GeoTrans Technical Workshop 2015
A workshop sponsored by the University of California Institute of Transportation Studies Multicampus Research Program and Initiative on Sustainable Transport: Technology, Mobility and Infrastructure.
TITLE: Pattern Recognition and Anomaly Detection for Modeling and Simulation in Sustainable Transportation
Workshop Description:
Sustainability policies, such as changes in land use to motivate decrease in travel, and market introduction of new technologies, such as hybrid electric cars or new fuels, have expanded the scope of travel behavior modeling to span numerous scientific fields. This expansion has demonstrated a need for new data and new techniques to perform pattern recognition across scales:
· At the micro-level: e.g. the behavior of persons and their households
· At the meso-level: e.g. traffic streams
· At the macro-level: e.g. activity and movement at the level of cities and regions.
Today we can count on a tremendous amount of data, often referred to as Big Data. Accumulation of Big Data offers unique opportunities to enhance traditional behavioral models that have been limited by the data collected in traditional household questionnaires. New information and new transportation services enable the development of more informative models. For example, by combining data from multiple sources, one can perform a variety of verification and validation tasks on existing data and develop forecasting models for behavioral facets previously neglected. These new opportunities entail risks of misrepresentation and illusionary inferences. Pattern recognition and anomaly detection allow us to differentiate the useful information and misleading results. In the first part of this workshop we will review examples of data fusion and discuss methods and techniques for pattern recognition and anomaly detection.
Transportation modeling and simulation is increasingly aiming at integration of different behavioral facets into internally consistent model systems. Examples include attempts to integrate land use with travel behavior, travel behavior with traffic assignment and route choice, and route choice with traffic simulation. Integration, however, is a general term that is often misused and misunderstood. This workshop will provide guidance in understanding integration of behavioral facets in models. Applications discussed in this workshop will span a wide spectrum. In this workshop, researchers from the University of California together with invited speakers will provide examples of integrated models they have developed. The discussion will include descriptions of policies examined, data used, algorithms and models for pattern development, applications, research gaps, and future work. We will also provide examples from fields outside the traditional domain of transportation. Other topics include modeling and simulation in land use and travel behavior, internet sources for travel analysis, time and space perception, and analyst derived spatial realities.
This technical workshop provides a forum to discuss the state-of-the-art in pattern recognition with potential applications in transportation research. The workshop will take place on Friday, February 27, immediately before the opening reception of the 2015 UCCONNECT Student conference. We invite faculty members, researchers, and graduate students, as well as practitioners from Metropolitan Planning Organizations and the State Department of Transportation (CALTRANS) to participate.
INVITED SPEAKERS INCLUDE:
Professor Mahdieh Allahviranloo - The City College of New York
Professor Richard Church - University of California Santa Barbara
Professor Konstadinos Goulias - University of California Santa Barbara
Professor Krzysztof Janowicz - University of California Santa Barbara
Professor Werner Kuhn - University of California Santa Barbara
Professor Hani Mahmassani - Northwestern University
Professor Alexei Pozdnukhov - University of California Berkeley
Professor Wilfred Recker - University of California Irvine
Professor Kenneth Rose - University of California Santa Barbara
Professor Kyriakos Vamvoudakis - University of California Santa Barbara
Ted Isbell - University of California Santa Barbara
TITLE: Pattern Recognition and Anomaly Detection for Modeling and Simulation in Sustainable Transportation
Workshop Description:
Sustainability policies, such as changes in land use to motivate decrease in travel, and market introduction of new technologies, such as hybrid electric cars or new fuels, have expanded the scope of travel behavior modeling to span numerous scientific fields. This expansion has demonstrated a need for new data and new techniques to perform pattern recognition across scales:
· At the micro-level: e.g. the behavior of persons and their households
· At the meso-level: e.g. traffic streams
· At the macro-level: e.g. activity and movement at the level of cities and regions.
Today we can count on a tremendous amount of data, often referred to as Big Data. Accumulation of Big Data offers unique opportunities to enhance traditional behavioral models that have been limited by the data collected in traditional household questionnaires. New information and new transportation services enable the development of more informative models. For example, by combining data from multiple sources, one can perform a variety of verification and validation tasks on existing data and develop forecasting models for behavioral facets previously neglected. These new opportunities entail risks of misrepresentation and illusionary inferences. Pattern recognition and anomaly detection allow us to differentiate the useful information and misleading results. In the first part of this workshop we will review examples of data fusion and discuss methods and techniques for pattern recognition and anomaly detection.
Transportation modeling and simulation is increasingly aiming at integration of different behavioral facets into internally consistent model systems. Examples include attempts to integrate land use with travel behavior, travel behavior with traffic assignment and route choice, and route choice with traffic simulation. Integration, however, is a general term that is often misused and misunderstood. This workshop will provide guidance in understanding integration of behavioral facets in models. Applications discussed in this workshop will span a wide spectrum. In this workshop, researchers from the University of California together with invited speakers will provide examples of integrated models they have developed. The discussion will include descriptions of policies examined, data used, algorithms and models for pattern development, applications, research gaps, and future work. We will also provide examples from fields outside the traditional domain of transportation. Other topics include modeling and simulation in land use and travel behavior, internet sources for travel analysis, time and space perception, and analyst derived spatial realities.
This technical workshop provides a forum to discuss the state-of-the-art in pattern recognition with potential applications in transportation research. The workshop will take place on Friday, February 27, immediately before the opening reception of the 2015 UCCONNECT Student conference. We invite faculty members, researchers, and graduate students, as well as practitioners from Metropolitan Planning Organizations and the State Department of Transportation (CALTRANS) to participate.
INVITED SPEAKERS INCLUDE:
Professor Mahdieh Allahviranloo - The City College of New York
Professor Richard Church - University of California Santa Barbara
Professor Konstadinos Goulias - University of California Santa Barbara
Professor Krzysztof Janowicz - University of California Santa Barbara
Professor Werner Kuhn - University of California Santa Barbara
Professor Hani Mahmassani - Northwestern University
Professor Alexei Pozdnukhov - University of California Berkeley
Professor Wilfred Recker - University of California Irvine
Professor Kenneth Rose - University of California Santa Barbara
Professor Kyriakos Vamvoudakis - University of California Santa Barbara
Ted Isbell - University of California Santa Barbara