Defining the concept of partial observability in network sensor location problems
Event details
| Date | 30.10.2013 |
| Hour | 12:15 › 13:15 |
| Speaker | Prof. Francesco Viti |
| Location |
GC B331
|
| Category | Conferences - Seminars |
The quality of information available on a network is crucial for different transportation planning and management applications. The problem of where to strategically obtain this information has long tradition, and normally can be subdivided into observability problems, focusing on the topological properties of the network, and flow-estimation problems, where (prior) information of traffic states, and/or the specification of which type of application is making use of the sensor data, is needed. This talk provides a broad view of the two categories, and focuses on the development of a new methodology and an intuitive metric for assessing the information quality of a set of sensors in a network in case of partial observability, i.e. when not all sensors characterizing full information coverage solutions are available. This become very useful if one considers that even in small sized networks the solution for full observability requires an exceedingly large amount of sensors. We show the natural interpretation of this new methodology both on toy networks and on a real-sized network, to show how the method performs and selects the most informative links where to install the sensors. Analysis of partial observability solutions shows that the local search algorithm succeeds in finding the links that contain the largest deal of information in a network, and to classify families of full observability solutions.
Bio: Prof. Francesco Viti (UL) is Associate Professor and holds the chair of the Transportation Engineering Lab at the Faculty of Science and Technology of the University of Luxembourg. His research covers a broad range of topics from travel behavior to multimodal network modeling and advanced data collection techniques, carried on at both the University of Luxembourg and at the Katholieke Universiteit Leuven. He teaches Traffic Planning and Management, Transport and Mobility and Infrastructure Design at the Bachelor of Engineering and the at Masters of Energy and Environment and Civil Engineering, and Supply Chain Management at the Master of Mechanical Engineering and as guest lecturer at the Master in Traffic, Logistics and ITS at the Katholieke Universiteit Leuven. He also acts as Expert for the European Commission and as advisor for various EU-funded projects. He has over 100 scientific publications with about 30 articles in journals with Impact Factor.
Bio: Prof. Francesco Viti (UL) is Associate Professor and holds the chair of the Transportation Engineering Lab at the Faculty of Science and Technology of the University of Luxembourg. His research covers a broad range of topics from travel behavior to multimodal network modeling and advanced data collection techniques, carried on at both the University of Luxembourg and at the Katholieke Universiteit Leuven. He teaches Traffic Planning and Management, Transport and Mobility and Infrastructure Design at the Bachelor of Engineering and the at Masters of Energy and Environment and Civil Engineering, and Supply Chain Management at the Master of Mechanical Engineering and as guest lecturer at the Master in Traffic, Logistics and ITS at the Katholieke Universiteit Leuven. He also acts as Expert for the European Commission and as advisor for various EU-funded projects. He has over 100 scientific publications with about 30 articles in journals with Impact Factor.
Practical information
- General public
- Free
Organizer
- Prof. Nikolas Geroliminis & Prof. Katrin Beyer
Contact
- Prof. Nikolas Geroliminis