This research explores the optimization of modular ribbed concrete floor systems as a strategy for circular construction. Ribbed floors can reduce material use and embodied carbon compared to flat slabs, but conventional customized solutions lack scalability while standardized waffle slabs are less efficient. To address this trade-off, a modular design approach is combined with a Deep Generative Design workflow using a Variational Autoencoder (VAE) and Gradient Descent (GD).

The discrete modular configurations are represented as bitmaps, decoupling geometry from structural performance and enabling efficient training of a simple VAE model. Trained on datasets generated in Grasshopper with Karamba3D, the VAE can predict and generate new designs while scaling to larger problem sizes. Optimization is performed by sampling in latent space and refining results with GD.

The workflow flexibly integrates new objectives and constraints without retraining, such as stock availability or embodied carbon. In benchmark tests, the VAE-based optimization outperformed a generic evolutionary solver, achieving lower elastic energy and better stock compliance. The approach demonstrates the potential of deep generative methods for scalable, constraint-aware optimization of modular ribbed floor systems, while challenges remain for extending to more complex structural models.

The construction sector is a significant contributor to the CO2 footprint in the Netherlands, emitting greenhouse gases that harm the global environment. To mitigate these emissions, all sectors must lower their CO2 emission. In construction, an effective strategy is reducing the use of primary materials through component reuse. However, flaws in glass components often result in the replacement of the entire panel. This research focuses on reusing glass panels to reduce waste and CO2 emissions. Through thermal treatment, aged glass could potentially be reused without compromising structural integrity. The glass used in this study is 15-year-old annealed soda-lime silica glass. The main research question is: To what extent can thermal treatment enhance the strength of naturally aged glass, affecting their potential for reuse?

Naturally aged glass, developing surface flaws over time from environmental exposure and human activity, was studied alongside artificially aged glass to assess differences in surface damage and thermal treatment effects. Microscopy and image analysis showed that naturally aged glass treated at 500°C and 600°C exhibited no clear healing trend. In contrast, artificially aged glass treated at 500°C caused minimal changes, while at 600°C, scratch width increased due to subcritical crack growth and a yellow discolouration occurred. Energy-dispersive X-ray spectroscopy revealed chemical changes on the air side of the glass, as the metal coating oxidizes during thermal treatment, altering the surface composition.

The strength of aged glass was evaluated using a coaxial double-ring test at 20 MPa/s, with results analysed via the 2-parameter Weibull distribution and weighted least squares regression. The tests revealed clear differences between naturally and artificially aged glass. Thermal treatment reduced the strength of naturally aged glass, with the 5% fractile decreasing by 28% after heating to 500°C and by 56% after heating to 600°C. In contrast, artificially aged glass improved, with the 5% fractile increasing by 13% after heating to 500°C and by 41% after heating to 600°C. SCALP-05 measurements assess surface stress in three groups: untreated glass, and glass treated at 500°C and 600°C. The average surface stresses measured were -6.49 MPa, -2.76 MPa, and -2.30 MPa. Although thermal treatment reduced surface stress, this effect was insufficient to influence overall strength conclusions.

The influence of thermal treatment on surface flaws differs between naturally and artificially aged glass. Microscopy showed minimal changes in naturally aged glass despite reduced strength, while artificially aged glass appeared visually worse but showed strength improvements in strength testing. This is likely due to the primary failure mode: artificially aged glass is strengthened by healing of dominant scratches, whereas naturally aged glass is mainly affected by overall material weakening.

This research assessed whether thermal treatment could restore the strength of naturally aged glass for reuse. However, the results demonstrated that thermal treatment not only failed to enhance the strength of naturally aged glass but also caused a noticeable reduction in strength. Additionally, yellow discolouration occurred during thermal treatment. In conclusion, thermal treatment compromised both the strength and appearance of naturally aged glass, limiting its feasibility for reuse in construction.