University of California Transportation Center: Grants and Administration

2006-2007 (Year 19)

Faculty Research:

The following projects, submitted by faculty members of the University of California, were evaluated and selected for funding based on a peer review process.

An Energy and Emissions Impact Evaluation of Intelligent Speed Adaptation

Principal Investigator:
Matthew Barth
UC Riverside
Email: [email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract: Excessive vehicle speed on today’s roadways often results in accidents, high fuel consumption rates, and excessive pollutant emissions. Traditional methods of limiting speed have only been moderately effective. Using the latest intelligent transportation technology, speed enforcement can be enhanced through vehicle speed management programs, often referred to as Intelligent Speed Adaptation (ISA). An ISA system monitors the location and speed of the vehicle, compares it to a defined set speed, and takes corrective action such as advising the driver and/or governing the top speed of the vehicle. ISA is an active research field in Europe where it is currently being evaluated. ISA also has the potential to smooth traffic flow during congested conditions, also leading to lower fuel consumption and emissions. In order to better understand the impacts of ISA on energy/emissions, it is proposed herein to carry out a careful analysis using both simulation tools and real-world experimentation. This project will take advantage of microscopic transportation/emissions modeling tools and other telematic system hardware previously developed by the principal investigator. The simulation tools will be used to examine different speed management algorithms under varying traffic scenarios and congestion conditions. For each evaluation, quantitative energy/emission impacts will be determined. To complement the simulation analysis, a set of limited real-world experiments will be performed using real-time traffic information provided to an ISA-equipped vehicle driving in traffic. Results will be compared to another non-equipped-ISA vehicle acting as a control, representing the general traffic flow. It is expected that a good deal of insight will be gained through both the simulation and real-world ISA experiments.

Key Words: dynamic speed control, intelligent transportation systems, energy and vehicle emissions impact analysis

Objective: The overall research objective is to evaluate the impacts that intelligent speed adaptation techniques have on vehicle fuel consumption and emissions using both simulation modeling tools and real-world experimentation.

Tasks:

Task 1: Literature/data review.
Task 2: Simulation Setup and Analysis;
Task 3: Conduct Real-World Experimentation.
Task 4: Reporting.

Milestones, Dates: Official start date Aug. 1, 2006, end July 31, 2008

Student Involvement: Graduate Student Researcher

Technology Transfer Activities: Publications will be posted on UCTC’s Website and distributed in hard copy, in most instances free of charge.

Relationship to Other UCTC Research: new project

Potential Benefits: The results from this project will allow us to quantify the potential benefits that intelligent speed adaptation will have on energy consumption and pollutant emissions; if favorable, this program may lead to a larger more comprehensive study.

Direct Cost: $47,389

Relieving Congestion by Real-time Monitoring of Traffic Conditions and Coordination of Traffic Signals across Zone Boundaries

Principal Investigator:
Michael Cassidy
Carlos Daganzo
UC Berkeley
Email:
[email protected]
[email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract: The current paradigm for traffic control relies heavily on forecasting models. Yet the models and data used to produce the outputs are unreliable. A robust strategy has recently been developed to avoid urban gridlock (queue spillovers that cause traffic to devolve to jammed state). The idea behind this strategy consists of dividing a metropolitan area into neighborhood-sized zones; monitoring macroscopic traffic conditions such as aggregate vehicular accumulations within each zone in real time; and controlling flow between zones. We propose to adopt this strategy in the settings of centralized traffic signal control and to develop algorithms that complement those already deployed in systems such as Los Angeles Department of Transportation (LADOT)’s Adaptive Traffic Control System (ATCS). We will develop plans to demonstrate the effectiveness of the strategy through a field test.

Key Words: traffic control, signal control.

Objective: Relieve congestion by real-time monitoring of traffic conditions and coordination of traffic signals across zone boundariesRelieve congestion by real-time monitoring of traffic conditions and coordination of traffic signals across zone boundaries.

Tasks:

Task 1: Obtain and analyze information about ATCS.
Task 2: Develop new guidelines for partitioning.
Task 3: Develop aggregation technique to monitor neighborhood level traffic state.
Task 4: Develop algorithms for implementing proposed strategy in ATCS.
Task 5: Plan for a field test in ATCS.
Task 6: Develop policies for coordination and synchronization of traffic signals across jurisdictional boundaries

Potential Benefits: Congestion relief and better utilization of transportation system capacities.

Direct Cost: $58,658

Robust Traffic Assignment via Convex Optimization

Principal Investigator:
Laurent El Ghaoui
UC Berkeley
Email: [email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract:The static traffic assignment problem with deterministic demand is often formulated as a linear, or, more generally, convex optimization problem. It has long been recognized that various uncertainties may affect the input data, such as origin-destination demands, or network topology. In turn, these uncertainties may greatly deteriorate the optimality of solutions to the traffic assignment problem. Thus, it is desirable to obtain a traffic assignment that is robust with respect to uncertainties affecting the model.

Recently, new approaches to decision-making under uncertainty have been proposed, under the name of robust optimization. The methodology has been successful in many areas of engineering, such as communications, filter design, control systems, and also in machine learning and statistics.

The goal of this project is to evaluate the potential benefits of using a robust optimization approach in the context of traffic assignment, both for static and dynamic problems.

It is expected that the proposed approach will provide a traffic assignment methodology that provides solutions that are far more robust than the original ones, yet give up relatively little in terms of performance. A more long-term potential benefit is that our research contributes to the growing interplay between robust optimization and transportation research.

Key Words: traffic assignment, convex optimization, uncertainties, robust optimization, performance

Objective: Development of a robust traffic assignment methodology that combines the efficiency of convex optimization algorithms and versatility of robust optimization models, offering solutions that are robust at little extra cost in performance.

Tasks:

Task 1: Literature review and impact assessment: an extensive literature review on traffic assignment under uncertainty will be performed, as well as and an assessment of the sensitivity of current models with respect to uncertainties.
Task 2: Static assignment, bounded uncertainty: we will investigate the static case, with bounded uncertainty (set-membership) models. A focus of this task is the development of models for extreme situations which catastrophically disrupt the network.
Task 3: Static assignment, stochastic uncertainty: this task is concerned with the case when uncertainty is modelled via random variables, with imperfectly known distributions.
Task 4: Dynamic assignment: this task is to formulate efficient robust optimization algorithms for the dynamic case, with focus on so-called linear recourse strategies. A focus of this task is again the context of highly disruptive perturbations.

Potential Benefits: The product of the project will enable to obtain solutions to traffic assignment problems that offer greater reliability, and behave better even in the case of large-scale, catastrophic perturbations.

Direct Cost: $70,240

Evaluation of the Information Needs of the Distributed Landside Port Planning in California

Principal Investigator:
Mark Hansen
UC Berkeley
Email: [email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract: Trucks are a vital and rapidly growing component of the California economy. This growth provides large statewide benefits, yet also strains the transportation system serving California’s ports. Despite the importance of truck travel in California and their relationship to port operations, current landside port transportation plans do not accurately account for movements of truck traffic due to a lack of deep understanding of demand. A sound transportation plan based on the planning decisions of truck operators and public agencies will allow California’s ports to grow without degrading the region’s transportation system. This research proposes to understand the informational needs of public agencies and private truck operators in planning for landside port operations and to define a demand analysis toolkit to provide this information efficiently. A prototype methodology for the demand analysis tool deemed most important to model truck traffic serving ports will be developed. It is expected that improved Origin and Destination tables to assist in the Trip Distribution stage of transportation modeling would provide the most benefit to both the public and private sector.

Key Words: freight, truck travel, California, landside port, transportation plan, truck operator, public agency, demand analysis

Objective: This research will investigate the distributed processes by which truck operators and port agencies plan for trucking activity in port areas, and initiate development of an analytical toolkit to facilitate and improve such planning.

Tasks:

Task 1: Evaluate Current Transportation System Service from the Perspective of Trucks at California Container Ports.
Task 2: Analyze Landside Port Planning and Identify Analytical Needs Related to Accommodating Trucking Activity.
Task 3: Initiate Development a Prototype Methodology for the Analysis Toolkit Identified.

Potential Benefits: A sound transportation plan based on information produced by the toolkit contemplated in this research will facilitate the planning decisions of truck operators and public agencies and allow California’s ports to grow without undue congestion or environmental degradation.

Direct Cost: $50,933

Taxi Drivers in Los Angeles: Profile of a Workforce Facing Change

Principal Investigator:
Jacqueline Leavitt
UC Los Angeles
Email: [email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract: Little is known about taxi drivers in the City of Los Angeles. In general, taxi drivers work long hours, without health insurance, for low wages. Nine taxi operators with a total of 2,303 cars have franchises each of which is assigned to one or more service areas. The L.A. taxi workforce is increasingly immigrant. Data about drivers conflates taxi and limousine drivers. Diversity by ownership or lease is not clearly defined; drivers include owner-operators, who may own multiple cabs and lease drivers who pay rent to an owner or fleet. Highly regulated, relationships with government bureaucracies and franchises determine fees and expenses including weekly leases, car payments, insurance, maintenance, etc. This study focuses on characteristics of taxi drivers -- demographics, income, health coverage, stress levels, and impacts on households – and the structure of their relationships with taxi companies. Findings are based on a survey of 400 taxi drivers and in-depth interviews of 30 taxi drivers. The findings will help the L.A. City Council in considering a proposal by taxi operators to shift to medallions and as franchises expire in 2010. Study findings can be used to clarify consequences for change in an industry whose workforce profile is changing.

Key Words: taxi drivers, City of Los Angeles, health insurance, low wages, immigrant, owner-operators, lease drivers, survey, in-depth interview, medallions

Objective: Develop a socio-economic profile of taxi drivers in Los Angeles, including working conditions, income, years of driving, health coverage, national origin and impact on economic support of households.

Tasks:

Task 1: Administer 400 surveys to taxi drivers in the City of Los Angeles.
Task 2: Conduct 30 in-depth interviews.
Task 3: Data entry and analysis.
Task 4: Review secondary literature.
Task 5: Draw from current research studies on network and sectoral analyses of taxi industry in LA

Potential Benefits: Provide a needs assessment about taxi drivers and information at a time when the City Council will face expiring taxi operator franchises and taxi industry is circulating a proposal to adopt a medallion system.

Direct Cost: $35,029

Subcontracting Decisions in California Highway Procurement Contracts

Principal Investigator:
Justin Marion
Gil Ricard
UC Santa Cruz
Email:
[email protected]
[email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract: Theories of the firm suggest that problems such as contractual incompleteness and hold-up lead firms to produce more inputs in-house rather than purchasing them from potentially more efficient suppliers. Repeated interactions between firms and their suppliers are often thought to relieve such problems, as the risk of putting in jeopardy future business opportunities often outweighs the short-run gains from providing suboptimal levels of non-contractible output or holding up production to capture more rents. The research described in this proposal will empirically examine the role of relationships, in the form of repeat interactions, between contractors and subcontractors in the California state highway procurement market. Data from auctions awarding highway construction and repair contracts will be used to assess several questions. First, what determines relationship formation? Second, how do such relationships improve firm productivity? And third, how do these relationships lead to improved performance after the contract is awarded? These results will then be discussed in the context of California Department of Transportation policies, such as the affirmative action program for disadvantaged subcontractors.

Key Words: Highway procurement, subcontracting, relational contracting.

Objective: The goal of this research is to provide insight into relationships between contractors and subcontractors, including factors influencing their formation, and how they affect firm efficiency and performance.

Tasks:

Task 1: Additional data collection of prices charged for items within contracts, and the actual value of payments to subcontractors by the general subcontractor.
Task 2: Data management including creating tracking variable for subcontractors across projects, generating contractor-subcontractor relationship measures, and generating geocoded firm locations and distance measures.
Task 3: Statistical and economic analysis, including estimating determinants of relationship formation, how relationships impact firm costs, and how relationships affect project performance.
Task 4: Write final report.

Potential Benefits: Better understanding of the impact of current and future policies relating to subcontracting, an important component of the production decision of highway construction firms.

Direct Cost: $58,641

Models for Evaluating General Truck Transportation Management Strategies

Principal Investigator:
Amelia Regan
UC Irvine
Email: [email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract: The objective of this research is to develop an effective and efficient general truck management strategies (GTMS) evaluation model that can concurrently reflect public and private sector standpoints. Microscopic traffic simulation models are the primary analytical tool employed in this analysis. Such models enable the examination of a wide variety of systems performance measures – measures that can be used to evaluate the impacts of various GTMS on transportation systems. Increased truck traffic has resulted from continued growth of national and international trade. This increase leads to traffic congestion, safety hazards, air pollution, and rapid pavement deterioration on highways and streets. In order to reduce these negative impacts, traffic agencies have studied various GTMS. The strategies are usually classified according to their primary goals: a) improving highway characteristics b) intelligent transportation systems c) operational strategies and d) enforcement/compliance. These are of great practical importance. However, they have mainly addressed operational and safety aspects of truck operations. These aspects are important but insufficient because they do not accurately describe the full effects of the implementation of GTMS. With this motivation, the evaluation models will be developed to analyze major challenges of truck traffic as well as to simultaneously reflect the standpoints of the public and private sector.

Key Words: Intermodal Freight Transportation System, Capacity Modeling, Truck Management Strategies

Objective: To develop an effective and efficient general truck management strategies (GTMS) evaluation model that can concurrently reflect public and private sector standpoints.

Tasks:

Task 1: Comprehensive Literature Surveys for the Prevalent GTMS in the U.S.
Task 2: Developing a Model For GTMS Evaluation.
Task 3: Analysis and Evaluation of the GTMS using Micro Simulation.
Task 4: Write final report.

Potential Benefits: This research will develop useful planning tools and also provide insight into freight bottlenecks in the case study region.

Direct Cost: $63,060

Approach to Real-Time Commercial Vehicle Monitoring

Principal Investigator:
Stephen Ritchie
UC Irvine
Email: [email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract: Vehicle classification algorithms allocate vehicles to predefined classes based on selected vehicle characteristics. Such algorithms have many important applications in transportation systems analysis and policy development. These include travel forecasting, goods movement studies, road design and maintenance, setting user fees, safety studies, traffic flow modeling, environmental impact analysis, traffic management and automated toll collection. This research will collect a large and unique dataset of commercial vehicle (CV) signatures using conventional inductive loops and a new wireless sensor with potential for cost-effective and widespread use. The data will be collected at the California Highway Patrol (CHP) I-5S Truck Weigh and Inspection Station in San Onofre, California. The cooperation of the CHP, Caltrans and others involved in the operation of this facility has already been obtained for this project. The data will be used to develop much more detailed and accurate vehicle classification algorithms for CVs, and will provide important insights into the strengths and limitations of a new wireless traffic sensor for vehicle classification and vehicle reidentification purposes.

Key Words: vehicle classification, algorithms, commercial vehicle, signature, inductive loops, wireless traffic sensor, vehicle reidentification

Objective: The objective of this research is to collect a large and unique dataset of commercial vehicle signatures for development of much more accurate and detailed commercial vehicle classification algorithms, utilizing a new traffic sensor with the potential to provide a low-cost, easily deployable, wireless alternative to conventional loops.

Tasks:

Task 1: Design of data collection setup, and assemble equipment.
Task 2: Collect signature data and video data at both locations.
Task 3: Investigate signature preprocessing requirements for Sensys data.
Task 4: Ground truth vehicle signature data.
Task 5: Investigate feature correlation and error analysis for vehicles belonging to similar classes for both loop and Sensys data.
Task 6: Develop vehicle classification algorithm for Sensys data.
Task 7: Develop improved vehicle classification algorithm for loop data.
Task 8: Recommendations for potential of loop and Sensys sensors for vehicle classification applications.
Task 9: Develop vehicle reidentification model for commercial vehicle traffic between both Sensys to Sensys and loop to Sensys sensors.
Task 10: Final Summary Report.

Potential Benefits: The expected benefits include new insights into the structure and formulation of improved vehicle classification algorithms for commercial vehicles, development of more accurate and detailed vehicle classification algorithms for commercial vehicles, key insights into the strengths and limitations of a new wireless traffic sensor for vehicle classification and vehicle reidentification purposes, and the potential to influence policy and investment decisions.

Direct Cost: $57,349

Modeling Transportation Networks during Disruption and Emergency Evacuations

Principal Investigator:
Zuo-Jun Max Shen
UC Berkeley
Email: [email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract: Emergency management has attracted a lot of research attention because of the importance as well as the complexity of the problem. Well-prepared transportation systems should be able to respond to natural and human-caused disasters in a timely and effective manner, and ensure the ability to move people and goods in times of crisis. We propose to model the highly uncertain and time-dependent transportation networks during disruptions and emergency evacuations, and propose efficient optimization algorithms to solve the resulting models. Specifically, two types of optimization models will be proposed and studied. The first one focuses on scenario analysis using risk management tools, and the second model deals with dynamic real-time decision making during actual evacuations. Possible results from this project can help the planners make quick and good decisions during evacuations.

Key Words: evacuations, optimization, risk management.

Objective: To provide optimization tools for the modeling of transportation networks during disruptions and emergency evacuations.

Tasks:

Task 1: Build optimizations models for scenarios analysis and dynamic decision making.
Task 2: Write final report.

Potential Benefits: The results from this research can help the planners making quick and good decisions during evacuations to save more lives and reduce the negative impacts of the disasters.

Direct Cost: $70,526

Congestion and Accessibility: What’s the Relationship?

Principal Investigator:
Brian Taylor
UC Los Angeles
Email: [email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract: This research examines how measures of transportation accessibility and congestion vary and relate in metropolitan areas. While congestion has been a perennial concern for transportation policymakers, planners, and researchers, traditional measures of congestion say little about the range and extent of opportunities that individuals are either gaining access to or missing out on because of the regional transportation system’s functionality. Using GIS-based methods, empirical measures of accessibility will be developed that account both for mobility constraints at a given location and the potential destinations accessible within those constraints. These measures of accessibility will be compared to common measures of congestion at the local and regional scales. We hypothesize that within a region, the effects of congestion on accessibility are likely to vary considerably across a single region. Because of these differences, empirical measures of accessibility may provide researchers, engineers, planners, and policymakers with different insights into the transportation system’s performance by emphasizing potential benefits for travelers rather than the mechanistic functioning of the infrastructure as do measures of congestion. This research, in other words, seeks to shift the unit of analysis in congestion measurement from the transportation network to travelers by focusing on accessibility instead of system performance.

Key Words: congestion, accessibility, GIS.

Objective: Determine how measures of transportation accessibility and congestion vary and relate in metropolitan areas.

Tasks:

Task 1: Literature review.
Task 2: Develop congestion measures.
Task 3: Develop accessibility measures.
Task 4: Congestion / accessibility comparative analysis.
Task 5: Production of deliverables.

Potential Benefits: This research will contribute to our understanding of how congestion relates to individuals’ access to destinations (or the inverse, spatial isolation), an important benefit as measures of congestion are often relied upon for guiding transportation planning and policy.

Direct Cost: $22,730

Mode Choice and Destination Choice: Estimations and Simulations for Airport Access in the San Francisco Bay Area, 2001/2002

Principal Investigator:
Kurt Van Dender
David Brownstone
UC Irvine
Email: [email protected]

External Project Contact : All UCTC projects are co-sponsored by Caltrans, Contact CoCo Briseno, Caltrans, 1120 N St., Sacramento, CA 94305, tel. 916 324-2440

Abstract:We propose to model and estimate how air passengers flying out of a San Francisco Bay Area airport choose a particular airport (as passengers often can travel to the same final destination from several Bay Area airports) and how they choose a transport mode for accessing the airport from their initial travel origin.

The estimation uses data from the 2001/2002 Airline Passenger Survey and a range of auxiliary sources, which in combination provide a detailed picture of the choice alternatives that are available to consumers with different trip origins and trip destinations, and of the time and money costs of these alternatives. In addition, the survey contains a number of useful socio-demographic variables.

The estimated model provides inputs relevant to models used in transport policy design. In addition the model itself can be used to simulate, for example, the effects of changes in airport access costs on airport market shares and on access modes’ shares. This seems especially relevant in the market for airport access, where transit modes could be expected to play a significant role. The model shows how airport and airline decisions affect passenger choices and the ensuing demand for ground transportation, in a region served by competing airports.

Key Words: urban transport, mode choice, discrete choice, mixed logit, multiple imputations, airport choice.

Objective: To estimate a model of airport choice and mode choice for access to the airport using data on passengers departing from the San Francisco Bay area in 2001/02.

Tasks:

Task 1: Construction of a dataset for estimation of the choice model, by combining the air passenger survey of 2001/02 with three or four additional sources.
Task 2: Econometric implementation of a mixed logit discrete choice model, with multiple imputation methods to account for measurement error.
Task 3: Simulation of counterfactual scenarios (policy analysis).
Task 4: Reporting.

Potential Benefits: To show how airport and airline decisions affect passenger choices and the ensuing demand for ground transportation, in a region served by competing airports.

Direct Cost: $74,489

UCTC 2006-2007 Dissertation Grant Abstracts

Intermodal Freight Network Design for Strategic Planning

Pruttipong “Palm” Apivatanagul, UC Irvine

Advisor: Amelia Regan

Freight planning is increasingly important because of steady increases in freight demand. Many states have realized the importance of the freight industry and have conducted freight planning studies in order to understand the existing problems in the freight network. Network integration is a significant factor in overall systems efficiency. Improvement projects should be directed to eliminate bottlenecks in the network. Our research focus is to develop an intermodal freight network design model that will serve as a decision support tool for allocating fixed budgets to such improvement projects. In order to analyze the benefits of improvement projects the model developed will combine a budget allocation model and a freight traffic assignment model in order to allocate the limited resources to the best set of improvement projects.

Key words: Network design, Budget allocation, Freight traffic assignment, Intermodal transportation, Bottlenecks, Integer programming

Determining the Impacts of Active Infrastructure Development on Travel Behavior and Physical Activity

Shaunna K. Burbidge, UC Santa Barbara

Advisor: Kostas Goulias

In recent years, there have been many studies correlating physical activity and travel behavior to exogenous factors. Although several have been conducted in attempt to achieve this goal, most either refer to this topic distantly in research designed for another purpose, or are methodologically weak and do little to address behavioral causality relating to the infrastructure’s impact on physical activity. The proposed research seeks to determine the connection between the installation of active infrastructure, and active travel behavior, by using surveying residents of a traditional neighborhood both before and after the installation of a class one trail using five time-point measurements of two consecutive day activity diaries. This proposed research will be the first of its kind using a tailored longitudinal (panel) study to analyze impacts of active infrastructure development on travel behavior and physical activity levels of neighboring residents. Through complex analysis methods, including regression modeling, structural equation modeling, spatial correlation, and additional qualitative methods; not only will this project contribute to the methodological “state of the art”, but it will also provide professional planners with information to justify the creation of additional active transportation infrastructure projects which could lead to an increase in physical activity and public health.

Key words: physical activity, travel behavior, active infrastructure, longitudinal (panel) study, structural equation modeling, public health

Omni-Directional Vision-Based Techniques for Traffic Surveillance and their Applications to Roadway Traffic Parameters Estimation

Meng Cao, UC Riverside

Advisor: Matthew Barth

During the past decade, a variety of traffic surveillance techniques and algorithms have been developed, primarily for observing vehicles from stationary rectilinear cameras mounted near roadways. However, less research has been carried out in observing lane-level vehicle activity operating around specific vehicles in the traffic stream. In this dissertation, we describe a unique orthogonal omni-directional vision system (ODVS) that has been developed to observe lane-level activity surrounding a vehicle, as well as the surrounding roadway geometry. This vision system uses a special catadioptric mirror providing a 360 degree orthogonal view of the environment. It is different from other catadioptric mirror-based omni directional vision systems in that it directly provides an orthogonal image without the need of warping a polar-coordinate based image to a perspective view. Based on this unique ODVS, a 3D model based roadway traffic surveillance system is being designed and implemented. It consists of three major parts: 1) lane-level surrounding vehicle detection; 2) vehicle tracking; and 3) localized traffic parameter estimation. The relationship between these time-space varying traffic parameters and the existing traffic data at fixed positions is being studied and described by a stochastic model. This model could be used to generate the estimation of traffic condition at any time and locations and the results will help to improve safety, increase traffic flow, reduce congestion and improve air quality through different traffic applications such as abnormal incident detection, accidents prediction, and lane-level vehicle behavior planning.

Key words: traffic surveillance, a unique orthogonal omni-directional vision system (ODVS), 3D model, vehicle detection, vehicle tracking, localized traffic parameter estimation, stochastic model, abnormal incident detection, accidents prediction, lane-level vehicle behavior planning.

The Impacts of Transportation Energy Policy on Fuel Consumption and Transportation Safety

Chun Kon Kim, UC Irvine

Advisor: Kenneth Small

Reducing transportation fuel consumption would not only enhance the country’s energy dependency but also help to reduce greenhouse gas emissions, improve air quality, and reduce other driving-related external costs. Therefore, a comprehensive review of transportation energy policy options to reduce transportation fuel consumption is necessary.
This research aims to provide a single comprehensive framework to evaluate and compare different pricing and regulatory policy options for reducing transportation fuel consumption in the United States. It examines various transportation energy policy instruments such as fuel tax, mileage based VMT tax, Corporate Average Fuel Economy (CAFE) standards, Pay-as-you-drive (PAYD) and Pay-at-the-pump (PATP) insurance premium to measure policy impacts through computerized policy simulations. By fully integrating three interrelated economic demand decisions – size of vehicle stock, use of the vehicle stock, and energy efficiency – it can predict short-run, long-run, and dynamic effects of a policy change.

In measuring policy impacts, this research pays attention to not only direct impact on travel demand and fuel consumption but also indirect (external) impacts on greenhouse gas (GHG) emissions and road transportation safety, which are often not taken into account and may modify policy outcomes by travelers’ behavioral reactions through the changes in other decision making factors.

Key words: transportation energy policy, fuel tax, VMT tax, CAFE standards, Pay-as-you-drive(PAYD), Pay-at-the-pump(PATP), policy simulations, transportation safety.

Model Development For Transit Planning

Ting Lei, UC Santa Barbara

Advisor: Richard Church

Transit has been advocated as an alternative means of transportation to mitigate problems brought by automobile use such as congestion, pollution and urban sprawl. Current systems in all but the largest cities are often limited in providing accessible service. Furthermore, there is a gap between the available planning tools and what is needed to effectively improve accessibility. This research aims at developing needed methodology for improving transit service in moderately sized cities and towns.

One problem facing transit planners is the lack of any comprehensive methodology to map and visualize current transit services, with respect to level of service measures, like trip time, arrival times, and latest departure times to accomplish trips within the time constraints of travelers. My dissertation research will develop methods for better evaluating and mapping patterns of levels of service of transit travel and demand.

This dissertation will also develop routing methodologies that take into account several important features of transit systems that are not captured by most of current methodologies. These include the coverage of both the origin and the destination of a trip and modeling interactions between routes, such as transfers and the cooperation and competitions between routes. Integer Linear Programming (ILP) methods will be tested and a heuristic will be developed realistically sized planning problems.

Key words: Transit planning, accessibility, routing, operations research, integer linear programming, heuristics.

Network-wide Signal Control with Distributed Real-time Travel Data

Ji-Young Park, UC Irvine

Advisor: R. Jayakrishnan

Though most existing signal systems include some optimization schemes, their performance suffers from inaccurate route travel prediction due to the limitations of data. In the Persistent Traffic Cookies (PTC) system being researched in UC Irvine, the path-based variables including path flow and path travel time can be obtained from the historical trip tables and current movement information stored in individual vehicles. The information is automatically updated by intersection wireless hardware every time the drivers return to those locations, and can be read by the same hardware, unless the drivers choose to withhold it. With such path variables data diary, the improved traffic control schemes can be introduced for a group of intersections called a sub-network. For network-level control, the path flows are estimated from the inference of individual vehicle movement to capture a movement along several intersections. The future path flow is predicted based on the current path flow and historical data. Based on it, we present a scheme to group a series of intersections as a sub-network for signal optimization. First, the interaction between any two intersections is estimated by the path flow between them. After choosing a Critical Intersection in a network, a group of intersections having a certain interaction with it is selected and formed a sub-network. The signal optimization in a sub-network is accomplished by a Mixed Integer Linear Problem with the objective to minimize the total delay and a set of constraints. To our knowledge, this is the first real-time traffic control optimization scheme developed using travel diaries.

Key words: Persistent Traffic Cookies (PTC), path-based variables, sub-network, network-level control, signal optimization, Mixed Integer Linear Problem.

Insuring the City: Rebuilding Boston and the American City

Elihu Rubin, UC Berkeley

Advisor: Paul Groth

Built on the site of a derelict railyard, the Prudential Center in Boston (1962- 1965) exemplifies the massive scale of postwar city-building. A 52-story office tower anchors the 31-acre site, which accommodates a network of housing, retail, office space, and a hotel and convention center, all of which delineate a series of semi-public plazas. An urban highway, the extension of the Massachusetts Turnpike (the “Pike”), was built in conjunction with the Prudential Center (the “Pru”), providing direct access to this “City within a City” and feeding the Pru’s three levels of parking garages. In this project, I analyze the Pike and the Pru as a single piece of urban infrastructure – a colossal urban form that is not only connected physically, but also as a political and economic entity. In doing so, I reconceive the established narratives of urban renewal, which commonly focus on local applications of federal urban policy. I argue that the politics of postwar highway building cannot be understood without looking closely at corporate patronage and organization of urban redevelopment. In this regard, the tension between toll roads and free roads is central; for both the Pike and the Pru relate to the increasing privatization of urban space in the postwar American city.

Key words: postwar city building, urban highway, urban renewal, federal urban policy, politics of postwar highway building, corporate patronage, toll roads, free roads, postwar American city.

Federal Transportation Earmarking and Metropolitan Planning: Coercion, Cooptation, or Collaboration?

Gian-Claudia Sciara, UC Berkeley

Advisor: Karen Christensen

This dissertation examines how Congressional earmarks, or porkbarrel projects, inserted in federal transportation spending bills impact the planning processes usually used to select urban transportation projects. In 1982, only 10 such projects were included in the multi-year funding bill. By 1998 there were over 1,500, and in the 2005 bill, earmarks numbered over 6,300. Earmarking is not new, but the sharp rise in this practice is. Without earmarks, federal funds are authorized and then appropriated in Congressional bills allowing state and metropolitan transportation agencies to contract to spend the money. Typically, these organizations select projects and programs according to local objectives and with public input. Their planning and decisionmaking processes are shaped by federal law and inspired by publicly deliberated goal setting, competitive project selection, and fiscal constraint. With earmarks, Congress hand-picks projects that may or may not reflect metropolitan priorities. Earmarks thus transfer spending discretion from metropolitan planning organizations (MPOs) and their local members to Congress. To date, scholars have not asked what this means for urban regions and their MPOs. Does earmarking erode or bolster MPOs’ authority? Do earmarks trigger competitive or cooperative responses among MPO member agencies and governments? This research examines these questions. Case studies of MPOs and their organizational responses to earmarking are the heart of the work.

Key words: earmarks, porkbarrel, federal transportation spending, planning processes, public input, project selection, fiscal restraint, metropolitan planning organizations(MPOs), organizational responses.

Real-time Inter-modal Substitution as an Airport Congestion Management Strategy

Yu Zhang, UC Berkeley

Advisor: Mark Hansen

This study introduces a real-time intermodal substitution as a congestion management strategy into national transportation system, in case of capacity shortfalls at major airports, temporary shutdown of airports due to security uncertainty, etc.

The study consists of four parts. First, a multi-modal network traffic assignment model is to be developed with delay approximation from deterministic queuing theory. By introducing surface transportation modes, the air transportation network is expanded with links although slow but not restricted by airport capacity. This forms a network traffic assignment problem with node capacity constraints, a general network problem with various applications in transportation. Second, the developed model is to be applied to air-coach intermodal substitution (IMS) as an airport congestion management strategy, with emphasis on airlines cancellation management and integrated airborne and ground vehicle routing optimization. The economic gains from the model application will be determined. Third, the impact of implementing IMS on surface transportation will be identified in the case of aviation system disruptions, especially those due to security crisis. Last, potential barriers for implementing the IMS will be identified and strategies for overcoming the barriers will be discussed.

Key words: congestion management, traffic assignment, intermodal substitution, system disruptions.

Vehicle Activity Analysis using Stochastic and Signal Processing Techniques and its Application to Energy and Emissions Modeling

Weihua Zhu, UC Riverside

Advisor: Matthew Barth

The focus of this dissertation research is the development and application of signal processing techniques to microscopic vehicle activity data, followed by the development of a stochastic model that relates microscopic vehicle activity data with macroscopic traffic data. This analysis and developed relationships will be valuable of a number of transportation applications that will be useful for surface transportation policy making.

The overall goal of this research is to use the stochastic model being developed for better emission/energy consumption estimation and for novel emission/energy-oriented trip planning techniques. To achieve these goals, the research is divided into three closely interconnected components. The first component is the analysis of microscopic vehicle data, which includes these subtopics: Trajectory-based road type recognition using wavelet analysis; map matching techniques for spatial trajectory analysis; and investigation of the characteristics of microscopic vehicle data. The second major component of this research is developing the stochastic model that can model key relationships between macroscopic traffic data and microscopic vehicle data. The third component of the research is applying the results to two new applications.

Through this dissertation research, we would be able to produce more accurate emission inventories from traffic data, implement real-time emission/energy estimation, and develop emission/energy-oriented trip planning techniques as an alternative method of navigation.