Edward Fok from FHWA at ITS

This coming Friday, March 6 there will be a seminar at the ITS Seminar room from 9:30 to 10:30 am featuring Edward Fok, from the Federal Highway Administration.

After the seminar, ITS students would be able to meet Edward Fok at the Student Conference room.

Drinking from the Advanced Transportation Firehose

This will be an overview of current development in urban transportation management technologies. I will describe some of the work begin done in predictive real-time operation, describe the goals of connected vehicle, examine some of connected vehicle’s impact of transportation operation, clearing the air between automated and autonomous vehicles, and discuss some possible impact of automation on transportation management. If I have time, I’ll also touch on some of the new challenges advanced transportation management systems are facing and could encounter in the future. The goal of this talk is to stimulate discussion and ideas where additional research will help.

Edward Fok is a Transportation Technology Specialist with the Federal Highway Administration (FHWA) Office of Technical Service/Resource Center. He helps public agencies apply advanced transportation systems and processes to solve mobility problems. He also helps researchers at Turner-Fairbanks and the Joint Program Office advance the state of the art in transportation operations. Ed is very active in many technical areas including Integrate Corridor Management, Connected Vehicles, Cyber Security, Automated Vehicles, and Advanced Freight Systems. Ed came to FHWA from the City of Los Angeles with 11 years of operations and research experiences and holds multiple professional engineering licenses.

Prof. Hani Mahmassani at ITS

On March 2nd Prof. Hani Mahamassani gave a seminar on:

Autonomous vehicles: Adoption Rates and Flow implications in mixed traffic streams.

We present a general conceptual framework to explore autonomous vehicle adoption. The traffic flow implications of different adoption rates are examined using a microscopic modeling framework of mixed traffic streams in which certain fractions of the vehicles are respectively autonomous, connected or both. We jointly model the properties of the peer-to- peer communication systems for different levels of message content. The framework is used in an exploratory analysis of the flow characteristics of the resulting mixed traffic stream, with particular attention to throughput and stability.

Professor Hani S. Mahmassani is the William A. Patterson Distinguished Chair in Transportation; Director, Northwestern University Transportation Center; Professor, Civil and Environmental Engineering, McCormick School of Engineering and Applied Science; and Professor (courtesy), Managerial Economics and Decision Sciences, Kellogg School of Management. Professor Mahmassani specializes in multimodal transportation systems analysis, planning and operations, dynamic network modeling and optimization, transit network planning and design, dynamics of user behavior and telematics, telecommunication-transportation interactions, large- scale human infrastructure systems, and real-time operation of logistics and distribution systems.

After the talk, ITS students had the opportunity to meet Prof. Hani Mahamassani, and from the ITS Graduate Student Association, we would like to thank him for this opportunity and the insightful discussion we had.

WEEK 5 – ITS Student Seminar

Travel patterns change and effectiveness of TDM in California, by Sungsu Yoon

The purpose of this research is to improve current travel demand model with up-to-date inputs and to contribute enhancement of needed sub-modules to forecast regional travel demands more accurately. Core structure of this research is deriving update Origin-Destination (OD) matrix for traditional regional four-steps travel demand model and then compare scenario results with major transportation demand measurements to see the impacts of inputs and update. Alternative scenario will have update OD matrix from the enhanced mode choice data based on the analysis of the latest or current travel behavior changes in the region. And other sources such as years of Household Travel Survey and employment based travel demand control programs also will be analyzed and be utilized by the needs. As an introduction, research will review two different travel surveys between 2001 and 2011 to see whether there is any travel mode pattern change among different survey years. Based on travel behavior changes (mode choice pattern changes and trip length, vehicle occupancy) from different demographic generations, there should be an updates on travel demand parameters and assumptions regarding travel demand modeling inputs and ratio of parameters. CHTS analysis will be direct source of the socioeconomic (SED) analysis but additional SED forecasts from SCAG would be supplemental resource to derive appropriate trip generation and distribution ratios in the demand modeling. Based on the update mode choice pattern and rate, research will derive update mode choice results by major modes (drive alone and carpools, active transportation) and OD matrix among research area by designated travel analysis zone (TAZ). In addition, to execute the latest travel mode share analysis, research will upgrade network systems which also include alternative active transportation systems such as biking or walking as needed. Regarding methodology, research will utilize previous modules and theoretical methods from well-established programs like as the TRIMMS (Trip Reduction Impacts for Mobility Management Strategies) and EPA’s COMMUTER model and other Travel Demand Management (TDM) measures. Analyzing and measuring impacts of specific TDM strategies and programs are important part of this research to derive update ratio of mode shares in each TAZ. After mode choice and OD matrix results are update, research will use those update OD results as seed matrices for assignment stage. And then once assignment results are ready, we could analyze impacts from travel pattern changes through comparing results of baseline case (current travel demand model result) and update scenario result which was developed from the latest trend of travel pattern. As a quantifying analysis, research will measure major travel demand factors (VMT, VHT and Delays) and major environmental measures (emission analysis) to know the performance of scenario plan.

Traffic congestion and fragmentation of metropolitan governance, by Gavin Ferguson

Metropolitan transportation planning in the United States is growing more fragmented and decentralized as a result of stagnant federal funding for transportation. How will the increasingly decentralized provision of transportation affect urban transportation systems? Although many argue that inter-jurisdictional conflicts will prevent the kind of coordination needed to address regional problems such as traffic congestion and air pollution, thus necessitating more centralized authority, there have been no econometric studies to support this claim. The present study begins to address this gap in the literature with an analysis of the relationship between decentralization and traffic congestion. A panel data analysis was conducted using data on U.S. metropolitan statistical areas spanning the years 1982 to 2011. The results showed a positive elasticity estimate of annual person-hours of delay per capita with respect to number of city governments per capita of 0.632. This estimate implies that decentralization contributed an extra 37 million hours of delay in 2011 across the 97 MSAs included in the analysis.

Week 4 – ITS Student Seminar

Measuring Perceived Travel Time Uncertainty In Transportation Modeling and Simulation by Gabriel Yu.

Measuring and incorporating perceived travel time uncertainty in transportation modelings is a desired topic in recent TRB research needs, which is especially critical in improving DTA, ABM, project(s) evaluation, and understanding better of travel behavior. This short speak will talk about motivations, existing methods, and the proposed methods in the efforts of measuring perceived travel time uncertainty. Sensation, category-based perception, and information will be introduced. The proposed Information entropy measurement will be followed by a simple application on static path-based traffic assignment and an extension to quantum cognitive models. 

Proactive vehicle routing with inferred demand to solve the bikesharing rebalancing problem by Robert Regué

Bikesharing suffers from the effects of fluctuating demand that leads to system inefficien- cies. We propose a framework to solve the dynamic bikesharing repositioning problem based on four core models: a demand forecasting model, a station inventory model, a redis- tribution needs model, and a vehicle-routing model. The approach is proactive instead of reactive, as bike repositioning occurs before inefficiencies are observed. The framework is tested using data from the Hubway Bikesharing system. Simulation results indicate that system performance improvements of 7% are achieved reducing the number of empty and full events by 57% and 76%, respectively, during PM peaks. 

Week 3 – ITS Student Seminar

Advocacy in Action: Understanding the Influence of Advocacy Organizations on Local Affordable Housing Policy in the U.S.

Financial support for affordable housing competes with many other municipal priorities. This work seeks to explain the variation in support for affordable housing among U.S. cities with populations of 100,000 or more. Using multivariate statistical analysis, this research investigates political explanations for the level of city expenditures on housing and community expenditures with a particular interest in the influence of housing advocacy organizations (AOs). Data for the model were gathered from secondary sources including the U.S. Census and the National Center for Charitable Statistics. Among other results, the analysis indicates that, on average, the political maturity of AOs has a statistically significant, positive effect on local housing and community development expenditures.

Anaid Yerena

Department of Planning, Policy & Design, University of California, Irvine.

Week 2 – CS Seminar

This week we are encouraging all ITS students to go to the seminar from the 2014 – 2015 CS Seminar Series featuring Dr. Franz J. Brandenburg from the University of Passau.

Date: Jan 30, 2015 from 11 to 12.

Location: Donald Bren Hall (DBH 6011).

Algorithmic Problems for the 2. Mobile Revolution.

Abstract:

There are autonomous or self-driving cars. Such a vehicle is capable of fulfilling the transportation capabilities of a traditional car, using techniques such as radar, lidar, GPS, and computer vision to sense its surroundings. There are prototypes by Google, Audi* and Mercedes* (* presented at CES Las Vegas 2015) and other companies and universities. 

My assumption: In 2030, the 2. Mobile Revolution is there! All cars are autonomous. How shall this work? The new way of transportation merges private, public and commercial transportation. A car comes on demand, you may change cars, you share a car with other passengers, and you’ll be dropped off at your destination.  

In the talk, I’ll present my transportation model. It poses new algorithmic problems, which arise from an aggregation of millions of shortest (cheapest) paths and result in extended versions of shortest paths and network flow problems.  These problems must be solved to make autonomous driving more convenient and cost-effective than nowadays transportation.

Week 1 – Poster Session

Hello everyone,

Tomorrow we will have a poster session from the students who presented a poster at the recent TRB conference. In this way you can quickly learn about what other students are doing.

We will hang the 5 posters at the Student Seminar room, AIRB 4000 and Prof. Jin will join us to present his research.

List of posters presented:

  • Daniel Rodriguez RomanAccounting for Population Exposure to Pollutants in the Toll Design Problem [pdf]
  • Zhe Jared SunSimultaneous State and Parameter Estimation in Newell’s Simplified Kinematic Wave Model with Heterogeneous Data [pdf]
  • Jiangbo Gabriel YuIncorporating Perceived Travel Time Reliability Into Transportation Planning and Simulation Models Using Information Entropy as the Measure [pdf]
  • Robert ReguéUnderstanding Household Priorities when Scheduling Activities [pdf]
  • Harya S. DillonThe Impact of Urban Form and Gasoline Prices on Vehicle Usage: Evidence from the 2009 National Household Travel Survey [pdf]

Welcome to the Winter 2015 ITS Student Seminar

The seminar provides an opportunity for graduate students to present their research and ideas in front of other graduate students. The intent is to foster discussion and share research among us.

We propose the following:

  1. Short Presentation: 10-15 minutes.
  2. Long Presentation: 45 minutes.
  3. Debates

1. Short Presentations

Short presentations are 10 to 15 min talks that can cover any transportation related topic. They are intended to be high level talks about individual research topics. At the end of each presentation there will be plenty of time for discussion, suggestions and comments. The presenter would then moderate the Q&A session and get valuable feedback and the research topic presented.

2. Long Presentations

This are longer presentations that will cover a research topic in more depth. It is intended for students willing to practice a presentation for a coming conference (such as TRB) or dissertation defense. Depending on time constraints and student participation they will be of 30 min or 1hr long.

3. Debates

We want to introduce discussions in the form of debates of transportation topics that affect our daily lives.

The format of the debate will be as follows:

  1. Brief introduction by a neutral moderator
  2. 10 minutes argument for the Pro or Position 1
  3. 10 minutes argument for the Con or Position 2
  4. 5 minutes rebuttal for the Pro or Position 1
  5. 5 minutes rebuttal for the Con or Position 2
  6. Questions from the moderator
  7. Questions from the audience
  8. Audience vote

During the first class we will present the list of possible topics and vote for two topics. After that each student enrolled will be assigned to a debate team and position to defend.

We encourage everyone to participate, it is a great experience to present in front of other students and you always get good suggestions and comments back!

If you would like to present your research, please email us. 

The first seminar is currently scheduled for Friday, January 23, 2015 (Friday after TRB) and it will simply consist of the posters that we presented during TRB (or posters from other conferences).

We look forward to seeing you in the next few weeks.

Best regards,

Robert, Roger, Daniel