Foredragsholdere
Maryam Lotfigolian is a PhD candidate at Oslo Metropolitan University, affiliated with the Faculty of Technology, Art and Design, the Department of Computer Science, and Mathematical Modelling. She is part of the ERC-funded REGAL project, where she works on computational methods in quantum chemistry, particularly density-functional theory. Alongside this research background, she works on optimization and operational decision systems in public transport at Ruter, with projects including vehicle staging for demand-responsive transport and bus bunching under real operational constraints. Her work combines mathematical modeling, classical optimization, and exploratory quantum approaches.
Aleksandar Davidov is a Data Scientist at Tet Digital AS, where he works on end-to-end machine learning and generative AI applications, from development through production. He has also led Tet’s quantum computing initiatives, including the development of some of Norway’s first production-ready quantum solutions.
Om sesjonen
Public transport is full of decisions that sound simple until you try to automate them. Where should self-driving vehicles wait between trips? How can we reduce bus bunching and keep buses running at more regular intervals?
In this talk, we show how we built two optimization systems for public transport at Ruter. The first uses historical demand to build a weighted graph and determine where self-driving vehicles should be staged between trips. We solve this problem using established optimization methods and explore where quantum computing could provide an alternative approach. The second focuses on bus bunching, using real-time data and predictions to support decisions such as holding a bus, skipping a stop, or bypassing part of a route.
We look at how different optimization approaches perform on these real-world problems, what quantum computing can realistically contribute, and how these decisions can help reduce passenger waiting and travel times. This is a practical engineering talk about modelling real transport problems, validating optimization results, and building systems that operators can trust.
Passer for
This talk is for developers, data scientists, data engineers, technical architects, and anyone interested in building decision systems from real operational data. The audience will see two concrete examples of how business problems can be modeled as optimization problems, and how to make those systems explainable, testable, and production-ready. No prior background in optimization or quantum computing is required, but some experience with software systems, data pipelines, or backend engineering will help.