Current Funded Research

A Measurement-based System for TMC Performance Evaluation
Investigators: Will Recker, Craig R. Rindt, James E. Marca
Support: California Department of Transportation

Caltrans has a significant investment in TMCs throughout the state. These operations centers are tasked with maintaining the safety and efficiency of California's highways by actively managing disruptions to the system caused by anticipated and unanticipated events that impact the available capacity and/or the demand to use individual facilities. Presently, however, no comprehensive methods are available to quantify the existing TMC deployments.

The proposed research will develop a TMC performance evaluation system that addresses this problem. The proposed system will allow TMC managers to evaluate the performance of various bundles of TMC technologies and operational policies by mapping their effects onto events in the system that can be measured using existing surveillance systems and daily activity logs. The resulting tool will provide managers with the long needed capabilities to:

justify valuable technology, personnel allocations, and maintenance costs,
identify technologies that aren't meeting their initial promise, and
identify gaps in current operational strategies that might be filled with new technology deployments.

This evaluation method used will primarily consider delay savings that are attributable to specific TMC actions, but will also seek to quantify benefits on other dimensions such as safety. All computations will be based on direct measurement of demand and capacity in the system using available sensors and will not rely on speculative simulation models requiring extensive assumptions.

The system will be implemented using a modular software architecture that can interface with a variety of existing or planned systems used by Caltrans. The initial deployment will make the evaluation system available to TMC analysts as a reporting system accessible from the Testbed Website. Possible future deployments could integrate the core performance evaluation models with real-time TMC management systems to assist operators with prioritizing response.

An Activity-Based Assessment of the Potential Impacts of Plug-ln Hybrid Electric Vehicles (PHEVs) on Energy and Emissions Using One-Day Travel Data
Investigator: Will Recker
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

With the success of Hybrid Electric Vehicles (REVs) in the automobile market, Plug-In Hybrid Electric Vehicles (PHEVs) are emerging as the next evolution of this attractive alternative. PHEV market penetration is expected to lead to lower gasoline consumption and less emission. The main objective of this project is to assess PHEVs' emission reduction and energy profile impacts based on simulation of vehicles used in activity and travel patterns drawn from the 2000-2001 California Statewide Household Travel Survey. Simulations replicating reported continuous one day data will be used to generate realistic emissions and energy impact assessment of PREV market penetration. A secondary objective is to estimate the decreased gasoline consumption and increased electricity demand in California. This will involve testing various electric pricing strategies designed to mitigate the recharging demands placed on the grid during periods of peak consumption. This assessment of emission and energy requirements of PHEV s will provide a benchmark that will assist in determining the impacts of future PHEV penetration into the automobile market. Specifically, the study will provide an upper bound on the potential demand on the existing grid, as well as categorize expected energy and emissions impacts by time of day and source.

Assessing the Impacts of Truck Traffic on Residential Property Values: A Southern California Case Study.
Investigator: Wei Li
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

With a 70% share of the nation's freight pool, trucking is the dominant mode of freight transportation today. Assessing the external effects associated with truck traffic has become increasingly important in recent years as policy makers, transportation planners and scholars have grappled with increasing traffic congestion, pollution and budget constraints confronting the nation's major metropolitan areas. For my dissertation, I will use hedonic models to investigate whether the volume of heavy truck traffic impacts neighboring property values, with an application to both single family and multifamily markets. My study site will be the Alameda Corridor in Los Angeles County .I will combine detailed GIS analysis with fixed-effects and spatial regression models in order to obtain unbiased estimates of model coefficients to test the hypothesis that the volume but also the nature of traffic matter; this latter aspect appears to have been ignored so far in hedonic studies applied to transportation. This work will likely have important implications for U.S. surface transportation policy, as negative externalities linked to trucking need to be quantified to investigate environmental justice issues and make infom1ed decisions about road infrastructure expansion, sound barriers construction, and truck traffic management

Assessment and Development of Commodity Flow, Logistics, and Other Relevant Goods Movement Data Sources to Facilitate Statewide Freight Modeling
Investigators: Stephen G. Ritchie and Amelia C. Regan
Support: California Department of Transportation

Caltrans has recently approached the University of California, Irvine (UCI) Institute of Transportation Studies (ITS) researchers with regard to applying novel data analysis and data mining techniques to develop a freight flow model for the State and to identify gaps in currently available freight flow data.

UCI ITS has resources that can serve Caltrans' needs in this area. First, we have experts in three key areas available to work on this project. These are freight transportation modeling, the interactions between housing, transportation and warehousing and data mining.

California ATMS Testbed Program: Phase V
Investigator: Will Recker
Support: California Department of Transportation

Research under this agreement focuses on linking existing resources available in the Advanced Transportation Management Systems (ATMS) Testbed into an open platform for the ongoing development, testing, and deployment of new traffic management systems centered on data-driven information processing and policy formation. By integrating existing ATMS Testbed assets, we intend to create a complete end-to-end ATMS Development Environment that can:

Support research and development and the evaluation of new advances in detection, information processing, and ultimately system-wide transportation management.
Explore potential transformations to Transportation Management Center (TMC) processes made possible by new advances.
Test those transformations in a risk free manner.

This generation of the ATMS Testbed is organized around four major task areas:

ATMS Development Environment—Supporting Development of the Next Generation TMC: Develop and maintain an implementation environment that enables independent testing and evaluation of ATMS products and technologies prior to full deployment.
Testbed Services and Technical Support: Research expertise and technical services to sustain operational capability of the ATMS Testbed infrastructure including the Transportation Management Center (TMC) Simulation Laboratory and associated traffic data communication network.
Testbed Integration, Promotion, and Maintenance of Berkeley Highway Labs (BHL): Manage and maintain the integrated ATMS Testbed-BHL facilities. Provide to researchers and practitioners the means to access and utilize the combined transportation research resources of the facilities through the Testbed website.
ATMS Testbed Management: University of California, Irvine, through its Principal Investigator (PI) will continue to be responsible for the overall management of ATMS Testbed operations. Acting on behalf of Caltrans Division of Research and Innovation (DRI), and in close coordination with DRI Testbed Manager, UCI will manage and maintain the overall day-to-day operations of the ATMS Testbed laboratories.

Can Technology Make US Greener Drivers? – An Investigation of the Potential Benefits of ACC and IVC
Investigators: Wenlong Jin and Jean-Daniel Saphores
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

The transportation sector accounts for nearly one third of the US greenhouse gas (GHG) emissions (Green and Shafer, 2003), and until recently, its emissions have kept on increasing. Transportation is also producing a large percentage of local pollutants such as PM, NO2, or CO. Much discussed solutions to address this problem include raising efficiency standards, blending low-carbon fuels with gasoline, or changing land-use patterns through urban design and planning. Another possibility, which has not attracted much attention so far, is to rely on new technologies such as adaptive cruise control (ACC) and inter-vehicle communications (IVC) to improve traffic flow and vehicle fuel economy while reducing the emissions of various pollutants. The purpose of this project is to address this knowledge gap. Using micro-simulation, we will investigate the impact on highway capacity and vehicle emissions when a subset of vehicles are equipped with ACC and IVC technologies, with an application to a section of the I- 80. We will then develop driving strategies to maximize energy efficiency of vehicles equipped with IVC and ACC. Quantifying the environmental benefits of these new technologies is important to understand if new technologies like ACC and IVC could contribute significantly to greening transportation.

Conceptual and Methodological Development of a California Statewide Freight Demand Model
Investigator: Stephen G. Ritchie
Support: California Department of Transportation

This proposal defines a work plan for an initial preparation phase to scope the conceptual and methodological development of a California statewide freight demand model. The objective of this initial phase is to gather all the necessary data and information, reach out to the appropriate stakeholders, and recommend a modeling framework for implementation.

A background is first presented to put the motivation and recent work in context. A broad set of objectives is laid out as a roadmap to drive the scope and tasks. Following that, a list of tasks is proposed along with an estimated time line and budget.

Data Delivery and Routing in Vehicular Networks for ITS Application
Investigator: Rex Chen
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

Communication in vehicular networks is an important component of Intelligent Transportation Systems (ITS). A network-centric vehicle has the ability to deliver accurate and real-time transportation information. Such communications capabilities provide an opportunity to address fundamental problems related to transportation safety and efficiency as well as allowing us to tackle issues such as the feasibility of dynamic ride sharing. Dedicated short range communication sets the roadmap for an inter-connected vehicular network and is an enabler of inter-vehicle communications. This dissertation research addresses an underlying problem in vehicular networks, specifically the study and development of routing protocols for vehicle communications. We take a holistic view - placing the problem in the context of both transportation (traffic flow dynamics) and communication networks (routing algorithms) to design network routing protocols to meet the needs of emerging ITS applications. We believe this approach will translate to a message relay protocol that is optimal and scalable to meet the needs of road networks that are time sensitive coupled with high vehicle mobility. Furthermore, this research in network protocol design can facilitate transportation policies in Vehicle Infrastructure Integration, with recommendations on optimal locations to deploy road-side stations as part of the ITS architecture for the future.

Deployment of a Tool for Measuring Freeway Safety Performance
Investigator: Will Recker
Support: California Department of Transportation

Inductive loop detectors (ILD) are the current state of the practice for monitoring traffic flow on freeways. ILDs are simple and robust, but the available data outputs (30-second volume and occupancy) have limited application. With the completion of PATH TO 5307, we have demonstrated that the data provided by simple loop detectors are sufficient to characterize valuable information beyond the usual volume, speed and density calculations.

The PATH TO 5307 research was undertaken to develop a tool for assessing the impacts of changes in freeway traffic flow on the level of traffic safety .Safety here is measured in terms of the probability of a reportable accident. The results were as follows:

A monitoring tool: the developed tool uses real-time 30-second data from single inductive loop detectors. The data are processed to produce continuous measurements of the odds of different types of accidents for all loop stations providing data. These measurements are then converted into ranges that can be displayed in terms of the familiar red-yellow-green markings that are used for levels of service-
A planning tool: the developed tool can be used to compare the safety level of changes in traffic flow over time.

Our results lead us to conclude that the analytical engine that drives the safety tool is based upon models that are highly effective in identifying those myriad aspects of traffic flow that are statistically related to accident probabilities. In short, loop detectors can be used for much more than estimates of speed, flow, and density.

However, owing to the highly sensitive nature of safety analysis, a follow-up study is essential. This continuation study has four primary goals:

Validate the models obtained in PATH TO 5307;
Seek ways to improve the accuracy of the safety predictions ( e.g., by adding geometric design factors as an input to our accident prediction models);
Extend the applicability of the modeling approach to other Caltrans Districts; and
Suggest appropriate uses, audiences, and modes for disseminating the safety predictions in real time

The most visible project outcome will be the deployment of a "PeMS-like" safety performance system based solely on loop detector data, together with real-time TMC maps displaying relative likelihood of traffic-related safety hazards.

Dynamic Traffic Management under Advanced Path-based Data Collection and Public-Private Multiple-Provider Information Supply and Guidance.
Investigator: Inchul Yang
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

A multi-agent Advanced Traffic Management and Information Systems (ATMIS) framework is proposed. Multiple private traffic information service providers cooperate or compete with each other in the A TIS market which most existing studies look at as a one-supplier system, mostly run by a public agency. Subscribers to private information services become probe vehicles and report traffic data as well as travel diaries {individual path-based data) to the private vendors. A Traffic Management Center (TMC), i.e., public agency optimizes network-wide traffic signal controls, for which a dynamic traffic optimization model {or algorithm) is developed in an integrated fashion at the network level on the basis of path- based data with traffic information supply and route guidance. In the framework, the private companies provide the path-based data used in the traffic signal optimization model to the public agency and in return get the optimized traffic signal control plans which are essential in predicting future traffic conditions. The study is simulation-based, as the path-based data collection systems are only expected in the near-future. The study results are expected to develop insights on the benefits from such cooperative public-private operations and provide the impetus for private and public investment in the necessary data collection infrastructure.

Generating a Database GIS Tool to Provide Roadway Link Vehicle Activity
Investigator: Will Recker
Support: California Department of Transportation and EPA Air Resources Board

In response to a solicitation by the California Air Resources Board (ARB), the UC Irvine Institute of Transportation Studies (UCI ITS) proposes the development of a tool that will help ARB improve the vehicle miles traveled (VMT) and speed estimates used in EMFAC for both trucks and passenger vehicles. We will integrate three important data sources: California's Weigh-In-Motion (WIM) system, the statewide Highway Performance Measurement System (PeMS), and the Highway Performance Monitoring System (HPMS). The three data sets will be integrated into a single, state-wide roadway link vehicle activity database (VAD). It is expected that the V AD will continue to expand to include other data sources and continue to be useful to the California ARB and to Caltrans in the years to come.

An important result of this project will be the development of a GIS tool that can display the merged data in a GIS application. This proposal presents a staged development of such a GIS application. First, each of the three data sets will be visible on their own in a GIS, without an explicit merging of the three sources. The second step is to create a "best source" layer, in which the three layers are filtered into an aggregate layer based on relative quality. The third step is to refine the integration of the three layers, including adjusting the source data based on observations from "better" data. So, for example, in this final integrated layer the HPMS average values would match up exactly with corresponding PeMS observations. The staged development that we propose will allow the ARB to have an input in the development of the GIS tool, as well as in the development of the data integration techniques.

The solicitation requested that the stock PeMS methods for inferring missing data be replaced. Our approach to this will be to focus on the merged data layer, rather than the PeMS data. We will use the raw data from PeMS, without any inference or imputation, as input to the merged data layer. We will include as part of the merging process notions of statistical confidence, and will simply drop data locations that do not meet minimum confidence bounds. Other data sources, including upstream and downstream PeMS detectors, will be used to "cover" these missing detector stations in the merged layer.

This proposal's emphasis is on building data synthesis and analysis tools. At the end of the project, the ARB will have the resources it needs to establish and maintain a merged data layer, including the ability to update the layer each year as new data becomes available. The ARB will also have the option to hire a third party to manage the data integration process, if they so desire.

This proposal is organized around the discrete tasks that build toward the final product. The ordering is not exactly chronological, as some of the subtasks overlap other tasks. A chronological ordering of the subtasks with target completion dates is given in the management plan. The tasks are numbered sequentially, with dot notation indicating subtasks.

High Occupancy Vehicle(HOV) System Analysis Tools: Statewide HOV Facility Performance Analysis
Investigator: Will Recker
Support: California Department of Transportation

Despite wide adoption of HOV facilities by many states, Metropolitan Planning Organizations (MPOs) and cities, there still remain questions on the effectiveness of HOV systems. Among relevant factors that may impact the operation and safety performance of HOV facilities, operation policy and HOV -access configurations are being actively evaluated by Caltrans. One idea under investigation to improve HOV lane operation is to reconfigure the HOV lane system from Full-time Buffer-separated to a Full/part-time continuous-access facility. Furthermore, a recent study aimed at evaluating safety performance of California freeways equipped with HOV facilities led to an interesting conclusion-notably, the safety performances of the buffer-separated, limited-access HOV lanes typically seen in Southern California, when compared to those continuous-access facilities, limited-hour operation of HOV in Northern California, appear to offer no safety advantages. This is contrary to the common belief that buffer separation and restricted entrance and exits will provide additional protection for traffic moving in the HOV lanes. As a result, whether to support the idea to reconfigure the HOV lane system from Full-time Buffer-separated to a Ful1/part-time Continuous-access facility depends on whether or not the continuous-access HOV facility can provide better operational benefits than the buffer-separate HOV facility. This proposal is designed to address this issue by providing a methodology based on combined real-world data evaluation and simulation analysis to derive further understanding and insights for future policy setting and operational guidelines for HOV facilities, with the ultimate goal of achieving efficient and safe highways.

Incorporating Vehicular Emissions into an Efficient Mesoscopic Traffic Model – An Application to Southern California
Investigators: Jean-Daniel Saphores, Wenlong Jin, and Stephen G. Ritchie
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

With the increasing public understanding of the health impacts of air pollution from motor vehicle operation, policy makers need better tools for analyzing the effectiveness of regional policies designed to reduce air pollution. Indeed, the four-step process, which is the current state-of-the-practice in transportation modeling, is inadequate for this task, and micro-simulation, while promising, is very time consuming. We therefore propose to incorporate vehicular emissions into a dynamic mesoscopic traffic model to create a tool that will generate information about traffic flow, vehicle speeds, and emissions of CO2 and various criteria pollutants (PM1O, NO2, and CO) on large scale networks. This tool will help evaluate the effectiveness of various policies, such as mandating a certain concentration of biofuels in gasoline or diesel, promoting hybrid electric vehicles (cars or trucks), or replacing older trucks with cleaner ones as is currently done for the San Pedro Bay Ports drayage trucks. In addition, our application to the road network linking the SPBP to the Inland Empire via downtown Los Angeles will provide new insights into various policies that could relieve the dual burdens of congestion and air pollution from communities affected by freight transportation to and from the SPBP .

Large Scale Real Options Models for Network Investment Planning and Operational Risk Hedging
Investigators: Amelia Regan and Joseph Chow
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

A real option model for investments made in a network is considered as a method for addressing managerial flexibility in transportation planning. A continuous network investment deferment model is formulated with longitudinal stochastic OD flows and each payoff is determined by the continuous network design problem. It is a bilevel program with an upper level Bellman equation for dynamic programming and a lower level based on the continuous network design investment allocation and user-optimal route choice. Each OD demand flow evolves as an independent, discretized geometric Brownian motion. A heuristic approach based on Monte Carlo simulation and Iterative Optimization Assignment is considered, using a sampling strategy to overcome it inherent computational inefficiency. Th e option value is decomposed in to the basic deferment option and a newly def in ed network option. Network exposure is expanded as an application to operational risk hedging to consider the impact of failed links on the expanded net present value. A solution for a Sioux Falls, SD network example with zero drift is compared to the stochastic demand scenarios in earlier literature as well as the standard exposure with the investment exposure.

Mitigating the Air Quality Impacts of San Pedro Bay Port Drayage Truck Operations on Arterials
Investigators: Stephen G. Ritchie, R. Jayakrishnan, and Jean-Daniel Saphores
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

The San Pedro Bay Ports (SPBP) of Los Angeles and Long Beach in Southern California comprise one of the largest and busiest container port complexes in the world, and contribute significantly to both our regional and national economies. However, the SPBP complex is a major contributor to air pollution. Local communities adjacent to the ports and associated intermodal facilities are believed to bear a disproportionate burden of the health impacts (respiratory and cardiovascular illness, and premature death) associated with this pollution. A particular concern is emissions from heavy duty diesel vehicles (HDDVs), or drayage trucks, which form an unusually large proportion of vehicles on nearby freeways and arterials and which carry port cargo (mostly containers) to and from the ports and intermodal rail and transloading facilities. Our existing UCTC research has developed a new modeling approach to evaluate the health and environmental justice impacts of various SPBP freight corridor operations and pollution mitigation strategies. However, our modeling of HDDVs only addresses the freeway network. This research will therefore complete a major missing link by modeling HDDV drayage truck operations on arterial roads, thereby providing new and more accurate insights into the emission impacts of alternative multimodal SPBP freight

Mitigating the Social and Environmental Impacts of Multimodal Freight Corridor Operations at Southern California Ports
Investigators: Stephen G. Ritchie, Jean-Daniel Saphores, R. Jayakrishnan, Oladele Ogunseitan, Amelia Regan, and Rudolph Torres
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

The San Pedro Bay Ports (SPBP) of Los Angeles and Long Beach in Southern California are one of the major container port complexes in the world: in 2004, for example, the SPBP processed over 36% of the U.S. container trade. However, the SPBP complex is also a major source of air pollution caused largely, on the land-side, by diesel locomotives and trucks that transport containers to and from the ports. The resulting annual health costs may exceed $2.5 billion. Low income and minority communities along the major Alameda corridor, a 20-mile railroad line that connects the SPBP to the transcontinental rail network east of downtown Los Angeles, are particularly affected. This study will create a tool that will quantify links between SPBP freight traffic, air pollution, and the health of local communities. This tool will help evaluate the effectiveness of various alternatives (such as congestion pricing to decrease peak container traffic flows, biofuels for trucks and locomotives, or intermodal and route shifting of container traffic) in order to mitigate the environmental and health impacts of SPBP activities. Expected results include new insights into the spatial, socioeconomic, public health, and social justice consequences of alternative SPBP multimodal freight operations strategies.

Multicampus Research Programs and Initiatives
Co-Directed by: Stephen G. Ritchie, director of the UC Irvine Institute of Transportation Studies, Samer Madanat, director of the UC Berkeley Institute of Transportation Studies, and Daniel Sperling, director of the UC Davis Institute of Transportation Studies.

The University of California (UC) Institute of Transportation Studies (ITS) Multi-campus Research Unit (MRU), with branches at Berkeley, Davis and Irvine, and with research collaborators at Los Angeles, Riverside and Santa Barbara, has received a major grant to create a new Program for Sustainable Transport. It has three research tracks: Technology, Mobility and Infrastructure.

The new program will be supported by an initial 5-year award of $6.25M from the UC Office of the President. It was selected under the University’s Multicampus Research Programs and Initiatives (MRPI) program through a competitive peer-review process. The ITS program will be co-directed by Professors Samer Madanat at UC Berkeley, Dan Sperling at UC Davis, and Stephen Ritchie at UC Irvine, with administrative headquarters at UC Irvine.

UC issued an MRPI Request for Proposals in January 2009 that was directed toward existing MRU’s and new MRPI’s proposed by campus groups. UC anticipated approximately $12M per year total would be available for the MRPI program, and that those funds would replace, not supplement, current MRU funding. The MRPI goals are to facilitate and support outstanding research and cutting edge discoveries that advance a field, impact the lives of Californians, increase UC’s competitiveness, inform public policy, and support innovative graduate student research. UC recently announced 28 MRPI awards under the new program.

Program funding to ITS will support cutting-edge research targeted at key societal needs. It will support key state initiatives unfolding over the next few years, as well as the training of the next generation of experts and leaders. It will bring researchers together from more than 30 disciplines on six campuses to seed multi-disciplinary initiatives, including collaborations between economists, geographers, ecologists, city and regional planners, public policy analysts, engineers from civil, environmental, electrical and mechanical engineering, computer scientists and experts in energy.

The theme of our successful proposal was sustainable transportation, with a focus on three integrated activities: 1) vehicles and fuels, 2) infrastructure investment and system management, and 3) land use and mobility planning. The objective is to leverage the substantial capabilities of ITS to design tools, policies, and programs to reduce congestion, oil use, local air pollution, and greenhouse gas emissions, in ways that contribute to economic growth and social well being.

Online Freeway Corridor Deployment of Anonymous Vehicle Tracking for Real-Time Traffic Performance Measurement
Investigator: Stephen G. Ritchie
Support: California Department of Transportation

Effective real-time traffic detection systems that meet operational needs are a prerequisite for obtaining accurate, detailed and timely freeway performance measures that are critical for efficient transportation system operations. Real-time performance measures provide the basis for optimal operation of the freeway network and an in-depth understanding of traffic impacts in congestion-prone metropolitan areas, the knowledge of which is essential to determine the best mitigating solutions. It is therefore imperative that Caltrans obtain experience with different detection and performance measurement systems in actual operating environments, as proposed here, in order to most effectively comply with state law regarding the efficacy of ongoing General Obligation bond fund expenditures and, more generally, to obtain the greatest benefit at least cost for California travelers on the state highway system.

Policies and Practices Related to Passenger Vehicle Greenhouse Gas Reduction: Evidence and Assessment
Investigator: Marlon Boarnet
Support: State of California - EPA Air Resources Board

Senate Bill 375 (SB 375, Steinberg, Chapter 728, Statutes of 2008) requires ARB to set greenhouse gas emission reduction targets for passenger vehicles and light trucks for 2020 and 2035. As part of the target setting process, SB 375 requires ARB to appoint a Regional Target Advisory Committee (RTAC), which will provide a recommendation to ARB on the factors and methods for consideration in target setting by September 30, 2009. In its draft recommendation, the RTAC recommends that ARB work with technical experts and practitioners over the next 4-6 months to develop and peer-review a list of land use and transportation policies and practices for reducing regional passenger vehicle greenhouse gas emissions, with opportunity for public input. ARB will begin the public process to develop this list in Fall 2009. The initial list will likely be extensive and include policies and practices for which empirical data may or may not currently exist. The RT AC also recommends that this list of policies and practices be incorporated into an analytical spreadsheet tool that could assess what greenhouse gas reductions may be possible by implementing some or all of the policies and practices identified in the list. The tool would assist in both near-term target setting and longer- term local planning and implementation.

In response to these recommendations, a team of researchers from the University of California proposes to provide technical support to ARB during its development and refinement of the list of policies and practices. Specifically, the UC researchers propose to: (1) identify which policies and practices on the list have supporting empirical data and evidence, (2) review and assess the evidence to provide conclusions on the degree of effect and other pieces of information (identified under tasks), and (3) provide easily understandable summaries of the review and conclusions, including degree of effect in reducing greenhouse gas emissions through VMT reduction.

Price and Frequency Competition in Freight Transportation
Investigators: Jan K. Brueckner and Nilopa Shah
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

Service frequency is an important quality dimension in passenger transportation, with frequent service allowing a good match between passengers' actual and preferred departure times. Frequency is thus a key dimension of competition between passenger carriers, along with the fare charged. Frequency is also a key quality determinant in freight transportation, since low frequencies require cargo to accumulate while awaiting shipment, imposing inventory costs on fi rm s. Frequent service, by contrast, allows inventory stocks to be kept low, reducing costs. While this trade-off has become more prominent in recent years with the advent of "just- in-time" production methods, the economics-oriented transportation literature lacks a conceptual framework for analyzing the provision of high-frequency freight service by competitive suppliers. The purpose of the proposed research is fill this gap by developing a theoretical model of frequency and price competition among the providers of freight services, paralleling the author's previous analysis of such competition in passenger services. While the two models will have a similar structure, key details differ given that the demand for freight services arises from firm profit maximization rather than consumer utility maximization. The research will provide a new conceptual framework for the analysis of competition in freight transportation.

Quantifying Effect of Local Government Actions on Vehicle Miles Traveled (VMT)
Investigator: Marlon G. Boarnet
Support: University of California, Davis (ARB Subaward)

California’s regions will soon be required to reduce transport sector greenhouse gas emissions to target levels to comply with SB375 requirements mainly through reducing vehicle miles traveled (VMT). In order for local and regional governments to properly prioritize actions; it is necessary to know how much each action would actually reduce VMT. The challenge is that the particulars of the local and regional context play a large role in determining which actions will be most effective where, but existing research provides little evidence on the role of context.

We posit that the extent to which people will reduce their VMT in response to local policies is a function of contextual variables, and it is this function that we propose to estimate m order to help local and regional governments prioritize actions. To do so, we will merge six large travel survey datasets from California with census information, transport network GIS layers, and land use GIS layers, creating a single large dataset that incorporates high variation along land use types, urbanization level, household sociodemographics, and travel type, covering the dominant dimensions that are relevant for VMT -reducing policy in the State of California. Using this dataset, we will estimate statistical models that explain VMT While controlling for "self-selection bias" in residential location choice. We will then use these models to estimate how elastioities and purposes, household types, and local contexts.

This information will be useful to -local governments as they decide which policies to implement today. As part of this project, we will develop a simple spreadsheet tool that allows local policymakers to immediately access the portion of our findings that is relevant to their particular community. This tool will enable more effective VMT-reduction policies to be implemented sooner at the local and regional levels. In addition, our findings will provide improved parameter estimates for use in land transportation scenario planning models, allowing these tools to more accurately predict the behavioral ro8ponse to policy scenarios.

Spatially Focused Travel Survey Data Collection and Analysis: Closing Data Gaps for SB 375 Implementation
Investigator: Marlon G. Boarnet
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

This research will help close a crucial data gap in land use - travel behavior studies. Current estimates of land use - travel behavior relationships are typically based on average effects for metropolitan areas or larger geographies. That gives little insight into the effect of small-area land use policies such as targeted infill development, transit-oriented land uses near stations, or similarly localized policies. In California, Senate Bill (SB) 375 requires that metropolitan planning organizations incorporate land use - transportation planning, but existing travel diary surveys have very few observations in areas of policy interest. This research will obtain a large number of travel diary surveys in small neighborhoods of high policy relevance for SB 375, providing data that will assist in SB 375 and related policy development. In addition to data and analysis that will directly benefit greenhouse gas emission reduction policy, the methods developed in this research will advance efforts toward low-cost, rapid travel data collection that can be used in before-and-after transportation program evaluations.

The Development of Web-based Training Modules for The Planning and Operational Analysis of Public Transit Systems
Investigator: Michael G. McNally
Support: California Department of Transportation

The Institute of Transportation Studies at UC Irvine has been developing a web-based training program for Caltrans and local agency transportation engineering and planning staff to provide a foundation for transportation analysis skills. Initial course module development has been directed toward micro simulation modeling, travel forecasting (Four Step Models), and systems analysis. Proposed herein is an extension of this web-based training to include course modules to address current and emerging modeling and analysis for technical assessment, travel forecasting, and operational analysis of public transit systems and alternative transportation modes. These course modules will be developed in the same format as current modules and introduced as courses in the Sakai-based web-site hosted at ITS Irvine.

Training Program for Corridor Systems Management
Investigator: Will Recker
Support: California Department of Transportation

In support of the Corridor System Management Plan (CSMP) process, Caltrans has identified a need for assistance in developing a program to move the state of the art in the application of planning and simulation tools to model transportation corridors into the state of the practice. The demands placed on commercially available modeling packages by the CSMP process are at or beyond the cutting edge of what are possible today.

The University of California Irvine Institute of Transportation Studies (UCI ITS) has resources that can serve Caltrans' needs in this area. Under a separate agreement, we are in the process of developing an online training website encompassing a broad range of topics in transportation planning and engineering. This website will eventually be open to Caltrans, MPO modeling staff, and UCI students. It will serve to connect academic material about transportation planning, engineering and modeling with the real-world experiences of practicing engineers and planners. Although the site is currently in its early stages, we envision the site facilitating a three-way dialog among UCI transportation lecturers, enrolled students in transportation programs, and Caltrans staff. Although the needs of the CSMP process go beyond what is available in the online website, the training website can serve to add significant value to any vendor-based training.

In this effort, we propose organizing, purchasing, and hosting vendor-specific training sessions to demonstrate the latest techniques available in commercial modeling packages for which Caltrans has licenses for solving the challenges of the CSMP process; we expect that provision of these classes will be particularly valuable to Caltrans District staff who are charged with CSMP projects in their respective districts. In order to minimize travel costs of Caltrans participants, we plan to distribute the vendor sessions roughly equally between the north (Sacramento) and south (Irvine) California regions At UCI ITS we have a complete traffic simulation training facility on-site--one that has been used successfully and extensively both by software vendors to hold training sessions in micro simulation as well as by Caltrans for "hands- on" training of TMC operators. Our understanding is that suitable training sites are also available in the Sacramento area (e.g., at UC Davis). These vendor-based sessions would be augmented as needed by practical, real-world applications of the different models to case studies within the California region, with emphasis on Orange County due to our extensive experience modeling this area. These practical sessions would be developed by UCI researchers-not by the vendors-and as such would be available to all Caltrans and MPO staff on our online training website. The training website would serve the participants in the training sessions both before and after the vendor training. Because the number of participants that can participate in any training session is small ( especially when compared to the numbers of planners and engineers in Caltrans' staff), we see the online training website as an ideal way to get as much value as possible out of the training sessions.

UCI-ITS Transportation Courseware: Web-based training for Caltrans Personnel
Investigators: Will Recker, Michael G. McNally, Lianyu Chu, Craig R. Rindt, James E. Marca
Support: California Department of Transportation

Caltrans has recently approached the University of California, Irvine (UCI) Institute of Transportation Studies (ITS) researchers with regard to setting up on-line training for their staff. The immediate problem faced by Caltrans is a need to increase staff capabilities in the area of microsimulation. In addition, Caltrans has a similar need for unifo rm training and a repository of best practices in all areas transportation engineering.

UCI ITS has resources that can serve Caltrans' needs in this area. First, we have recently developed courses in microsimulation, and can post those lectures on-line. Second, UCI ITS has past experience with on-line learning management systems (LMS). Third, UCI ITS has an active interest in setting up an on-line teaching web site for use with our graduate and undergraduate students.

Variability of Travel Behavior of Harbor Community Residents
Investigator: Douglas Houston
Support: U.S. Department of Transportation and California Department of Transportation/University of California Transportation Center

The Harbor Communities Time Location Study (funded by a 2007-08 UCTC grant, PI Paul Ong) demonstrated innovative methods for using GPS-enhanced travel and activity monitoring to measure, analyze, and verify highly resolved travel patterns of low-income households for multiple modes and days. We monitored 10-14 days of activities and locations of 51 adult residents (21-65 years old) of the Wilmington area of the City of Los Angeles, California and the western portion of the City of Long Beach. California in early 2008, and have completed initial analysis of travel patterns for the 2-3 days for which both self-reported time-activity logs and passive location tracking with a portable GPS device. We found that participants did not report about half of the location/travel identified in the GPS-enhanced data. The current research will refine methods to process/classify the remaining 7-11 days monitored through GPS tracking only in order to analyze the variability of travel patterns of low-income households in a major goods movement corridor.

Visual PFE Software Enhancements for Planning Applications in Small-Sized Communities
Investigator: Will Recker
Support: California Department of Transportation

The goal of this project is to enhance existing Visual PFE (Development of a Path Flow Estimator for Deriving Steady-State and Time-Dependent Origin-Destination Trip Tables) and adapt it to perform planning applications in small-sized communities ( especially in rural areas) where the limited planning resource hinder the development and applications of 4-step models.

Visual PFE, and PFE- TD are software packages for estimating travel demand in a general urban network developed in PATH TO 4135 and PATH TO 5502, respectively, where PFE refers to Path Flow Estimator, the approach used in estimating path flows and hence O-D trip rates in these two projects. Visual PFE provides a graphical user interface (OUI) for and implementation of a static version of PFE and a few other static O-D estimators, while PFE- m implements a dynamic, or time-dependent (TD) version of PFE with a OUI transplanted from Visual PFE. In TO 4135, it was expected to:

Implement and extend Bell and Shield's original PFE model to obtain planning level O-D trip tables for general networks
Develop other static O-D estimation algorithms (namely bi-level type of algorithms) for comparison with PFE
Develop a graphical user interface, Visual PFE, to facilitate the use of the developed algorithms
Test and evaluate the developed algorithms using a real world example.