UCTC Research Abstracts 2004-5

Dissertation Research

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.  

Auto-mobility, Spatial Isolation, and the Poor

Principal Investigator:
Evelyn Blumenberg
UC Los Angeles
Email: eblumenb@ucla.edu

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

Abstract: This research examines the role of transportation access in explaining the spatial isolation of metropolitan residents. Numerous studies suggest that low-income households tend to be concentrated in resource-poor, central-city neighborhoods, isolated from employment opportunities, consumer goods, services. Surprisingly, only a small sub-set of this spatial isolation research examines how automobile availability and transit service quality affect knowledge of and access to opportunities, goods, and services.

To examine this issue, we will draw on the literature on cognitive models of geographic space to examine how access to automobiles and high-quality transit service affect peoples’ (1) knowledge of their city and (2) the physical boundaries of their daily activity spaces. Specifically, this research focuses on three questions: First, are low-income households with automobiles less spatially constrained than transit-dependent low-income households? Second, are low-income households more spatially constrained than higher income households, controlling for access to household vehicles? And, third, does living in a job- and/or transit-rich neighborhood diminish spatial isolation among those dependent on public transit?

This research will contribute importantly to our understanding of how mobility influences metropolitan residents’ knowledge and perceptions of opportunities, goods, and services. Such understanding is important to the development of both transportation and economic policy.  

Key Words:  low-income, spatial isolation, automobile availability, cognitive models

Objective:  develop strategies for understanding of how mobility influences metropolitan residents

Tasks: Review previous work on the topic, assemble data, analyze data, and prepare reports.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005 

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:  improved transportation and economic policies

Direct Cost: $39,339

 

Housing-Retail Balance, Travel Demand, and Physical Activity

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

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

Abstract:  Jobs-housing balance is today being actively pursued as a land-use strategy for reducing vehicle miles of travel. Since travel for shopping and personal services usually accounts for over twice as many motorized trips as journeys-to-work, this research will examine “housing-retail” balance as a potentially more effective land-use management strategy. Two hypotheses will be tested. One holds that retail-housing balance significantly reduces VMT for shopping and consumer services, with the largest benefits accruing for convenience and non-durable good purchases. The second holds that retail-housing balance increases non-motorized travel, providing physical activity benefits. Using data from BATS 2000, daily activity records will be used to specify 24-hour shop trip incidences, vehicle miles, and durations. Isochronic measures of retail accessibility and housing-retail diversity indices will be measured using 2000 CTPP Part 2 based on two-digit retail job occupational codes. Introducing various statistical controls, nested logit and multiple regression models will be used to test hypotheses, generate travel/land-use elasticities, and provide order-of-magnitude comparisons to findings on VMT reductions associated with jobs-housing balance strategies. Qualitative case work involving interviews of neighborhood residents will elicit attitudinal responses regarding the desirability and design aspects of community retail activities and their influences on travel choices and physical activities.

Key Words:  Housing-Retail Balance; Accessibility; Mode Choice; New Urbanism; Smart Growth; Logit Analysis; Case Studies.

Objective: To measure the degree to which housing-retail balance, as embodied in New Urbanism community designs and smart-growth planning, yields motorized-travel-conserving and physical activity benefits, especially in relation to the more widely embraced strategy of jobs-housing balance.

Tasks: Compile travel data for Bay Area sampled households from BATS 2000; obtain place of employment data on stratified retail jobs from 2000 CTPP Part 2; using GIS and statistical tools, develop isochronic measures of retail-services accessibility and housing-jobs balance indices; compile necessary control variables; test hypotheses by estimating nested logit and multiple regression models; screen candidate neighborhoods for case-based research; select cases and conduct intercept surveys of residents, shoppers, and merchants; review and digest case findings; examine public policy considerations that are informed by the research results; prepare research report and journal articles.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005 

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:  Measure motorized-travel-conserving and physical activity benefits of retail-housing balance for various shopping functions, and provide order-of-magnitude insights of benefits relative to jobs-housing balance strategies.

Direct Cost: $49,425

 

Transaction-Cost Economic Analysis of Institutional Change toward Design-Build Contracts for Public Transportation

Principal Investigator:
David Dowall
UC Berkeley
Email: dowall@berkeley.edu

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

Abstract: This research is a transaction-cost economic analysis of recently completed transportation projects, informing a comparative evaluation of the institutional change in public contracting from design-bid-build to design-build project delivery. Design-build, in which design and construction services are bundled together, is an alternative form of public contract recently adopted by transportation departments in 24 states. With this method, lower production costs may be obtained by beginning construction before design is complete. Such savings, however, may come at the expense of organized labor and public participation, and could reflect higher transaction costs than traditional methods. At issue is the question of whether or not California’s Department of Transportation should also engage in design-build contracting. This research will produce pair-wise case studies and a quantitative database explaining the benefits and cautions of these two modes of delivery. Research techniques will include semi-structured interviews, the review of documents and archival records, and mining online legal and news sources. Analysis will proceed by triangulating evidence to validate or refute propositions from transaction cost economics against rival theoretical interpretations of institutional change. 

Key Words: transaction-cost economics, public contracting, project delivery, design-build

Objective:  Comparative evaluation of recently completed surface transportation projects developed according to design-bid-build and design-build methods, assessing actual transaction and production costs as well as relative impacts to organized labor and public participation, with reference to the question of whether or not the State of California should pursue a policy of programmatic design-build contracting.

Tasks: 1. Case Selection 2. Develop Case Study Protocol 3. Data Collection 4. Analysis and Writing

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005 

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:  Results should provide decision-makers in California with plausible explanations of the benefits and cautions of design-build contracting while enriching academic discourse on the topic of project delivery with the theoretical depth available from the literature of new institutional economics.

Direct Cost: $78,272

 

Activity-Oriented Scheduling/Activity Survey and Analysis Via a Unified Real-time Data Collection Framework

Principal Investigator:
Reginald Golledge
UC Santa Barbara
Email: golledge@geog.ucsb.edu

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

Abstract: In the previous research, we have developed a conceptual model of real-time activity scheduling/implementation data collection system. It is operationalized as a wearable computer complete with GPS recorder and wireless WAN card. The wearable computer features with real-time activity decisions tracking and activity pursuit recording in field. It empowered the transport researchers a unique research means to 1) identify the temporal-spatial decision making structure embedded in activity scheduling 2) the linkage between activity decision-making and associated actual activity execution. The goals of this research are: (i) to further improve the real-time system that incorporates the extraction of activity scheduling and execution information within one unified data collection framework with an up-to-date equipment and system functions; (ii) identify a unified conceptual ontology to explore and explain the dynamics and interaction of activity scheduling and execution; and (iii) most importantly, explicitly define the mechanism in which the formulation of people’s activity schedules are subject to the influence of the social-demographic and temporal-spatial constraints that gradually lead to the activity-travel patterns detailed by passive, observing survey methods.  

Key Words:  Real-Time Data Collection; Wearable Computer; Activity Behavior

Objective:  To test and evaluate the potential for use of a real time wearable data collection system developed in a previous UCTC funded project.

Tasks: 1. To develop additional Pocket PC-based real time data collection devices 2. Collect data for 40 participants (in real time) of one week’s daily activity patterns and to analyze the data.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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:  Should reduce or eliminate data loss usually missed by surveys and post-hoc interviews.

Direct Cost: $74,040

 

Wet Pavement Accidents on California Highways: Causes, Concentrations, and Potential Means for Reduction

Principal Investigator:
Thomas Golob
UC Irvine
Email: tgolob@uci.edu

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

Abstract: The proposed research involves an a statistical analysis of accidents that have recently occurred on California Highways during inclement weather. There are well-known countermeasures for reducing the number and severity of wet pavement accidents, and the key is to identify where to apply each countermeasure to achieve maximum benefit. Caltrans already has processes in place for identifying treatment projects, and this research begins by becoming familiar with these procedures and their supporting data. Working together with Caltrans, a sample of roadway segments of the California State Highway System will be chosen, and an historical dataset will be developed by combining accident data with detailed roadway characteristics, weather conditions, and traffic. The analysis phase will then focus on determining how the propensity for accidents (by type and severity) is related to roadway geometrics, pavement factors, and the weather and traffic conditions prevailing at the time of each crash. The final phase of the project will focus on means of integrating the new results into performance monitoring and planning procedures.

Key Words:  Traffic accidents, highway safety, wet pavement, accident reduction

Objective:  The objective is to improve the process for identifying projects that improvement roadway safety under conditions of wet-pavements.

Tasks: Process Review; Select a Sample of Roadway Sections; Gather Data on Roadway Characteristics and Weather Conditions; Match with Accident Data; Analyze Relationships Among Accidents, Weather, Roadway Characteristics and Traffic Flow; Compare Notes with Caltrans Personnel to Interpret Results in a Planning Context

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005 

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 is intended to aid Caltrans and other State Federal and State Agencies in identifying problems related to traffic safety during wet weather conditions and it is intended to provide guidelines for designing countermeasures to reduce the number and severity of traffic accidents.

Direct Cost: $66,814

 

The Davis Bicycle Studies

Principal Investigator:
Susan Handy
UC Davis
Email: slhandy@ucdavis.edu

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

Abstract: As a means of transportation and as a form of physical activity, bicycling generates benefits to the bicyclist as well as to the community as a whole. Bicycling now accounts for less than 1 percent of all trips for all purposes in the U.S., but evidence from other western countries suggests that under the right conditions, bicycling levels can be significantly higher. The experience of Davis, California suggests that it is possible to create conditions conducive to higher levels of bicycling in the U.S. However, the extent to which public policies have contributed to bicycling levels in Davis has not been rigorously assessed. This project aims to fill that gap through a quasi-experimental study of bicycling behavior in Davis and comparison communities designed to determine the influence of bicycle infrastructure and mixed land-use patterns relative to individual preferences, community culture, and other factors. To provide a policy context for the behavioral analysis, the project will document the history of bicycle policy in Davis and compare bicycle infrastructure in Davis to other bicycle -oriented college towns. This project is planned as the first in a series of complementary studies of bicycling behavior focused on Davis.

Key Words:  bicycling, bicycling behavior, bicycle policy, bicycle infrastructure, mixed land use patterns, preferences, culture

Objective:  The objective of this project is to rigorously assess the extent to which public policies have contributed to bicycling levels in Davis using a quasi-experimental design.

Tasks:
1. Literature review
2. Development of the survey instrument
3. Development of sample
4. Implementation of survey
5. Analysis of survey data
6. Historical review of Davis policies
7. Inventory of infrastructure and policies
8. Final report

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005 

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 contribute to a stronger empirical basisfor policy decisions about bicycle infrastructure.

Direct Cost: $71,501

 

Testing Spatial Mismatch: A Structural Equations Modeling Approach

Principal Investigator:
Robert Johnston
UC Davis
Email: rajohnston@ucdavis.edu

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

Abstract: We will critique past studies of the Spatial Mismatch Hypothesis and then apply three structural equations models to data for the Sacramento, California region. We will estimate both cross-sectional and dynamic models and we will use a network-based travel model to measure accessibility to jobs.

Key Words:  Spatial Mismatch, structural equations models, cross-sectional and dynamic models, accessibility to jobs.

Objective:  The objectives of this study are to: 1. examine the relations between employment, auto ownership, income, job accessibility, and other variables using structural equations models; 2. to test whether simultaneity exists; and 3. to determine if simultaneity results in biased estimates, by comparing these results to those from multiple regression models.

Tasks: 1. Develop two multiple regression models by using pooled data, whites only, and blacks only, on 1990 and 2000 datasets as comparison bases. 2. Develop separate cross-sectional structural equations models for 1990 and 2000 for pooled data, whites only, and blacks only. 3. Compare the models developed in Task 1 and Task 2. 4. Develop unconditional change-score structural equations models for pooled data, whites only, and blacks only. 5. Develop two-wave structural equations models for pool, whites only and blacks only.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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:  To our best knowledge, we will be the first researchers to use a dynamic nonrecursive structural equations model to test spatial mismatch.

Direct Cost: $38,756

 

Street Trees and Intersection Safety

Principal Investigator:
Elizabeth Macdonald
UC Berkeley
Email: emacdon@berkeley.edu

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

Abstract: For at least 250 years, the finest of streets the world over have been associated with trees. Elm or oak shaded residential and commercial main streets remain as memories, but seldom as realities, of the best American urbanism. In the automobile age, a real concern with safety has resulted in tree spacing standards in the United States that dictate long setbacks from intersections, ostensibly geared to achieving unobstructed sight lines for drivers. This research proposal starts with a premise that sidewalk street trees should not be restricted unless it can be shown unequivocally that they create unsafe environments. The research will: investigate standards in California communities to see how they interpret engineering guidelines on tree placement at intersections; use new three-dimensional computer modeling and drive through animation techniques and Geographic Information Systems tools to model and analyze a variety of typical urban intersections to determine the impacts of trees and other intersection paraphernalia (e.g. newsracks and parked autos) on drivers’ ability to see approaching cars; conduct controlled experiments to ascertain what drivers notice at intersections; and explore available data on tree-related accidents at intersections. Findings and conclusions will be published in an Institute of Urban and Regional Development Working Paper.

Key Words:  Intersection design, street standards, street trees

Objective:  To use three-dimensional modeling techniques and GIS spatial analysis tools to test whether street trees near intersections significantly block a driver’s visibility of approaching vehicles.

Tasks: Task 1: Gather street design standards from California cities, analyze restrictions on street trees and other objects at intersections, and compile the data in tabular form.
Task 2: Create three-dimensional computer models of typical urban intersections where a minor road intersects with a major road. For each typical intersection, create versions without and without street trees, with and without parked cars, and combinations of each.
Task 3: Create snapshot images of what a driver on the minor road would see when stopped at each simulated intersection, looking to the left and to the right. Import these images into a GIS spatial database, and calculate areas of visibility.
Task 4: For each modeled intersection, create a drive-through simulation that represents what a driver would see when moving along the minor road, stopping at the intersection, and scanning the roadway to look for approaching vehicles.
Task 5: Conduct controlled experiments in which participants (a minimum of 100) are asked to view the drive-through simulations and indicate when they notice approaching cars, and analyze the data.
Task 6: Gather accident data for the city of Oakland, California, and analyze whether or not there is any correlation between high accident rates and intersection street trees.
Task 7: Prepare a final report outlining the study findings, including proposals for revised policy recommendations. Prepare as well the Final Summary Report for UCTC.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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 research will result in revised policy advises regarding street trees at intersections, new three-dimensional methods for modeling and testing intersection design, and new objective GIS measures for calculating areas of visibility.

Direct Cost: $47,113

 

Robust Optimal Maintenance and Rehabilitation Policies in Asset Management

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

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

Abstract: Robust optimization is a modeling methodology to solve optimization problems in which the data are uncertain and only known to belong to some uncertainty set. The proposed research will use this modeling methodology to obtain robust maintenance and rehabilitation (M&R) policies for individual infrastructure assets. Using field and laboratory data, alternative methods will be used to characterize the uncertainty with regards to infrastructure facility deterioration. By considering the defined uncertainty set, a robust counterpart of the original infrastructure maintenance problem will be created. Efficient solution algorithms will be developed to solve the robust counterpart or good approximations thereof. Finally, case studies will be performed to evaluate the usefulness of the proposed approach in reducing M&R expenditures. The proposed research is the first adaptation of robust optimization methods to asset management. The research will also contribute to the literature on robust dynamic programming in the context of Markov decision processes. The results of this research will improve the confidence of public works agencies in asset management systems and thus facilitate the acceptance and deployment of these systems.

Key Words:  Robust optimization, uncertainty set, maintenance and rehabilitation, infrastructure assets, asset management, Markov Decision Process.

Objective: The objective of this research is to develop a prototype of an asset management system that uses robust optimization to produce M&R policies that are less sensitive to the input data.

Tasks: Literature review; Infrastructure facility uncertainty modeling (for highway pavements); Investigation of alternative uncertainty models; Formulation and solution of robust optimization problem; Extension to infinite horizon problems

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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 proposed research has the potential to reduce the costs associated with errors in modeling deterioration rates while managing infrastructure assets, which will help realize the full potential of asset management systems and thus facilitate the acceptance and deployment of these systems by public agencies.

Direct Cost: $42,735

 

Modeling the Adoption of Teleshopping

Principal Investigator:
Patricia Mokhtarian
UC Davis
Email: plmokhtarian@ucdavis.edu

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

Abstract: Fully understanding the potential transportation impacts of new and old shopping alternatives requires investigating the adoption of the various alternatives. This multi-year study proposes to design, administer, and analyze an original survey of shopping attitudes and behavior, leading to a model of shopping mode choice. To reduce the heterogeneity of shopping behavior, we focus on one or two frequently-purchased product classes. We define alternatives in terms of the dimensions of pre-purchase behavior (with store, catalog, and Internet modes) and transaction behavior (store, phone, mail, and Internet modes, distinguishing auction sites from conventional e-tailers). Research questions include: (1) For the selected product class(es), what are the advantages and disadvantages of each shopping mode? (2) Can market segments with different propensities to use alternative modes be identified? (3) To what extent are there perceived to be viable alternative modes for a given shopping occasion? (4) Are the various shopping modes substitutes, or complements? Offering the option of paper or web-based surveys, we plan to obtain about 2,000 responses. The first year of the study is mostly devoted to survey design, data collection, and cleaning, with some preliminary descriptive analyses. Future years will involve various multivariate statistical analyses and multidimensional discrete choice modeling.

Key Words:  shopping mode choice, teleshopping, e-shopping adoption, B2C e-commerce

Objective: To better understand the circumstances under which the alternative shopping modes of store, catalog, and Internet are chosen, which has implications for the future transportation impacts of teleshopping.

Tasks: 1. Literature review: ongoing; 2. Survey design, paper and Internet versions: August 04 – January 05; 3. Sample extraction and recruitment: November 04 – January 05; 4. Administer survey: February 05; 5. Data entry and cleaning (paper surveys): March – April 05; 6. Initial descriptive analyses: May – June 05; 7. Write up findings: July 05

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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: Increased insight into individual responses to ICT-based shopping alternatives, having direct travel and indirect urban form implications.

Direct Cost: $81,379.30

 

Estimating Activity Rates and Emissions from Heavy-Duty Construction Equipment

Principal Investigator:
Debbie Niemeier
UC Davis
Email: dniemeier@ucdavis.edu

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

Abstract: The proposed research will help Caltrans estimate emissions from transportation project construction activities. The research will expand upon existing work at UC Davis (UCD) to develop a construction emissions spreadsheet tool. Using construction diaries created by Caltrans, the research team will estimate a range of construction equipment activity for six project types representative of virtually all of the transportation projects completed by Caltrans. In addition, the research team will synthesize existing literature regarding estimating construction activity, and provide guidance for project analysts charged with estimating emissions from specific projects. The work products will illustrate how to use construction equipment activity data to estimate emissions, using a new constructions emissions spreadsheet tool developed by UCD under Caltrans sponsorship. Air quality management districts recognize the growing importance of non-road mobile source emissions, and are increasingly asking Caltrans to estimate emissions from non-road equipment. The research will enable Caltrans to respond to these information requests, and to test the sensitivity of emissions estimates to various project elements.

Key Words:  non-road activity, construction emissions, non-road inventory

Objective: To estimate a range of construction equipment activity for six project types representative of virtually all of the transportation projects completed by Caltrans.

Tasks: 1) Interview Caltrans staff & identify representative projects; 2) Define projects to be evaluated; 3) Obtain and evaluate construction diaries; 4) Analyze and evaluate construction activities; 5) Summarize existing resources; 6) Update construction emissions modeling spreadsheet; and 7) Prepare guidance document

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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 project work products will, for the first time, document the range of equipment activities associated with transportation construction projects undertaken in California.

Direct Cost: $47,082

 

Modeling Car Ownership Rates, and Age and Value of Vehicles

Principal Investigator:
Paul Ong
UC Los Angeles
Email: pmong@ucla.edu

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

Abstract: The proposed study develops new approaches to modeling car ownership rates, and the age profile and average value of the automobile stock using aggregated data for small geographic units (census tracts or zip-code areas). This type of information is critical to large-scale urban transportation models and models of air pollution from mobile sources. Ideally, these models should be based on understanding the underlying causal factors that determine the number and characteristics of household vehicles. Current models using census data are limited to ownership rates without the ability to examine age and value, and the current analytical approach has a serious econometric problem with the endogeneity of household income as a dependent variable. The proposed study overcomes these limitations by combining census data with non-census data, and by using an instrumental variable approach to examine variations across tracts or zip-code areas in Los Angeles County. Socioeconomic and demographic characteristics come from the 2000 census, the exogenous cost of car ownership comes from insurance quotes, and the age and value information is based on a special tabulation of data from the Department of Motor Vehicles.

Key Words:  car ownership, age and value of cars, instrumental-variable approach

Objective: The research objective of the proposed study is to estimate three equations using an instrumental-variable approach – automobile ownership rate, the proportion of the vehicle stock over ten years old, and the average value of the vehicle stock.

Tasks: The project will have three major tasks. The first is assembling the required data, which will be done primarily by a graduate student under my supervision. The second task is analyzing the data, which will be done by me with assistance from the graduate student. The third task is writing up the results, which will be done by me. If the graduate student is interested and capable, she or he will be invited to co-author a paper.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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 research will provide develop insights that will ultimately enhance key components of transportation and pollution models, and will enable policy analysts better able to examine what factors can be influence to enhance transportation resources for households.

Direct Cost: $49,748



Similarity Analysis for Estimation of an Activity-based Travel Demand Model

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

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

Abstract: In this research, we propose to develop an estimation procedure for a particular mathematical programming activity-based model in order to estimate the relative importance of factors associated with spatial and temporal interrelationships among the out-of-home activities that motivate a household’s need or desire to travel. The method will employ a genetic algorithm to estimate coefficient values of the utility function, based on a particular multidimensional sequence alignment method to deal with the nominal, discrete, attributes of the activity/travel pattern (e.g., which household member performs which activity, which vehicle is used, sequencing of activities), and a time sequence alignment method employing an inner product metric to handle temporal attributes of the activity pattern (e.g., starting and ending time of each activity and/or travel). The estimation procedure will be tested on data drawn from a well-know activity/travel survey.

Key Words:  activity-based, estimation, sequence alignment, activity pattern

Objective: This research will establish a consistent metric and develop a procedure for activity-based travel demand model estimation.

Tasks: 1) Adaptation of existing sequence alignment techniques to examine similarity among the Activity, Person, and Vehicle dimensions; 2) Development of Activity Sequence (or Order) similarity indices; 3) Development of indices measuring the overlap in time spent on out-of-home activities without respect to specific activity-person and/or specific activity-vehicle linkages; 4) Development of inner product metric for temporal similarity; 5) Testing of similarity results under various weight scorings to determine the critical weights for general cases; 6) Estimation and validation on a sample drawn from households in the “so-called” Portland Activity data set.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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 research will establish a mechanism for operationalizing consistent estimation household activity scheduling problem.

Direct Cost: $47,450



Capacity Modeling for Large Scale Urban Multimodal Freight Transportation Systems

Principal Investigator:
Amelia Regan
UC Irvine
Email: aregan@uci.edu

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

Abstract: This research is designed to develop an analysis tool that can estimate the capacity of multimodal freight transportation systems. Lack of sufficient capacity for freight transportation is increasingly becoming a major issue in metropolitan areas throughout the United States, particularly in Southern California. In supporting economic expansion goals, it is important to consider how transportation investments can sustain the continued growth of the economy. Traditional approaches to capacity preservation and expansion, especially in urban areas, have proven to be inadequate, mainly due to the high cost of land use, environmental concerns, physical barriers, and opposition from local communities. There is broad recognition of the need for comprehensive multimodal approaches that leverage the competitive advantages of each mode. Existing methods for capacity analysis, however, do not adequately address the distinct features of multimodal systems. Because of the complexity relevant to multimodal plans and projects, new methods to evaluate current usage and potential capacity of multimodal systems are desired to be developed. The model we intend to develop would assist transportation planners and infrastructure managers in making the most efficient use of existing capacity and in improving their decision-making related to transportation planning and investment.

Key Words:  Intermodal Freight Transportation System, Capacity Modeling, Multiple Commodity Network Flow Problems

Objective: To develop new capacity modeling tools.

Tasks: 1) A comprehensive literature survey and gap analysis will be conducted with the relevant references and guidance materials in order to improve our knowledge baseline on multimodal freight transportation systems and capacity analysis. 2) A mathematical formulation and algorithm will be developed in the second phase 3) The reasonableness and applicability of the developed model will be tested and assessed.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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 research could develop useful planning tools and also provide insight into intermodal freight bottlenecks in the case study region.

Direct Cost: $55,348



Cruising for Parking

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

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

Abstract: Where curb parking is cheaper than off-street parking but all curb spaces are occupied, drivers who want to park their cars are presented with a choice: they can spend time cruising for curb parking or spend money to park off-street. Since curb parking is under priced, drivers have an incentive to search for curb spaces. Cruising is individually rational but collectively harmful because it increases traffic congestion, air pollution, fuel consumption, and accidents. In the proposed research, we will study the effects of cruising for under priced curb parking. We will measure the time it takes to find a curb space in Westwood Village, a commercial district adjacent to the UCLA campus, and estimate the share of traffic that is cruising for parking. From these findings we will measure the congestion effects directly related to under priced curb parking. To accomplish these objectives, we will examine the following: 1) the average cruising time before finding a vacant curb space, 2) driver’s strategies in cruising for parking 3) the average parking duration at curb spaces, 4) the share of traffic that is cruising for parking, and 5) the transportation improvements that would occur if cruising were reduced by correctly pricing curb parking.

Key Words:  parking, cruising, congestion

Objective: To quantify the effect of under priced curb parking..

Tasks: Review previous work on the topic, assemble data, analyze data, and prepare reports.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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: improved understanding of driver behavior when parking - greater understanding of the dynamics of curb parking

Direct Cost: $53,978

Why Do Inner City Residents Pay Higher Premiums? The Determinants of Automobile Insurance Premiums

Principal Investigator:
Michael Stoll
UC Los Angeles
Email: mstoll@ucla.edu

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

Abstract: The proposed study examines the relationship between traffic density, vehicular accident and automobile insurance premiums across sub-metropolitan areas. It is widely known that inner city residents pay higher premiums, holding car characteristics constant, than others, but there is very little systematic research to explain why. We propose to disentangle two competing explanations for these higher premiums: the higher rates are the product of racial discrimination (“red lining”), and the higher rates are due to a higher cost for insuring inner-city residents as a result of their greater risk. We will examine whether inner city residents are exposed to greater vehicle risks and whether these greater risks can account for their higher premiums. These risks include greater exposure to automobile accidents because metro area vehicle traffic is much more dense there than elsewhere, and greater exposure to high crime neighborhoods (higher car theft risks), among other factors. The proposed study will use multivariate econometric models to test these hypotheses, after accounting for other relevant factors. The analysis will examine variations across small geographic areas within Los Angeles City using both census and non-census data. This topic is important because the recent literature has shown that higher insurance premiums adversely impact inner city residents’ ability to purchase and maintain cars, which in turn has been shown to have an important influence on their ability to gain employment.

Key Words:  Auto Insurance, Redlining, Vehicle Accidents, Inner city residents

Objective: To improve understanding of the factors that account for the higher auto insurance premiums paid by inner city residents.

Tasks: assemble and geocode data, analyze data, prepare research article

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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 research will provide greater insight into the potential barriers to car ownership of inner city residents, which in turn could improve their employment opportunities

Direct Cost: $47,305



Motor Fuel Price and Expenditure Effects on Vehicle Use in California

Principal Investigator:
Martin Wachs
UC Berkeley
Email: mwachs@berkeley.edu

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

Abstract: Motor vehicle fuel costs lie at the intersection of several national transportation policy debates. Transportation efficiency and equity concerns are a common thread in these debates. They include how best to reduce gasoline consumption; how to understand rising personal mobility costs and burdens on low income households; and how to address the quiet revolution in the nation’s transportation finance system, shifting from user fees to general taxes and borrowing to support investment. A clear understanding of how vehicle fuel prices and household fuel expenditures affect household transportation choices and costs would shed needed light on these discussions and related policy choices. New data collected by the 2001 National Household Travel Survey (NHTS) provide far better estimates of household vehicle fuel economy, annual fuel expenditures, and vehicle miles traveled (VMT) than available previously. These NHTS data, together with the debates outlined, motivate this effort. This study will use the data to: 1. identify variation in fuel prices and annual household fuel expenditures; 2. estimate the value of specific household trip types; 3. model the price sensitivity of demand for fuel among California households; and 4. suggest how different policy alternatives, including a potential motor fuel tax increase, could affect fuel consumption, vehicle acquisitions and travel among California households.

Key Words:  National Household Travel Survey, Fuel Price, Fuel Expenditure, Motor Fuel Tax, Equity

Objective: Using newly available NHTS data, this study will study how vehicle fuel prices and household fuel expenditures affect household transportation choices and costs and consider the implications for policy choices that address transportation finance concerns.

Tasks: Applying standard statistical analysis and modeling techniques to the NHTS data, this study will address the following tasks:
1. Analysis of Variation in Fuel Prices and Household Fuel Expenditures;
2. Estimation of Trip Costs and Value of Specific Trip Types;
3. Model of Price Sensitivity of Demand for Fuel; and
4. Discussion of Policy Alternatives, including a motor fuel tax increase, could affect fuel consumption, vehicle acquisitions and travel among California households.

Milestones, Dates: Official start date Aug. 1, 2004, end July 31, 2005

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 project would provide a far more reliable picture of variation in household fuel expenditures and fuel prices than previously available. Comparing fuel expenditures with household and driver demographic information as well as with trip characteristics will allow for a nuanced view of how different households and drivers value different trips and of the price elasticity of demand for trips and travel in the face of fuel price increases.

Direct Cost: $33,129

 

Dissertation   Research:

 

The Economic Effects of Transportation Infrastructure on the Value of Time

James Barrett, UC Davis

Advisor: Douglas Larson

Construction of transportation infrastructure, designed to relieve congestion, can lead to increased driving and sprawl. To justify expenditures and negative externalities associated with infrastructure construction, transportation studies must deduce the value of time saved. Methods used to determine the value of time include discrete choice and contingent valuation or revealed preference and stated preference methods. There is little comparison between these methods and the current level of transportation infrastructure is typically considered exogenous. My dissertation will contribute to improving our understanding of the differences and similarities between models used to determine values of time. I will examine the effects of the current level of transportation on these values. Specifically, my dissertation asks:
• Are there differences in the values of time calculated for each method?
• Are the methods consistent in the order of time values for various activities?
• What is the relationship between values of time and demographic characteristics?
• Is the order of time values for different demographic characteristics the same for each model?
• What is the relationship between transportation infrastructure and values of time for various activities, including non-transportation related activities?
• Is the order of time values between regions the same for each model?

 Key words:  transportation infrastructure, congestion, value of time, discrete choice,  contingent valuation, revealed preference, stated preference

 

Causality between the Built Environment and Travel Behavior: Static and Dynamic Structural Equations Modeling Approaches

Xinyu Cao, UC Davis

Advisor: Patricia Mokhtarian

The link between the built environment and travel behavior has long been of interest to urban planning. Although new urbanists advocate that neo-traditional developments offer an effective way to reduce auto dependence, our understanding about this causal link is still tentative due to the influences of residential self-selection. In addition, the use of cross-sectional data remains one important obstacle among others that hinder current understanding. Using a survey of 1682 respondents in Northern California, this study examines the causal relationships between the built environment and travel behavior using structural equations modeling in static and dynamic contexts. This study represents a step beyond static description and simple regression models of relationships between the two, to an exploration of how they interact in a dynamic and integrated model. The results of this study are expected to provide new insight into these relationships (especially the panel model) and the extent to which these relationships are due to genuine causality or third-party correlation with other factors. Such results will offer a more realistic picture to policy makers and urban planners on how to promote and replicate traditional neighborhood designs, and contribute to more accurate prediction of the effectiveness of such policies and designs.

 Key words:  built environment, travel behavior, neo-traditional developments, self-selection, structural equations modeling, panel model.

 

D isaggregate Structural Relations Between Residential Location, Job Location, Travel Behavior, and the Use of Telecommunications

Gustavo Oscar Collantes, UC Davis

Advisor: S usan Handy

 The research described in this proposal focuses on the analysis of the relationships between individual and household choices about residential location, job location, travel behavior, and the use of information and telecommunications technologies (ICT), at the disaggregate level, from a transportation policy perspective. Structural equation modeling will be used to explicitly account for the endogeneity of these general variables, which will constitute an improvement over the extant literature. This methodological approach will enable the analysis of multidirectional directions of impact between the endogenous variables. The main data source is the 2000 San Francisco Bay Area Activity Survey. Useful information on the impact of ICT on travel behavior can be obtained from these data. An additional improvement relative to previous related structural models resides in the more comprehensive conceptualization of residential location, which will consider regional as well as local accessibility measures. By testing for structural relationships and quantifying the magnitude of those relationships, this research will provide a basis for the evaluation of potential impacts of alternative land use and transportation policies.

 
Key words: residential location, job location, travel behavior, information and telecommunications technologies, structural equation modeling, land use and transportation policies

 

An Operating Technique to Reduce Regional Port Traffic and Improve Port Productivity

Anne Victoria Goodchild, UC Berkeley

Advisor: C arlos Daganzo

 This research address two key transportation issues; constrained capacity at intermodal terminals, and traffic congestion on urban road networks. This research will demonstrate the costs and benefits of a port operations technique that has the potential to reduce land-side traffic and improve port productivity. The technique, called double-cycling, can improve the utilization of land-side vehicles and quay cranes when used to unload and load container vessels by reducing the number of unproductive movements. Currently, about half of all chassis trips are unproductive, these include trips between the quay crane and local storage, often a distance of 10 miles or more. Similarly, approximately half of all quay crane moves are empty. With careful planning, these wasted movements can be converted into productive movements by unloading and loading vessels simultaneously. Many mathematical tools will be used to understand the relationships between benefits and port parameters. The result will be useful formulas and planning tools that can be used at both the planning and operational level to design systems, plan operations and quantify benefits. Research to date indicates that significant reductions can be made to regional port traffic and ship turn-around time. Specific savings depend on the details of each port but 25-30% reductions would not be unusual.

 
Key words: intermodal terminals, traffic congestion, port operations, port productivity, quay cranes, container vessels, chassis, regional port traffic, ship turn-around time.

 

Robust and Optimal Strategies for Asset Management in the Presence of Epistemic Uncertainty

Kenneth David Kuhn, UC Berkeley

Advisor: S amer Madanat

 Public works agencies use asset management systems to optimize returns in M & R expenditures. These system are built around a series of assumptions regarding infrastructure decay, and no guarantees can be made about system performance if these assumptions are violated. The goal of this research is to use newly developed robust optimization techniques to consider epistemic uncertainty and potential errors in initial assumptions when making maintenance and rehabilitation decisions. Research tasks will include representing uncertainty in a meaningful and understandable way and finding computationally tractable solution algorithms for robust optimization of a network level asset management problem.

 
Key words: asset management systems, robust optimization, epistemic uncertainty.

 

Moisture Sensitivity of Hot Mix Asphalt Pavements

Qing Lu, UC Berkeley

Advisor: John Harvey

Moisture sensitivity in hot mix asphalt (HMA) pavements is one of the leading pavement performance related issues facing highway agencies. During the past two years, it has been the number one issue and priority for the Asphalt Concrete Task Group of Caltrans. Current test methods for identifying moisture sensitive mixtures, however, do not accurately predict field performance. The objectives of this research are to investigate the effect of moisture on the fundamental properties of asphalt concrete, and to develop a performance based test and evaluation procedure for identifying moisture sensitivity of HMA and evaluating the effectiveness of anti-stripping treatments. The research includes two phases: field data collection and analysis, and laboratory test procedure development. The first phase involves collecting complete information, including mix design, pavement structure, traffic and environmental factors, construction quality and performance, for around 150 pavement sections randomly distributed in California. Statistical analysis on this data set will reveal the possible main factors contributing to moisture damage in the field. The second phase will focus on evaluation of current moisture sensitivity test methods, and development of new test and evaluation protocols incorporating the main factors. This phase includes and specifically addresses material selection, specimen preparation, specimen conditioning and testing, and recommendations as to evaluation criteria. Implementation of this research will aid in the identification of moisture sensitive mix and proper selection of corrective treatments, thereby economically increasing the life of HMA pavements.

Key words: moisture sensitivity, hot mix asphalt (HMA), pavement performance, test and evaluation procedures, anti-stripping treatments

 

Children’s Travel: Patterns and Influences

Noreen McDonald, UC Berkeley

Adviser: Elizabeth Deakin

Children’s travel lies at the intersection of policy debates on transportation planning, public health, and urban design. Dramatic shifts in how children get to school – 18% used autos in 1969 compared to nearly 50% today –and increased attention to the effects of the built environment on travel and public health have led to policies which promote walking and biking for the school trip. However, policy is being made without a solid understanding of how youth travel and why they use these methods. I propose to address this hole in our understanding by describing and analyzing the general patterns of children’s travel. The goal of this research is to establish a baseline knowledge of the patterns of children’s travel and to identify the key factors in deciding how children travel, particularly to school. Statistical techniques, i.e. Poisson regression and logit modeling, of activity diaries for 15,000 households in the Bay Area and group discussions with parents and children form the core of the research methodology. The results will be useful in developing policies around school transportation and models of children’s travel behavior.

 Key words: children's travel, transportation planning, public health, urban design, Poisson regression, school transportation

 

Auction Mechanisms in Electronic Freight Transportation Service Procurement for Spot Markets

Srinivas Phaneendra Nandiraju, UC Irvine

Adviser: Amelia Regan

B2B e-Commerce facilitates the reduction of supply chain intermediaries, reduces transaction costs and efficient fleet utilization. This revolution has spawned a number of online marketplaces for freight transportation service procurement. Tendering freight in these marketplaces is becoming increasingly popular, but there is a dearth of decision tools to aid the shippers and carriers or the 3PL logistic firms in these procurement processes. Our research focuses on the online procurement markets in truckload (TL) and less-than-truckload (LTL) in a spot market environment. We will explore the development of auction mechanisms for multi-attribute transportation auctions in TL and LTL service networks; and also formulate carrier-bidding strategies in these auctions. Multi-agent simulations will be performed on carefully designed problem sets to capture the interaction of shippers and carriers in these auction processes.

 Key words: Business-to-Business marketplaces, Freight Transportation, Service Procurement, Auctions mechanism design, Competitive bidding.

 

Does Walkability Influence Walking Travel Behavior? Defining the Effect of Walkability on Mode Choice, Acceptable Walking Distance, and Perception of Danger in Home-to-Transit Station Trips

Sungjin Park, UC Berkeley

Adviser: Michael Southworth

This research will test the claimed but unproven effect of walkability on walking travel behavior. Through three case studies focusing on transit station areas, this research will test three hypotheses: a higher degree of neighborhood walkability (1) increases transit users’ likelihood to choose walking over other travel modes to the station, (2) increases transit users’ acceptable walking distance and (3) reduces transit users’ perception of danger in their trip to the station. Each case study will have either a pair of neighborhoods sharing a single station or two station areas, and they will be compared in terms of walkability and small-scale commercial activities while controlling other socio-economic variables. There will be four phases to my research: station area profiling, physical observation, a station-based user survey, and a focus group study. The first phase will produce the most comparable pairs of study sites for my case studies. The physical observation will evaluate the level of walkability by measuring a full spectrum of walkability indicators. The station-based user survey will measure the travel behavior of transit users’ trips to the station through a self-administered questionnaire. Finally, the focus group study will yield a more detailed explanation of possible differences in travel behavior.

 Key words: walkability, walking travel behavior, travel modes to the station, acceptable walking distance, perception of danger.

 

A Bottom-up Reliability-Based Bridge Management System

Charles-Antoine Robelin, UC Berkeley

Adviser: Samer Madanat

The goal of the project is to develop a tool to help transportation agencies make decisions regarding bridge inspection and maintenance. The proposed bridge management system is a reliability-based optimization model of inspection and maintenance for a system of highway bridges. The objective in the formulation is to minimize the total expected social cost, including the expected cost of failure. The frequency of inspections is included as a decision variable. The probability of failure is explicitly taken into account in the constraints. A bottom-up approach is used, which allows for bridge-specific details to be taken into account. Most existing network-level models assume that component deterioration is memory-less; however, this assumption is relaxed in this project, and time dependent deterioration models are used. The formulation is flexible enough to accommodate different types of facilities, deterioration processes, and failure modes. Proposed extensions include the use of reinforcement learning techniques to take into account the epistemic uncertainty in the deterioration model, and the formulation of the problem in a multiobjective framework. Parametric studies will be conducted in order to assess the performance of the proposed models, and compare their performance to that of existing models.

 Key words: inspection, maintenance, bridge management system, reliability-based optimization, bottom-up, reinforcement learning, multiobjective.

 

A Self-Organizing Decentralized Traffic Information System Based on Adaptive Power and Interval Control Broadcast in Inter-Vehicle Communication

Huaying Xu, UC Riverside

Adviser: Matthew Barth

Advanced Traffic Information System (ATIS) can help drivers adjust their routes, avoid congestion, and thus to improve the efficiency of existing roadway system. Traditional traffic information systems typically involve a central authority that collects data from the street network, processes them in traffic management centers and disseminates traffic analysis result to the drivers. Since a central unit covers a relatively large area and due to the limited bandwidth for transmitting the traffic messages, the broadcasted information needs to be general and cannot include specific details on the area close to the current position of the driver. Based on inter-vehicle communication, a self-organizing traffic information system can be fully decentralized. It can disseminate local detail traffic conditions in a very short time and can complement conventional traffic information systems with detail local traffic information. Suitable communication protocol should be utilized in the systems so that vehicle state information and analyzed traffic information can be reliably and efficiently disseminated without the aid of infrastructure. Three innovation elements are included in this research dissertation. First, we propose a self-organizing scheme for traffic information system design based on adaptive power and interval control broadcast. In our design, each node can adapt their transmission power and transmission interval according to the local environment in order to avoid collision in high traffic density situations or network fragmentation in low traffic density situations. Second, an integrated environment for both traffic and network simulation will be implement. Communication networking and vehicle traffic simulation tools exist separately so far. It is crucial to have an integrated simulation environment to evaluate both communication performance of our protocol and its effectiveness to improve roadway efficiency. Third, we also aim to estimate radio propagation models for different traffic environments in this research. Most of the recent channel propagation measurements were performed in the indoor environment and little attention about channel measurement has so far been paid to traffic environment. We will use our channel models in the integrated simulation environment.

 Key words: Self-organizing Traffic Information System, inter-vehicle communication, power control, transmission interval, propagation, broadcast