CPU Lab researchers publish pioneering work on cities, design, and complexity

Researchers from the [CPU]lab at Manchester School of Architecture have had several articles published in academic journals in 2025, reflecting the breadth of research specialism within the group.

Prof Ulysses Sengupta, who leads the [CPU]Lab at MSA, says: “We have been lucky to come together as an interdisciplinary group of researchers in the [CPU]lab recently. The outputs advance our focused research agendas around complexity, cities and design studies utilising emerging technologies, while exposing opportunities for collaborative research with new members who bring their own research agendas and frameworks. We look forward to exploring collaborations at the reflected intersections of the humanities, social sciences and design sciences. ”

The first article,“Exploring city dynamics through tweets: A framework for capturing urban activities as complex spatiotemporal patterns” by Mahmud Tantoush, Ulysses Sengupta, and Liangxiu Han, introduces a novel framework for mapping urban activity using geolocated Twitter/X data. Focusing on Manchester, the researchers reveal how space (urban form) and place (human activity) interact in complex, often non-linear ways. Their model treats cities as Complex Adaptive Systems and shows promise for comparative analysis across cities using social media data. The article has been published in Cities, a global top 10 Urban Studies journal.

Guillermo Sánchez Sotés, alongside Thomas Fischer (Southern University of Science and Technology) and Christiane M. Herr (Xi’an Jiaotong-Liverpool University, China), critically examine architectural theory in “The Appropriation of Autopoiesis in Architecture.”, published in Constructivist Foundations.  They unpack how Patrik Schumacher’s influential but controversial application of the biological theory of autopoiesis serves more to advance his architectural agenda than to clarify theory for his readers. Through text analysis, they argue that Schumacher’s appropriation lacks conceptual rigor, highlighting the pitfalls of borrowing from other disciplines without full fidelity.

Hamid Khalili’s “The Cinema of Béla Tarr: The Architectonics of Time, Movement and Hapticity”, published in Architecture Theory Review, delves into the spatial and temporal richness of Tarr’s “slow cinema.” Drawing on both film and architectural theory, Khalili constructs a compelling framework linking the tactility of Tarr’s visuals to concepts like montage in modernist architecture. With original material, including an unpublished nine-hour interview with Tarr and archival research on Sergei Eisenstein, Khalili reveals how Tarr’s filmmaking challenges conventional perceptions of time and space in cinema and architecture alike.

Kristen Zhao, alongside Robin Chang (RWTH Aachen University), and Guibo Sun (University of Manchester), explore the failure of urban experimentation in Hong Kong in an article published in the Journal of Urban Mobility.“Experiment unlearned: Unpacking leadership and learning of key actors in a Hong Kong Street experiment” evaluates the Healthy Street Lab 2.0 project and identifies a lack of leadership and institutional flexibility as central to its shortcomings. Despite well-funded, co-creative efforts between government and civil society, most prototypes were eventually dropped. The authors point to entrenched power dynamics and a reluctance to learn as key barriers to transformative urban change.

Finally, Ulysses Sengupta, alongside Solon Solomou (University of Nicosia) examine the novel use of decision making intelligent agents (AI) within cutting edge urban simulation in “Simulating Complex Urban Behaviours with AI.” They compare traditional logic-based agent models with new cognitive architectures that mimic human-like decision-making for location choice. Their findings demonstrate how cognitive  agents can better reflect the complex behaviours seen in real-world urban economies and decision making processes involving  unpredictability, memory, and learning. The research advances the use of emerging technologies to plan, design and understand comutationally simulated future city scenarios.