary of the input data for the model and then focus on an in-depth discussion and analysis of model output and the calibration process. Current results put 80% of the 322 calibration points spread across the network within an acceptable bound of error, but the project found that it was also important to consider alternative metrics of network performance so as not to neglect other aspects of model realism. In the future, the DTA model described here could aid in evaluating important policy decisions and infrastructural development in the context of the macro/meso-scale network operation. Additionally, this project serves as a proof of concept for the Australia region and may provide valuable insight to other practitioners interested in emerging areas of transport planning and traffic modeling.
Mohammad Nurul Hassan
Co-Authors: Taha Hossein Rashidi, S. Travis Waller, Neema Nassir and Mark Hickman.
Title: Do Travellers Strategize When Selecting Their Transit Stop?
Abstract: This research aims to study the use of behavioral travel modeling approaches in transit users' stop choice behavior. Several model structures like MNL, NL, Mixed MNL and Mixed NL are used for this purpose where stops are segmented based on travel schemes. Stop choice attributes considered in the modelling exercise include facility attributes, path attributes associated with each stop, correction factors to address the correlation between the paths and users' socio-economic factors. The data used in this research was taken from the household travel survey (HTS) of 2009 in Southeast Queensland (SEQ), Australia.
Co-Authors: Vinayak Dixit, Samer Hamdar
Title: Estimating risk Attitude and Risk perception in Car following
Abstract: Theoretical developments in decision-making research enable us to consider the car-following behaviour from a new angle. However, the application of these theories in modelling car-following behaviour has been rarely seen. This research studies different car-following situations based on the vehicle type of leader and follower. Established on the Expected Utility Theory, a new car-following model incorporating risk attitude and perception, two important concepts from behavioural economics is shown. Risk attitude and perception are then explicitly defined in the context of transport research. Constant Relative Risk Aversion (CRRA) model is adopted for the estimation of risk attitude. Risk perception is assumed to be related to the stochasticity of the spacing perceived by drivers. The risk attitude and perception are successfully estimated using the Next-Generation Simulation (NGSIM) data. We find that drivers are risk averse while following a vehicle. The values of risk attitude of most drivers are around 0.5, which is consistent with a previous study by Vinayak et al. (34). Differences of risk attitude and perception among drivers are found in the current study. The proposed model is also used for the prediction of speed in different types of car following with reasonable accuracy. The future study will focus on the macroscopic properties of the model, such as flow-density relationships, average travel times and average delay. Further, the lane-changing behaviour will be considered in this stochastic framework.
Co-Authors: Vinayak Dixit.
Title: Accounting For Transport Impacts On The Economy: An Integrated Computable General Equilibrium And Transport Model
Abstract: The use of separate transport and economic models in urban planning provides a limited view of economic impacts, restricts the testing of network design options and lengthens the planning process. Furthermore, the standard methodology for economic appraisal assumes partial economic equilibrium and cannot determine the distribution of impacts from the transport sector to particular households. Computable general equilibrium (CGE) models can capture general equilibrium effects and measure welfare at the household level, but mostly lack integration with transport models and do not represent all trip generators. This paper develops an integrated traffic assignment and spatial CGE model in nonlinear complementarity form. The CGE submodel generates commuting, shopping and leisure trips as inputs into the transport submodel, which then assigns trips to the network according to user equilibrium. The resulting travel times then feed back into household prices and freight margins. Households and firms fully account for travel times in decisions on where to shop, how much labor to supply and where to source production inputs. Calibration and applications of the model are demonstrated for 14 regions and 2 industries across Sydney using GAMS/PATH on the NEOS server. The welfare of various network improvements is 16 measured using equivalent variations.
Co-Authors: Chen Cai, Lauren Gardner, Vinayak Dixit, S. Travis Waller, Fang Chen.
Title: A Strategic User Equilibrium For Independently Distributed Origin-Destination Demands
Abstract: This paper proposes an extension of the strategic user equilibrium proposed by (1) and (2). The proposed model relaxes the assumption of proportional OD demand, as it accounts for users' strategic link choice under independently distributed OD demands. The convexity of the mathematical formulation is proved when each OD demand is assumed to follow a Poisson distribution independently; link flow distributions and users' strategic link choice are also proved to be unique. Network performance measures are given analytically. A numerical analysis is conducted on the Sioux Falls network. A Monte Carlo method is depicted to simulate network performance measures, which are then compared to the results computed from the analytical expression. It is illustrated that the model is capable of accounting for demand volatility while maintaining computation simplicity.