In 2022, the building and construction sector was responsible for more than 7 percent of global carbon emissions, a staggering footprint considering the scale of this industry. A critical question arises: how many of the materials used in erecting homes, bridges, and other infrastructures are truly necessary? The answer lies in reimagining design efficiency and material usage, and recent advances in computational engineering are paving the way toward a future where structural designs not only meet functional and safety standards but also drastically reduce environmental impact.
Topology optimization, a computational technique, has emerged as a powerful tool to minimize material use in structural design. This method algorithmically distributes material within a given space to achieve maximum strength with the least weight. However, while topology optimization excels in generating lightweight, efficient designs at the micro-scale, such as in 3D printing, its application in large-scale construction has been limited. The challenge is straightforward: the optimized designs tend to be overly complex and impractical for conventional construction methods, clashing with the realities of time, budget, and buildability.
Bridging this divide, researchers at MIT have developed an innovative framework that endows topology optimization with practical constraints, making it suitable for real-world construction projects. Presented in a recent publication in Automation in Construction, this framework integrates buildability concerns directly into the optimization process. By allowing designers to impose limits on structural complexity, such as capping the number of components converging at a single point or defining minimum part sizes, the resulting designs become more attainable for contractors and engineers alike.
A remarkable aspect of this new approach is its capability to incorporate multiple materials, including timber and steel, and intelligently assign parts based on their mechanical properties. Where steel excels in bearing compressive loads, timber offers advantages in reducing carbon footprints. The framework balances these materials, distributing them within the design to optimize both performance and environmental impact. This multi-material optimization represents a meaningful advancement in how sustainable construction can be conceptualized from the ground up.
The MIT team’s work, spearheaded by Josephine Carstensen and civil engineering PhD student Zane Schemmer, tackles a fundamental gap in structural engineering: the integration of sustainability within design algorithms that have traditionally prioritized strength and weight alone. Using mixed-integer linear programming, the model makes discrete decisions such as selecting material type for each component and ensuring connection strengths meet construction standards, rather than relying on fractional or approximate assignments.
Unlike 3D printed designs where component assembly is less constrained, conventional construction methods require adherence to established joinery rules and material-specific connection techniques. Timber and steel, for example, demand different approaches to part connections, which the framework meticulously accounts for. This level of detail enhances the feasibility of the optimized designs, ensuring that the theoretical benefits can translate into actual built forms without prohibitive complexity or cost.
An illustrative application of the framework is the reimagining of the Lockport truss bridge, famously spanning the Erie Canal near Buffalo, New York. By selectively applying constraints such as minimum angles between connected components and minimum component sizes, researchers produced simplified yet efficient truss designs that uphold structural integrity while remaining practical to build. These optimized variants included timber-only, steel-only, and hybrid timber-steel configurations, each reflecting distinct trade-offs between carbon emissions and strength requirements.
The insights from this work suggest that multi-material trusses can strike a superior balance: leveraging timber’s lower embodied carbon where feasible, and employing steel’s strength only where structurally critical. This nuanced strategy could unlock significant reductions in the construction sector’s carbon footprint, advancing emissions targets without compromising safety or durability.
Performance-wise, the framework is computationally more demanding than traditional topology optimization methods, due to the added complexity of constraints and discrete choices. Nonetheless, the researchers demonstrated that these demands remain manageable on standard computing devices such as a MacBook Pro, pointing to broad accessibility for civil engineering firms and design professionals. With increasing computational power and optimization software improvements, scaling to larger and more diverse projects is within reach.
Looking forward, the MIT team plans to physically realize scaled-down versions of the optimized designs. Such prototypes will serve to validate computational predictions, offering empirical evidence of constructability and performance. Additionally, ongoing efforts aim to refine and extend the framework’s constraints, enhancing user-friendliness and integration into engineers’ existing workflows.
This research highlights an essential shift in engineering education and practice. As Schemmer notes, sustainability principles have not historically been a core part of structural design curricula. Embedding these principles into early design stages through computational tools presents an unprecedented opportunity to reduce material waste, lower carbon emissions, and align construction with climate action goals.
Funded by the MIT Morningside Academy for Design, this work underscores the emerging intersection of civil engineering, applied mathematics, and computer science in advancing sustainable infrastructure. By moving topology optimization from theoretical exploration to practical implementation, it offers a blueprint for transforming the built environment while addressing one of humanity’s most pressing challenges: climate change.
Subject of Research: Sustainable structural design and topology optimization in civil engineering.
Article Title: “Minimum Carbon Trusses: Constructible Multi-Component Designs with Mixed-Integer Linear Programming”
News Publication Date: Not specified in the content.
Web References:
https://www.sciencedirect.com/science/article/pii/S0926580526003262?dgcid=author
Image Credits: Courtesy of Josephine Carstensen and Zane Schemmer
Keywords:
Construction engineering, Civil engineering, Structural engineering, Bridge construction, Building construction, Algorithms, Sustainability, Computer science, Computer modeling
