As part of RE-ROUTE project, we are organising a talk by our visiting researcher, the details of the talk is as follows:
Date and Time: Wed 17th at 13:00
Venue: CS Boardroom
For the catering purposes, please RSVP here: lunch and coffee will be provided.
Title: New stochastic highway capacity estimation method and why product limit method is unsuitable.
Abstract: Highway capacity is a key variable in the traffic engineering field. It is used for assessment of existing or planned infrastructure or in intelligent transport systems (e.g. for queue prediction). The original concept of a single-value capacity, although still used widely by practitioners for its easy applicability and interpretability, has been overcome with the concept of stochastic capacity. The capacity varies over time being dependent on a broad range of quasi-random variables, such as traffic flow composition, weather, drivers’ behaviour, and others. Kaplan-Meier estimate, also known as the product limit method, has been previously used for estimation of the capacity probability distribution. I will briefly present the product limit method and describe why it seems not fit for such use. Then I will present an alternative method that overcomes the discussed issue. Both methods will be compared and the new method will be analysed from the perspective of reliability and sensitivity to the amount of available data. The new method was also used in a case study to evaluate the effect of a traffic flow speed harmonisation system for freeway work-zones. Its results will be presented, too.
Speaker Bio: Igor Mikolášek received the B.Sc. and M.Sc. equivalent degrees in Structures and transport infrastructure from Brno University of Technology in Brno (BUT), Czech Republic in 2014 and 2017, respectively. He spent two semesters at the Technical University of Denmark in 2015/2016. He continues his education as a doctoral student at BUT. He is currently employed by the Transport Research Centre CDV, Brno, Czech Republic at the Department of road traffic and ITS. His research is focused on traffic flow bottlenecks, especially work zones, and modelling their capacity and impacts, and development of related ITS systems to provide traffic information and traffic control with the aim of improving road safety and optimizing the traffic flow to maximize the capacity.