SASNet Training Workshop: Data Management for Urban Transport Operations
- Friday, 10th June 2016
- 9:15am - 1:15pm
- Sir Alwyn Williams Building - Level 5, Lilybank Gardens, Glasgow, UK G12 8QN get directions
The half-day workshop aims to introduce data management methods and modelling for urban transport planning, using specific case studies to illustrate the use of sensors and high performance computing. The workshop will be delivered by renowned leaders in the field of urban informatics, in two focused sessions:
- Macroscopic traffic flow modelling and control of heterogeneous cities with mult-sensor data
- Traffic Management Enabled By High Performance Computing
The free workshop will likely be of interest to postgraduate students, early career researchers, and anyone with an interest in urban data management, especially as it relates to transportation, including transportation government operators and private companies.
***Please note: Registration on Eventbrite is required for catering purposes. Refreshments and lunch will be provided at the end of the workshop over networking. If you have any questions about the workshop, please contact: Keith.Maynard@glasgow.ac.uk
9:00am - 9:15am Coffee/tea and pastries provided
9:15am - 9:20am Introduction by Prof. Piyushimita (Vonu) Thakuriah, Director, Urban Big Data Centre
9:20am - 10:20am Dr Konstantinos Ampountolas lectures on ‘Macroscopic traffic flow modelling and control of heterogeneous cities with multi-sensor data’ and will be based around his research as described in the following abstract:
Mobility and transportation are two of the leading indicators of economic growth of a society. Traffic congestion has a significant impact on our daily lives as it directly affects the productivity, health and environment, which call for drastic and radical solutions. Control of traffic congestion has attracted a lot of research attention during the past few decades. Conventional traffic management faces limitations but the introduction of Intelligent Transportation Systems (ITS) technologies and new sensing hardware promise significant progress in reducing the congestion level in cities. Nevertheless, modelling and control of heterogeneous cities with multiple attractions/regions of congestion remain a big challenge, due to the high unpredictability of choices of travellers, the uncertainty in their reactions to the control, and the spatiotemporal propagation of congestion. In this talk, we macroscopically describe the traffic dynamics in heterogeneous cities with multi-sensor data.
We present results on two modelling variations for unimodal (car only) and bimodal (car and bus) cities that can be integrated in advanced traffic management schemes for zone or perimeter control. The heterogeneous network of Downtown San Francisco is used to test the proposed models and control schemes. Our research seeks to shed some light in the realistic modelling and efficient control of traffic flow for overcrowded heterogeneous networks.
10:20am - 10:30am Q&A discussion
10:30am - 10:45am Comfort break
10:45am - 12:45pm Dr Anuj Sharma provides an introductory level training workshop session ‘Traffic Management Enabled By High Performance Computing’, focusing on traffic management of a wide area network using multiple data streams.
In the past, only a limited number of data streams have been available from infrastructure mounted sensors for use in traffic management, but with the advent of smart phones, connected cars, etc., several non-traditional sources of data are being made available for analysis. However, these heterogeneous data streams arrive at very high rates and often can have spurious data. This exponential growth in relevant data streams has brought new opportunities and challenges in the realm of traffic management. Increased data enables improved monitoring, prediction, and management of traffic, but only if automation is employed to handle the data volumes. These volumes would otherwise exceed the abilities of humans to process.
This course will start with the introduction of data streams and high performance computing. Then we will dive into data stream quality control. Finally the course will give an overview of conducting performance estimation, incident detection and traffic control using multiple sources of data. Towards the end we will discuss a glimpse in future and role of innovative technologies.
12:45pm - 1:15pm Refreshments and lunch provided over networking
This training is being provided as part of the new ESRC-funded Social Analytics Strategic Network (SASNet). The network, jointly founded by The Urban Big Data Centre (UBDC) and Business and Local Government Data Research Centre (BLGRC) focuses on capacity-building for social analytics of emerging heterogeneous forms of data, including big data. This event is presented by two SASNet fellows from the SASNet Fellowship Programme, a scheme developed to allow academics and non-academic visitors to join the research community at either BLGRC in Essex or UBDC in Glasgow. Please visit the websitefor more information as well as details on how to apply to become a SASNet Fellow.