INVITED SPEAKER

Dr. Chi-tathon Kupwiwat

Bio: Dr. Chi-tathon Kupwiwat is a Lecturer in the Department of Architecture, Faculty of Architecture, Chulalongkorn University, Thailand. He received his Ph.D. in Engineering from Kyoto University, Japan, where his research focused on reinforcement learning–based optimization of truss and frame structures. His work lies at the intersection of computational structural design, graph-based modeling, and artificial intelligence, with an emphasis on representing building structures as learnable systems for performance-driven design. His recent research investigates graph-based reinforcement learning frameworks for structural optimization, including geometry, topology, and sizing problems for trusses, lattice shells, and building-scale structures. He has contributed to the development and evaluation of multi-task and multi-agent reinforcement learning approaches, as well as the integration of non-structural performance criteria such as life-cycle environmental impact into learning-based design workflows. His current research focuses on robustness, out-of-training application, and policy transfer across structural typologies, with the long-term aim of developing transferable and scalable learning-based optimization models for structural design.

Speech Title: Graph-Based Reinforcement Learning for Structural Design: From Representation to Transferable Optimization Policies

Abstract: This research investigates graph-based reinforcement learning (RL) as a framework for structural design and optimization, with the aim of moving beyond problem-specific numerical solvers toward reusable, learning-based design strategies. Structural systems are represented as graphs, enabling variable-sized and topology-dependent structures to be processed using graph neural networks, while design modifications are formulated as sequential decision-making actions. Within this formulation, reinforcement learning has been applied to geometry, topology, and sizing optimization problems for lattice shell, truss, and frame structures. Recent and ongoing studies demonstrate that graph-based RL can effectively handle complex structural systems and outperform conventional heuristic optimization methods under trained conditions. Extensions of this framework include multi-task and multi-agent reinforcement learning formulations, as well as the integration of non-structural performance criteria, such as life-cycle environmental impact, enabling simultaneous consideration of structural efficiency and sustainability within a unified learning model. Building on these results, future research will focus on improving robustness under parametric variation, systematic evaluation of out-of-training application, and policy transfer across structural typologies.

Assoc. Prof. Ahsen Maqsoom

Bio: Dr. Maqsoom is the Program Head for Construction Engineering Management and an Associate Professor at the Green Tech Institute, UM6P. He holds a Postdoctoral degree in Sustainable and Safe Construction from Chulalongkorn University, Thailand, and earned his Ph.D. and M.Eng. in Construction Engineering and Infrastructure Management from the Asian Institute of Technology (AIT), Thailand. Before joining UM6P, he served in various organizations, including international academic institutions and multinational industrial firms across Thailand, Pakistan, South Korea, and Oman. With over 11 years of teaching and 6 years of industry experience, Dr. Maqsoom brings expertise in project and construction management. He has authored more than 130 international journal and conference papers, contributed to several book chapters, secured and led multiple research and industry-funded projects, and supervised numerous Ph.D., Master’s, and undergraduate students. Dr. Maqsoom is an Editorial Board Member of the Sustainability journal, a Professional Engineer, and a Member of the American Society of Civil Engineers (ASCE). As a Convener, he has played a leading role in establishing Civil Engineering and Construction Engineering programs at the Bachelor’s, Master’s, and Ph.D. levels across several universities. His research centers on sustainable and resilient construction systems, green buildings, construction safety, and infrastructure management, while also examining the role of digital technologies in transforming the construction industry. He further explores energy-efficient and low-carbon construction, circular economy practices, Industry 4.0 adoption, and risk-based project management to advance smarter, safer, and more sustainable built environment solutions.

Speech Title: Decarbonising the Built Environment Pathways Toward Net-Zero Construction

Abstract: The building sector is becoming one of the most decisive areas of the global warming. In rapidly urbanising developing economies, constituting between 11 and 28 percent of national GHG emissions, operational carbon intensity targets 21 kg CO 2/m 2/year in Nigeria, and 52 kg CO 2/m 2/year in China. This comparative synthesis involves ten emerging and developing economies, India, China, Brazil, Nigeria, and Indonesia, South Africa, and Mexico; Morocco, as well as evaluating operational emissions, embodied carbon, renewable energy penetration, green building adoption, policy strength, investment readiness, and transition risk. The comparison shows that there is major difference in the paths of decarbonisation. China and Brazil have a relatively good policy framework and mobilisation of investment, and Morocco and Mexico have a moderate readiness with the help of renewable growth and green bond markets. Pakistan and Nigeria on the other hand have serious structural setbacks such as low levels of green buildings (1-2 per cent), a gap in financing (over 14-18 billion a year) and poor regulatory frameworks. Most economies have a low penetration of renewable energy in buildings which means a reliance on fossil fuel. The intensity of embodied carbon, which is mainly caused by the use of concrete and steel, is a burning issue. Although the role of low-carbon cement is on the rise in China (28% and Brazil (22%), adoption is minimal in some of the lower-income economies. The recycling rates and the integration of timber have not been developed according to the standards of net-zero, which is limiting the decarbonisation of the lifecycle. China has the highest multidimensional readiness index, which includes the emissions performance, the strength of its policies, the financial capacity, and the adoption of technology (68/100), the next is Brazil and Morocco, with Pakistan and Nigeria having a high transition risk. The results provide a significance on the aspect that decarbonisation of construction in emerging markets is not necessarily a technological challenge but a systemic revolution that necessitates concerted climate finance, policy change, material innovation, and institutional building capacity. The risk of net-zero lock-in in the long-term is that as construction growth outpaces infrastructure requirement in the Global South, it becomes increasingly difficult to keep up with the net-zero trajectories. On the other hand, resilient, future-ready urban development can be unlocked through the focused investment in green building standards, building systems using renewable energy, low-carbon materials, and blended finance tools. This comparative synthesis provides a policy roadmap to policymakers, investors and industry leaders to speed up the process of the net-zero transition in the built environment.