I am Yahya, I am a postdoctoral researcher at the University of Glasgow, Urban Big Data Centre, and I am working within the AI for Collective Intelligence (AI4CI) hub. Before joining the UBDC, I was a postdoctoral researcher in King’s College London (KCL) and University College London (UCL). Prior to this, I acquired my PhD from the University of Manchester with a focus on urban simulations.
My interests are twofold: first, I am curious about individual behaviours in urban systems and how they collectively can lead to emergent patterns in cities; second, I am interested in urban simulation methods that can represent such individual behaviours and support exploring and understand urban systems further. My research involves both, where I aim to extend the boundaries of urban simulation methods to answer real world questions and inform social policies in cities. So far, my academic career has been driven by this passion for urban simulation methods, with an endeavour in exploring different urban systems – including housing, health and transport systems.
Being in the AI4CI hub has been a great opportunity to explore applying AI methods within urban simulations, while maintaining a meaningful impact on real world policies.
What are you working on?
I am currently working mainly on a research project to understand the impact of transport interventions in Scotland (such as closing a highway) on the daily plans of different demographic groups. The research involves the development of a transport Agent Based Model (ABM) that involves using AI methods to simulate individual daily activity scheduling and modifications in case of transport interventions.
I am also working on research where we explore the use of transformer-based models (such as LLMs) to simulate human behaviours in ABMs – with a case study using a housing model.
What excites you most about your field of research?
The thing that excites me the most is the potential to make impacts on policies, that can end up shaping our social and physical urban environment as better places to live in. For instance, it will be very rewarding in our current work to inform on the impacts of planned transport interventions on specific demographic groups.
Another aspect that equally excites me is building the capacity to embed new methods (including AI) within urban simulations to further make a real-world impact. In this case, it is rewarding to be able to clearly communicate and justify the use of such new methods to generate trust in the models and their outcomes.
Who has influenced your academic journey?
When I was formulating my PhD proposal, I came across two books that strongly influenced my research: “Complexity: Theories of cities have come of Age” by Juval Portugali; and “Agent-based models of geographical systems” by Alison Heppenstall, Andrew Crooks, Linda M See and Michael Batty. The former book drove my passion towards a different understanding of cities as complex systems that are subject to emergent patterns. This drove me to exploring the latter book as it was fascinating to understand how Agent Based Models (ABMs) can be used to explore and understand different components of complex urban systems.
Can you recommend a paper you think we should read?
I recommend “Building a science of cities” by Michael Batty – the godfather of the field of urban simulations, and for a longer in-depth read I recommend the book “The new science of cities” by Michael Batty. I like those two reads as they showcase how our understanding of cities has evolved into cities as complex systems. They also highlight complex system aspects within cities, such as interactions and emergence, which is overall a great entry point for the curious minds around urban simulations.
What are your ‘Top Tips’ for early career researchers?
I would provide three key tips to ECRs, particularly for those seeking an academic career. First, follow research and acquire skills in something you are passionate about. This passion will show to some extent in the quality of your work and in your academic interactions, and this can open opportunities ahead. Second, be open to explore domains which intersect with the filed you are passionate about. ECRs can have that flexibility which enriches their interdisciplinary experiences. Third, try to connect with researchers who successfully navigated the ECR phase to understand such process. Reach out and attend ECR seminars – you would be pleasantly surprised at how much some people are willing to help and guide.