Browse by author
Lookup NU author(s): Emmanuel AYODELE, Professor Margaret Carol Bell CBE, Dr Neil Thorpe, Dr Fabio Galatioto
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
The current increased car use causes congestion resulting in delays and elevated carbon emissions alongside unacceptable levels of pollutant concentrations. Congestion problems affect the economy and exacerbate air quality with detrimental effect on human health and calls for improvement in network efficiency and a reduction in vehicle kilometres travelled. Fundamental to both, is the need for much better informed decisions not just in the context of traffic management from day to day but also in more long-term strategic planning and policy changes. Autonomic Computing (AC) is capable of managing highly complex networks with minimum human intervention and allows recurrent network problems such as congestion and air pollution hotspots to be identified thus reducing costs without compromising efficiency. However, key components of an autonomic system are not only algorithms based upon which the system identifies problems but also formulated advice on the selection of solutions to the problems identified. Status monitoring algorithms require continuous analysis of traffic and pollution levels as well as meteorological conditions from both the static and dynamic perspective. This leads to problem identification. In order to identify effective solutions, origin and destinations of routes in the network, also, are needed. Whilst SCOOT - Split Cycle Offset Optimisation Technique delivers the former, Bluetooth technology offers the opportunity to fulfil the latter. This paper explores the various ways through which Bluetooth data provides the relevant traffic information to support intelligent decisions in an autonomic environment. The research focuses on the analysis of vehicle by vehicle data collected from Bluetooth sensor arrays deployed across three geographical locations in Manchester captured over a period of six months and delivers the hour by hour origin – destination matrix across the network. The accuracy of the origin-destination is shown to be sufficient to derive performance measures such as journey time and carbon emissions, and thus to inform effective solutions. Finally, how Bluetooth data informs the self-cleansing, self-healing, self-protecting, self-configuring and self-optimising properties of an Autonomic Decision Support Systems (ADSS) is demonstrated. Normal 0 false false false EN-GB ZH-CN X-NONE
Author(s): AYODELE E, BELL MC, THORPE N, GALATIOTO F
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Unpublished
Conference Name: 46th Annual UTSG Conference
Year of Conference: 2014
Publisher: Newcastle University
