University of California Transportation Center: Grants and Administration

2005-2006 (Year 18)

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.

Congestion control for highway network systems

Principal Investigator:
Alexandre Bayen
UC Berkeley
Email: bayen@ce.berkeley.edu

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 project will build on available sensing technology used to monitor highway traffic, and on power computing for large scale simulations, in order to derive accurate congestion forecast strategies and efficient congestion control schemes for highway systems. The study will rely on a mathematical formulation of the physical laws responsible for highway congestion. We will build a mathematical analysis framework mapping the Lighthill-Whitham-Richards partial differential equation (traditionally used to model highway traffic) into a Hamilton-Jacobi equation, which is easier to analyze and control. Using the Hamilton-Jacobi equation, we will identify traffic flow patterns leading to congested situations, and devise control strategies to apply at the on-ramps of the highway in order to alleviate congestion. We will develop a simulation toolbox in order to validate our results against highway data available from the PeMS system. We will run test scenarios to assess the accuracy of our forecasts and the efficiency of the devised control schemes. The results will be available in the form of a report and software developed for simulation purposes. The report will include the scientific contributions of the project, as well as documentation of the software.

Key Words: sensing technology, large scale simulations, mathematical analysis framework, traffic flow patterns, simulation toolbox, PeMS system.

Objective: The goal of this project is to establish a mathematical and a computational framework for the forecast
and control of congestion propagation in highway network systems.

Tasks:

Task 1: Mathematically pose the problem of congestion propagation.
Task 2: Generate software to solve the problems posed.
Task 3: Use previous work in order to tune the models for relevant scenarios, using the PeMS system.
Task 4: Run the software on specific California highway scenarios, using the PeMS system.
Task 5: Document software and generate library of test cases.

Milestones, Dates: Official start date Aug. 1, 2005, 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: This project will provide new analysis and computational tools to forecast highway congestion and control the propagation of congestion fronts on the highway.

Direct Cost: $75,473

An Empirical and Theoretical Study of Freeway Weave Analysis

Principal Investigator:
Michael Cassidy
UC Berkeley
Email: cassidy@ce.berkeley.edu

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: A major source of freeway bottlenecks is driver lane-changing activity at freeway weaving sections formed when an on-ramp is followed rather closely by an off-ramp. Current methods for analyzing traffic conditions on these types of sections have shown poor ability to reproduce empirical observations and have produced inconsistent results. More than half a century after the first publications, there is still no consensus on which, if any, of the existing weaving analysis methods is reliable.

This research proposes an alternative approach to study weaving phenomena based on recent empirical and theoretical findings. Recent field experiments uncovered mechanisms of driver lane change behavior that cause the reduction in freeway system capacity. Further, we have developed a parsimonious extension to the kinematic wave theory (the multilane hybrid model) that reproduces this behavior by explicitly incorporating the effects of lane-changing maneuvers. These findings suggest a parsimonious method can be devised for predicting weaving section capacity based on the cause and effect mechanisms that govern this type of bottleneck.

This research will collect high-fidelity observations at multiple (i.e., three or more) freeway sites to better understand the operations of weaving sections. The findings should lead to the development of a new method that will improve upon current methodologies.

Key Words: freeway bottlenecks, lane-changing, freeway weaving sections, field experiments, kinematic wave theory

Objective: To test a new theory for better understanding the operation of freeway weaving sections and developed a new method for analysis.

Tasks:

Task 1: Literature review: further review the state of the art in weaving analysis. Fully document limitations and inconsistencies reported in the literature.
Task 2: Data collection: site selection, data collection and processing of at least 3 freeway sites.
Task 3: New method: a new method will be developed based on the team’s recent findings on the connection between lane changes and traffic congestion.
Task 4: Final report.

Potential Benefits: This would be the first theory/method able to replicate real-world observations on weaving section operations.

Direct Cost: $73,966

Freeway Deconstruction and Urban Renewal: Land Market and Transportation Impacts

Principal Investigator:
Robert Cervero
UC Berkeley
Email: robertc@berkeley.edu

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: Freeway “deconstruction” marks an abrupt shift in urban priorities in places like San Francisco, Portland, and Milwaukee. Priorities are shifting away from designing cities to enhance mobility toward promoting economic and environmental sustainability, livability, and social equity. This project will investigate the economic trade-offs and distributional implications of freeway demolitions, drawing from case-study experiences and quantitative analyses. Case-study work will probe the institutional and planning contexts of freeway removals based on experiences with the Park East Freeway in Milwaukee and the Central Freeway in San Francisco. The quantitative analyses will focus on San Francisco’s two notable yet different freeway demolitions – the Embarcadero Freeway and Central Freeway. For the Embarcadero corridor, matched-pair methods will be used to investigate pre- and post- changes in commercial real-estate market conditions, including vacancy rates and average rents relative to downtown comparison areas (beyond the freeway impact zone). For the Central Freeway corridor, matched-pair methods will be used to study changes in residential sales values between nearby neighborhoods and otherwise similar ones off the freeway corridor. Matched-pair comparisons will also be used to study changes in roadway levels of service, delays, and pedestrian accident rates for major streets and intersections impacted by the Central Freeway demolition.

Key Words: Freeway; Economic Development; Land Market Impacts; Safety; Level-of-Service; Hedonic Price Modeling; Matched-Pair Comparisons; Case Studies.

Objective: To evaluate the impacts of freeway demolitions on the land markets and economic standing on affected neighborhoods as well as traffic and safety conditions to gain an understanding of efficiency and distribution-equity trade-offs.

Tasks: conduct a literature review; conduct case study in San Francisco and Milwaukee, using secondary sources and conducting interviews with stakeholders; obtain data on land-use changes in affected corridors; acquire residential and commercial real-estate market data; conduct matched-pair comparisons of change sin property values and other real-estate market performance indicators between impacted neighborhoods and control sites; estimate hedonic price model that estimates sales prices as a function of proximity to former freeway corridor and other controls; compile traffic and accident data before and after freeway removals; measure impacts on highway levels of service, accident levels, and distribution shifts (by mode and route); synthesize the findings and address appropriate public-policy recommendations and responses; prepare article summarizing key research findings.

Milestones, Dates: Official start date Aug. 1, 2005, 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: Gauge the impacts of freeway removals on economic development and land markets of affected communities as well as traffic performance and safety levels, probing net efficiency versus distribution-equity implications.

Direct Cost: $52,592

How Much Do Low Income And Foreign-Born Households Use Public Transportation?

Principal Investigator:
William Clark
UC Los Angeles
Email: wclark@geog.ucla.edu

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: At one level the answer to this question is an unqualified “a great deal” - of course low-income populations use public transportation. How else would many of these workers get to their jobs? And, there is an implicit, and often explicit, sense that cities need to provide increased public transit and access to that transportation for low-income populations. However, recent research and commentary has raised questions about the use of public transportation by low-income populations. Studies of mostly welfare population have suggested that while public transportation is not unimportant, the car is a much more critical factor in moving from welfare to work. This paper conducts research with a detailed geo-coded data set to examine the low-income population in general and the foreign born population in particular. How much do low-income households in the Los Angeles metropolitan area use public transportation in their journey to work, schools and other activities? Are the foreign-born population particularly intensive users of public transportation and can we determine the relative trade-off of public transit and car use by these households? The study will extend previous studies that were targeted to specific welfare populations, to the low-income population at large.

Key Words: journey to work, public transportation, low-income populations, foreign-born populations

Objective: To evaluate the relative role of transit and auto use by low-income populations.

Tasks: (a) Using logistic regression to estimate the probability of being employed or becoming employed as a function of car versus transit use for individual households, (b) evaluating the use of transit by residential location.

Milestones, Dates: Official start date Aug. 1, 2005, 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 study will generalize and confirm, or question, earlier studies which have suggested that transit use does not significantly benefit welfare to work transitions.

Direct Cost: $52,119

Is the Gender Gap History? Revisiting Sex Differences in Driving using a National Panel, 1985 to 2005

Principal Investigator:
Randall Crane
UC Los Angeles
Email: crane@ucla.edu

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: Dozens of studies indicate that while women as a group drive more each year, and tend to take more trips than men, men still drive more overall. Explanations include women’s rising labor force participation and their enduring disproportionate share of domestic and child-oriented responsibilities. As these two trends continue to evolve, much remains unclear about how they and other changes in key demographic, family, and economic factors translate into changing driving patterns.

In 1990 San Francisco, working men reported driving more than working women in all age groups but 23-39. In 2000, by contrast, travel times were the same by sex for all age groups but one. Additionally, women traveled less in single, childless adult families in 1990, but more by 2000.

These trends have not been analyzed for individual-level, longitudinal national data containing rich detail on family structure. It is thus unknown whether the so-called gender gap is shrinking nationwide when using proper statistical controls. We will analyze the National sample from the American Housing Survey, comprising 11 waves from 1985 to 2005. The model specification conforms to urban form theory, the model estimation uses panel techniques, and the potential endogeneity of wages and land costs will be addressed statistically.

Key Words: Commute, Journey to Work, Sprawl, Decentralization, Suburbanization

Objective: An improved understanding of how sex influences travel outcomes.

Tasks:

Task 1: Development of theoretical model of individual travel demand
Task 2: Development of statistical model of household travel behavior and urban form, based on choice theories of home and job behaviors.

Potential Benefits: Inform public policies aimed at influencing the determinants & impacts of urban development.

Direct Cost: $44,565

A Spatial Analysis of Self-serving and Altruistic Travel Behavior

Principal Investigator:
Konstadinos Goulias
UC Santa Barbara
Email: goulias@geog.ucsb.edu

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: Formulation and specification of activity and travel analysis models require better understanding of time allocation behavior. This is particularly important when studying time allocation of persons in joint activities and travel. However, very little is known about the perceived selfish and altruistic behavior and how this relates to travel behavior in time and space. In this project we attempt to offer a first analysis using structural equations models and data from a special type of activity participation diaries. The data from these diaries are detailed lists of activities pursued, times spent in each activity, and travel information from activity to activity (including travel time, mode used, and so forth) linked to individual and household characteristics. We also use detailed information about the persons that were served by the activity of each participant. The analysis will identify the correlation structure underlying behavioral indicators such as amount of time and frequency of activity and travel episodes for family and friends as well as for self-serving reasons. Then, the temporal and spatial relationships with activity participation and travel with others and alone as well as modal split will also be studied and correlated with altruistic and self-serving behaviors.

Key Words: activity and travel behavior, egoistic and altruistic behavior, modal split

Objective: Create a model system that explains modal split as a function of altruistic and self-serving behavior.

Tasks: Task 1: Develop First Model Set (October 1, 2005 to January 30, 2006); Task 2: Develop Second Model Set (January 31, 2006 to May 15, 2006); Task 3: Write report and papers for publication (August 1, 2005 to July 31, 2006)

Milestones, Dates: Official start date Aug. 1, 2005, 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: Achieve better understanding of travel behavior to design better policies favoring environmentally friendly modes.

Direct Cost: $64,447

Emission and Air Quality Impacts of New Diesel Engine Control Technologies

Principal Investigator:
Robert Harley
UC Berkeley
Email: harley@ce.berkeley.edu

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: Diesel engines are now responsible for half or more of the total nitrogen oxide (NOx) and exhaust particulate matter (PM) emissions from on-road vehicles nationally There has been recent progress in developing control technologies such as selective catalytic reduction, lean NOx traps, NOx storage catalysts, and diesel particulate filters. Although these technologies have not yet been widely deployed in on-road vehicles, their use will soon be required due to new heavy-duty engine emissions standards that will take effect in the U.S. starting with the 2007 model year. In this research, we will review control technologies available for future use, assess control efficiency, cost, durability and robustness in service, and the potential for undesirable side-effects on exhaust emissions. Emissions estimates will be developed for historical, current and future year scenarios that span a 20-year time period. We will use a 3-D photochemical model to predict the effects of changes in vehicle emissions on air pollution levels in the Los Angeles area.

Key Words: air pollution, diesel, emission controls, nitrogen oxides, particulate matter

Objective: Assess changes in emissions and air quality likely to occur in the next 10 years due to new emission control requirements for diesel engines.

Tasks:

Review of new diesel exhaust control techniques
Estimate emissions for historical, current, and future year scenarios
Assess air quality impacts of changing emissions in the Los Angeles area

Potential Benefits: Quantify expected improvements in air quality over the next decade as a result of major investments in diesel emission control technologies.

Direct Cost: $44,834

Life-Cycle Environmental Assessment of Passenger Air and Rail Transportation

Principal Investigator:
Arpad Horvath
UC Berkeley
Email: horvath@ce.berkeley.edu

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 passenger transportation modes of heavy rail and air are critical systems relied upon for business and leisure. When considering their environmental effects, most studies focus on the fuel use of the vehicles, and ignore the energy and other resource inputs and environmental outputs from the life cycles of involved infrastructures, fuels, and vehicles.

The goal of this project is to develop comprehensive life-cycle assessment (LCA) models to quantify the energy inputs and emissions from heavy rail and air transportation in the U.S. associated with the entire life cycle (design, raw materials extraction, manufacturing, construction, operation, maintenance, end-of-life) of the infrastructures, fuels, and vehicles involved in these systems.

|The research results will assist decision-makers in private and governmental organizations in making environmentally friendlier investment and other decisions. A computer-based decision-support tool will be created to aid the decision-making process. An analysis of the proposed high speed rail line connecting San Francisco and Los Angeles will provide an opportunity to apply the research results to a case study.

Key Words: air transportation, rail transportation, energy, emissions, environment, life-cycle assessment

Objective: Develop comprehensive life-cycle assessment (LCA) models to quantify the energy inputs and emissions from heavy rail and air transportation in the U.S. associated with the entire life cycle (design, raw materials extraction, manufacturing, construction, operation, maintenance, end-of-life) of the infrastructures, fuels, and vehicles involved in these systems.

Tasks:

Literature research -- It is anticipated that the following aspects will be included in the literature research and will shape the study’s theoretical framework: composition of and data on the infrastructure serving air and heavy rail transportation, energy inputs, emissions data from each life-cycle phase, fuel use by these transportation modes, details of the fuel cycle, end-of-life treatment of infrastructure and vehicles.
Develop LCA models -- This task will include the identification of processes to be analyzed by LCA for the infrastructures, fuels, and vehicles in heavy rail and air transportation. A hybrid LCA model for each of these modes will be newly developed.
Collect data -- Collect data from literature, air and rail transportation organizations, the Ecoinvent database, and EIO-LCA.
Develop computer-based decision-support tool -- Based on the theoretical models and the data collected, an MS Excel-based computer tool will be developed in order to aid environmental analysis and enable the comparison of air and rail transportation modes with different sets of data for various decision-makers. We will design this tool to be expandable to analysis of other transportation modes in the future.
Case study -- Apply the developed LCA model to a case study, most likely to the proposed Los Angeles-San Francisco high-speed rail line which can be compared to the existing air transportation mode in the same markets.

Potential Benefits:

Direct Cost: $47,444

The Effect of In-Transit WiFi Internet Access on The Value of Travel Time Implications for Mode-Choice Models

Principal Investigator:
Adib Kanafani
UC Berkeley
Email: kanafani@berkeley.edu

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 study explores the effect on valuation of passenger travel time (VOTT), of providing of wireless wireless internet access (WiFi) on public transportation modes. Specifically, we aim to formulate a model of the effect of WiFi on trains on time value and eventually on market share.

VOTT is the amount that a passenger would pay to reduce travel time by unit quantity. Amenities that enhance possibility of other activities during travel reduces the disutility of travel time, subsequently reducing VOTT. For most work and leisure activities, of internet access during the commute enhances the quality and productivity of work and leisure activities performed during travel.

Utility maximizing mode choice models may be used for quantify the effect of WiFi on VOTT and on market share. In these models, the ratio of the Lagrangean multipliers of the travel time and income constraints results in the same value of VOTT as the ratio between marginal utilities of time & cost estimated as part of modal utility in a discrete travel choice model. Internet access can then be modeled as an attribute in attribute specific choice models. The extent of reduction of VOTT can be used as a guideline for pricing WiFi on trains.

Key Words: Information Technology, Value of Time, Mode Choice, WiFi on Trains

Objective: Evaluate the effect of internet access in public transportation modes on the value of travel time and on mode choice.

Tasks:

Theoretical Model Development
Data Assembly
Estimation of Time Values
Adapting Mode Choice Models for Internet Access Attributes

Potential Benefits: Understanding feasible business models for the provision of WiFi services on public transportation systems, and estimating the effect of that on market share.

Direct Cost: $52,969

How Safe Is the Ride? Evaluation of Design and Policy Responses to Women’s Fear of Victimization in Transportation Settings

Principal Investigator:
Anastasia Loukaitou-Sideris
UC Los Angeles
Email: sideris@ucla.edu

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 relationship between women’s fear and the built environment has been the subject of much research with clear findings that women feel unsafe in many locations. Cities and municipalities around the world have addressed this issue by implementing different programs to assess and remedy safety gaps in the built environment. Some of these programs have looked at transportation settings, but little academic research has specifically focused on this aspect of women’s safety. Whether traveling by bus, automobile, or other mode, women’s fear of transportation facilities – such as parking lots, buses, and bus stops – in turn affects the way women engage in travel. This study will focus on the safety concerns and needs of women riders by first reviewing the literature on women’s fear in public settings. Subsequent research activities will include: 1) interviews with representatives from feminist organizations and women’s advocacy groups to ascertain their perceptions of women’s safety issues in cities, 2) a survey of U.S. transit operators to document the programs and activities they have implemented to make their systems safer for women riders as well as their assessments of the efficacy of these programs, and 3) case studies of model programs targeting women’s safety issues in transit environments.

Key Words: Women, safety, transit travel

Objective: The objective of the research is to identify the needs of women groups regarding transit safety; assess if these needs are met by transit operators; and identify model programs and best practices from the U.S. and overseas that address women’s concerns about safe transit travel.

Tasks:

Literature review: Compilation of literature on women’s fears and concerns about safety in public environments with a particular emphasis on transit settings. Documentation of programs and creation of an inventory of municipal policies in the U.S., Canada, Great Britain, and Australia that have as their explicit focus the safety of women in cities.
Interviews with representatives of women’s groups in the U.S., such as the National Organization for Women, American Association of University Women, Center for Women Policy Studies, Coalition of Labor Union Women, Feminist Majority Foundation, League of Women Voters, National Council of Women’s Organizations, etc.
Web-based survey of transportation agencies: The survey will target all 259 U.S. transit agencies that according to the National Transit Database maintained by the Federal Transit Administration operate at least 50 vehicles in peak period service.
Case studies of model programs: Selection of 3-5 case studies of international programs targeting women’s safety issues in transit environments for in-depth study.
Report writing. Discussion of research findings and compilation of best practices for making transit travel safer for women.

Potential Benefits: Development of policies and design recommendations for making transit travel safer for women

Direct Cost: $43,095

Bottom-up Bridge Management System

Principal Investigator:
Samer Madanat
UC Berkeley
Email: madanat@ce.berkeley.edu

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: Infrastructure Management Systems support agencies in developing efficient policies to monitor, maintain, and repair deteriorating facilities in transportation infrastructure networks. In the case of bridges, two approaches exist for the optimization of resource allocation: top-down and bottom-up. In the top-down approach, the optimization is done first at the network level, which does not provide bridge-specific recommendations. In the bottom-up approach, the optimization is done first at the facility level. Current systems using a bottom-up approach present a major limitation: the optimization is done on a portion of the life cycle, which is equivalent to a short planning horizon. Finally, current state-of-the-art systems use time-independent deterioration models, which may not be appropriate.

A Bridge Management System is proposed, which would optimize the lifecycle maintenance while using time-dependent deterioration models and preserving bridge-specific details. A bottom-up approach can be used. Moreover, robust optimization techniques can be applied at the facility-level, in order to mitigate the epistemic uncertainty in the deterioration process. Recommendations for each component can be aggregated and selected to meet network-level constraints. It is expected that the proposed approach will provide the following benefits: long-term maintenance costs savings, improvement of long-term performance and mainstreamed maintenance management.

Key Words: Infrastructure Management Systems, Maintenance, Resource Allocation, Deterioration Models, Robust Optimization.

Objective: Development of a Bridge Management System that accounts for facility-specific attributes and system-level considerations

Tasks:

Literature Review: An extensive literature review will be performed in two main directions, bridge component deterioration models and M&R optimization models
Model formulation and solution: This task is divided in three major subtasks: subtask 2a (general problem formulation), subtask 2b (deterioration model synthesis) and subtask 2c (optimization model development).
Extensions: Two extensions to the model are proposed: simultaneous optimization of inspection and maintenance decisions (to address inherent uncertainty) and robust optimization (to address epistemic uncertainty).
Application Case-studies: The methodology developed in the previous tasks of the research will then be implemented using data from existing bridges

Potential Benefits: The product of the project will provide state DOTs a prototype for the next generation of Bridge Management Systems, which goes beyond the constraints of current systems.

Direct Cost: $44,360

A Model of Activity/Travel Scheduling/Rescheduling Decisions in an Uncertain Environment

Principal Investigator:
Wifred Recker
UC Irvine
Email: wwrecker@uci.edu

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 so-called activity-based approach to analysis of human interaction with the social and physical environments dates back to the original time-space geography works of Hägerstrand and his colleagues at the Lund School in 1970. Despite their obvious theoretical attractiveness, activity-based approaches to understanding and predicting travel behavior have suffered from the absence of an analytical framework that unifies the complex interactions among the resource allocation decisions made by households in conducting their daily affairs outside the home, while preserving the utility-maximizing principles presumed to guide such decisions. The objective of this research is to develop a computationally-tractable system, based on an extension and modification of some rather well-known network-based formulations in operations research, to model human dynamics in uncertain environments. The research will build on the mathematical programming formulation of the Household Activity Pattern Problem (HAPP) by embedding in the household activity schedule decision process a means of capturing uncertainty and preferences, and by introducing the dynamics of rescheduling. The proposed model will be estimated and validated using data derived from an on-line survey/diary of household members, documenting the scheduling and dynamic rescheduling of all (in- and out of- home) activities over the course of a one- week period.

Key Words: activity-based, human dynamics, uncertain environments, household schedule

Objective: The objective of this research is to develop a computationally-tractable modeling system that incorporates uncertainty factors within the dynamics of activity/travel rescheduling/adaptation.

Tasks: 1. Formulate “fuzzified” version of household activity/travel decision model; 2. Develop framework for producing skeleton activity plans; 3. Incorporate dynamic rescheduling based on similarity measures; 4) Estimate and verify model using existing REACT! Data.

Milestones, Dates: Official start date Aug. 1, 2005, 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 resulting framework would greatly increase our ability to address the household activity scheduling problem.

Direct Cost: $46,435

The Political Calculus of Congestion Pricing

Principal Investigator:
Donald Shoup
UC Los Angeles
Email: shoup@ucla.edu

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: Although traffic congestion has worsened and tolling technology has improved, motorists and elected officials continue to oppose congestion tolls on crowded roads. While transportation economists have long recommended tolls as the most promising way to reduce traffic congestion, economic efficiency alone has been insufficient to galvanize public policy. The political viability and public acceptance of congestion pricing depends on creating legitimate claimants who will fight for the revenue, and logical claimants for freeway toll revenue are the cities through which the freeways pass. Distributing toll revenue to freeway cities will compensate these cities for bearing the local external costs of a regional system, and will also create well-organized claimants who support for the tolls. This research will explore how distributing toll revenue to cities can work. There are three aspects to this proposal: First, we will estimate the distribution of toll revenues to cities in Los Angeles County, and interview city officials to explore their attitudes towards congestion pricing if the revenues are returned to cities. Second, we will replicate our estimates of revenue distribution in other metro areas. Third, we will conduct a review of the health implications of freeways on adjacent communities.

Key Words: congestion, congestion pricing, distribution of toll revenue

Objective: To explore how distributing toll revenue to cities can work.

Tasks: Task 1: estimate the distribution of toll revenues to cities in Los Angeles County, and interview city officials to explore their attitudes towards congestion pricing if the revenues are returned to cities. Task 2: replicate estimates of revenue distribution in other metro areas. Task 3: conduct a review of the health implications of freeways on adjacent communities.

Milestones, Dates: Official start date Aug. 1, 2005, 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 political viability and public acceptance of congestion pricing depends on creating legitimate claimants who will fight for the revenue, and logical claimants for freeway toll revenue are the cities through which the freeways pass. Distributing toll revenue to freeway cities will compensate these cities for bearing the local external costs of a regional system, and will also create well-organized claimants who support for the tolls.

Direct Cost: $47,452

Estimating Potential Exposure to Diesel Truck Freeway Emissions in Southern California: A Methodology for Assessing Air Pollution Impacts of Goods Movement

Principal Investigator:
Arthur Winer
UC Los Angeles
Email: amwiner@ucla.edu

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: It is now well documented that air pollution and health impacts of diesel vehicle emissions are highly concentrated near major roadways. This study will develop a methodology to disaggregate the potential impacts of diesel truck traffic from all other traffic in order to characterize the geographic extent and magnitude of diesel pollutant impacts along freeways in Los Angeles County, with an emphasis on those freeways with a relatively large fraction of diesel truck traffic. Recognizing the potential benefits of expanding the goods movement capacity of the freeway system, this project will provide transportation planners and policy makers with a more comprehensive understanding of local pollutant impacts, and enhance their ability to develop informed and defensible strategies to avoid and mitigate adverse impacts of heavy-duty freight vehicles.

Key Words: air pollution, diesel truck traffic, emissions, goods movement, heavy-duty freight, local pollutant impacts

Objective: To develop a methodology to characterize the geographic extent and magnitude of diesel pollutant impacts along freeways that will inform strategies to avoid and mitigate adverse impacts of heavy-duty freight vehicles.

Tasks: Review Existing Studies; Collect, Assess and Compile Truck Count Data; Assign Truck
Counts to a GIS Freeway Map System; Identify and Characterize Potential Air Pollution Impacts

Milestones, Dates: Official start date Aug. 1, 2005, 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: This project will aid transportation and air quality agencies in assessing and evaluating the potential local air quality impacts of heavy duty diesel truck emissions immediately adjacent to goods movement freeway corridors.

Direct Cost: $37,094

Dissertation Research:

Environmental Knowledge, Environmental Attitudes, and the Environmental Impact of Automobile Ownership and Use

Bradley J. Flamm, UC Berkeley

Advisor: Martin Wachs

    A wide gap has emerged in recent decades between Americans’ commitment to protecting the natural environment and the environmental impact of their automobile ownership and use. In the aggregate, it appears there is an inverse relationship in which the stronger Americans' pro-environmental attitudes grow, the more resource-intensive their auto ownership and use become.
    Why is it that stronger environmental attitudes have not resulted in smaller environmental impacts from automobile ownership and use? What explains the discrepancy between environmental knowledge and attitudes and the environmental impacts of automobile ownership and use?
    In this study, the relationships between environmental knowledge, environmental attitudes, and the environmental impacts of automobile ownership and use will be explored. The principal research method will be the statistical analysis of responses to a short knowledge-attitudes-behavior (KAB) questionnaire administered in the Sacramento metropolitan region of California.
    Previous research suggests that environmental knowledge and attitudes will have positive, but weak associations with auto ownership and use. Extensive social, economic, political, and cultural constraints are believed to discourage individuals from behaving in ways that reflect their environmental attitudes. Yet were stronger pro-environmental attitudes reflected in automobile ownership and use decisions, many vexing urban transportation planning challenges could be more effectively addressed.

Key words: environmental knowledge, environmental attitudes, automobile ownership and use, knowledge-attitudes-behavior (KAB), transportation planning

Cartographic Displays of Congestion in ATIS

Kirk Goldsberry, UC Santa Barbara

Advisor: Keith Clarke

According to the 2004 Urban Mobility Report (Shrank and Lomax, 2004), "Congestion occurs during longer portions of the day and delays more travelers and goods than ever before." The report presents undeniable evidence that the rends across the United States are toward more delays. One problem is that despite ubiquitous and growing congestion events, urban travelers, including commuters, truckers, and visitors to the urban environment do not have sufficient resources for congestion avoidance. My dissertation addresses a gap in Advanced Traveler Information Systems (ATIS) research. Previous research on information provision has included work by planners, engineers, and behavioral geographers. My dissertation approaches information provision in ATIS from a cartographic perspective. Graphic displays within ATIS must be clear, legible, portable, and versatile. Current dynamic traffic displays fail in each of these areas. The general objective of this research is to create the knowledge necessary to correct these failures. The dissertation develops methods in two categories: telecartographic dissemination, and cartographic design. The dissemination methods increase the portability and versatility of dynamic traffic maps. The cartographic design methods enhance the communicative efficiency ATIS displays. This dissertation applies transformation techniques from analytical cartography and findings from behavioral scientists to create maps that are more congruent with the ways people perceive congestion and network environments.

Key words: ATIS, Traffic maps, SVG, Telecartography, Mobile mapping, Topological Diagrams.

Experimental Congestion Topics with Various Heterogeneous Profiles on Subjects

John Hartman, UC Santa Barbara

Advisor: Ted Bergstrom

A standard congestion experimental design allows subjects to choose between an uncongested route where they pay no toll but lose a set number of points for every time unit they spend traveling or a congested toll route that charges a fixed price per trip in addition to the time point cost. A combination of point deductions and a toll can ensure an equilibrium outcome that minimizes the total travel time of all participants. However, previous experiment results (see Selten et all 2004 and Hartman 2005, for example) find that while equilibrium is attained, it is unstable in the sense that even after many experimental rounds, the outcomes bounce around the equilibrium rather than settling on it. One possible explanation for this instability is the assumption of homogeneous travel costs. It is on this margin that I propose to extend the model with tolls by assigning multiple time travel point deductions instead of one - this reflects the idea that different drivers have different opportunity costs due to lost wages. In this environment, heterogeneous travel costs should help us to attain the equilibrium faster, and be more stable once achieved, because deviations are more costly for high wage earners in terms of high time costs and high for low wage earners on the toll margin.

Key words: congestion, toll, experiment, instability, heterogeneous travel costs.

Vehicle Re-identification System for Travel Time Information and Route Guidance

Shin-Ting (Cindy) Jeng, UC Irvine

Advisor: Stephen G. Ritchie

The ILD-based Vehicle ReIDentification system (ILD-VReID) tracks vehicles via vehicle signature re-identification algorithms. Although ILDs are not without problems as a traffic sensor, they are widely used for historical reasons and the sunken investment in the large installed base makes their use in this research highly cost-effective. This research attempts to present the development and the improvement of path-based ILD-VReID algorithms for both homogeneous and heterogeneous detection systems. A new approach for ILD-VReID using an interpolation method is proposed. Moreover, vehicle feature clustering and inductance transform methods will be developed for the studies of heterogeneous detection systems. It was found in the preliminary study that the interpolation method was comparable with the lexicographic method and was even superior when applied to the single loops case. The potential contribution for the proposed method is application to square and round single loop (which in California are far more common than double loop installations) vehicle re-identification and performance measurements, while avoiding issues associated with re-estimation or transferability of the speed models. The potential applications of this research are OD estimation, travel demand forecasting, and path travel time estimation. The estimated travel times and the path information are useful for the route guidance system.

Key words: homogenous, heterogeneous detection systems, interpolation method, vehicle feature clustering, inductance transform, performance measurements, OD estimation, path travel time estimation, route guidance system

A Household Activity Scheduling Model Incorporating a Task Allocation Process with Week-based Learning Mechanisms

Hee-Kyung Kim, UC Irvine

Advisor: Michael G. McNally

Intra-household interactions and group decision-making mechanisms have been receiving increasing research attention in the modeling of the activity participation. This research aims to investigate the interactions of household members and collective decision-making mechanisms in the modeling of maintenance activity participation decisions, based on task allocation and time use mechanisms within the household level. To account for these task allocation and time use mechanisms, this research proposes a week-based cycle model rather than a day-based model and a learning behavior model to see how individuals and households adapt to highly dynamic and stochastic situations with imperfect information. A preliminary study proposed a household task allocation and time use model based on some assumptions. The rationale behind the model is that a household tends to pursue the equality of time expenditures among household members to serve obligatory activities for household sustenance. This research plans to extend the preliminary study to make the model more general. First, the proposed research will address the activity chain choice problem with multiple paths, which can be applicable to the general network in the task allocation and time use model. Second, the proposed research with further extend the dimension to the spatial-temporal choice problem to completely schedule the household activity pattern.

Key words: intra-household interactions, group decision-making, maintenance activity participation decisions, task allocation and time use week-based cycle model, learning behavior model, equality of time expenditures, activity chain choice problem, spatial-temporal choice problem.

Study of Disaggregate Control of Vehicular Traffic under In-vehicle Speed Advisories using Kinetic Theory of Vehicular Flow

Riju Lavanya, UC Irvine

Advisor: R. Jayakrishnan

Traffic control applications can be divided into the following categories: ramp metering, in-vehicle route guidance systems, and speed control systems. Compared to its counterparts, relatively little attention has been paid to speed control systems. Encouraged by the growth of traffic monitoring systems and computing technology, we believe that we now have the necessary tools to tackle this rather intractable problem. Instead of providing static information about speed limits, one could consider providing dynamic speed limits to the vehicles, via traditional Variable Message Signs (VMCs) or roadside beacons that can relay dynamic speed information to in-vehicle communication devices. None of the existing speed control schemes have considered the possibility of speed advisory messages being provided individually to drivers and this is the theme of this dissertation. This is a preliminary foray into the less examined problem of traffic control using in-vehicle speed advisories under both macro and micro level of traffic detail. The theoretical developments in control could be based on the kinetic theory of traffic flow, and the research is expected to make significant use of microscopic simulation to study the efficacy of such schemes.

Key words: traffic control, speed control, kinetic theory, microscopic simulation

Work and Non-work Accessibilities in Polycentric Urban Forms: A Test of Their Impacts on Housing Prices and Land Use Changes

Guangyu Li, UC Berkeley

Advisor: John Landis

Accessibility is the key to transportation and land use connection. According to traditional location theory, transportation investments that improve local accessibility will enhance property value and promote intensive land use. Recent studies, however, show weak or even mixed evidence on such impacts. Several reasons can explain for this, such as the possible diminishing return to new investments resulted from the overall improvement in accessibility level; variations in specific transportation projects, variations in study context, and methodology limitations or data constraints faced by individual studies. This dissertation will improve our understanding of the significance of accessibility in modern development processes by examining both work and non-work accessibility in current polycentric urban form. It will contribute to the field on four aspects. First, instead of measuring the distance or time to nearby transportation facilities, it will evaluate accessibility based on the spatial distribution of desired opportunities; Second, both regional and local accessibility will be constructed; Third, in addition to job accessibility, it will examine the importance of non-work accessibility that was often ignored in past studies. Forth, with three carefully selected case study regions, it will explore how accessibility affects households’ housing consumption in different urban contexts, thus highlighting the influence of local factors.

Key words: work accessibility, non-work accessibility, regional accessibility, local accessibility, polycentric urban form

Valuing the Benefits of Reductions in Cancer Risk from Mobile-Source Air Toxics: A Regression Discontinuity Approach

Scott Elliot Lowe, UC Santa Barbara

Advisor: Antonio Bento

The goal of this research is to provide credible estimates of the benefits derived from automobile-related regulations to reduce air toxics emissions. The proposed research will explore correlations in housing prices and air toxics cancer risk data to infer a value for reductions in the risk of cancer from exposure to air toxics. A detailed spatial dataset of air toxics cancer risk along with housing attributes and amenities are used to estimate a hedonic price function for housing values. This price function, after being correctly adjusted for statistical problems (Endogeneity), provides an accurate estimate of the value that homeowners place on reduced cancer risk. Using a regression discontinuity estimation approach, I am able create a natural discontinuity between houses that fall in cancer risk hotspots and those that don’t. By doing this I will be able to identify the pure effect of reductions in risk separate from unrelated changes in the housing market. Using these estimated benefits, I can compare these to the costs of the implemented policy mechanisms—increased gasoline and diesel production costs— which will shed light on the cost effectiveness of the policies. The valuations will also potentially provide justification for the proposed regulations to reduce diesel PM emissions, and thus the associated reductions in cancer risk from air toxics.

Key words: regulation, cancer risk, spatial dataset, regression discontinuity, diesel PM emissions

Light-Duty Hydrogen-Fuel-Cell Vehicle Adoption: Early Markets and Vehicle-to-Grid Power in California

Brett D. Williams, UC Davis

Advisor: Daniel Sperling

I propose to examine the commercialization of hydrogen-fuel-cell vehicles (H2FCVs) in California in the context of two strategies employed to facilitate technology adoption: 1) creating innovative value from novel product attributes and 2) strategic niche marketing. The innovative value explored is “vehicle-to-grid (V2G) power”: employing parked vehicles as electrical generators to support the complex operation of the grid. The strategic niche pathway explored is: 1) use of V2G-capable H2FCVs in fleets, 2) production of V2G power by organizational and private vehicles at aggregation sites (“parking-lot power plants”), and 3) H2FCV use by California households at home. The proposed research integrates several methodologies and theoretical frameworks and comprises three phases: Phase I identifies a subset of California households “pre-adpated” to early H2FCV adoption through a “constraints analysis.” Phase II applies an enhanced technology-diffusion framework to the proposed strategic market niche pathway. Phase III uses “recombinant innovation” theory to examine the creation of V2G markets using ICE hybrids and evaluates the potential of parking-lot power plants. Collectively, these research phases evaluate an illustrative pathway of H2FCV market development and are designed to inform public and private decision-makers about the market dynamics of commercializing H2FCV technologies.

Key words: hydrogen-fuel-cell vehicles (H₂FCV), strategic niche marketing, vehicle-to-grid (V2G), innovative value, fleets