Data-Driven Design Optimization

An abandoned basketball court reimagined into a modern loft — optimized using AI-driven design and data. Would you live here? This transformation isn’t just visual. AI-based space optimization tools were used to model how people actually live, move, and use space: 1,000+ layout simulations evaluated for circulation efficiency, light access, and privacy 20–30% reduction in wasted space by optimizing zoning and vertical volume A raised bedroom increased usable floor area by ~15% without expanding the footprint AI daylight simulations improved natural light penetration by 25–35% across the day Storage and furniture placement optimized to reduce movement friction by up to 40% The outcome: A space that feels significantly larger, brighter, and calmer — without adding square meters. Why this matters: In dense cities, every m²/foot² saved can reduce construction cost by 8–12% AI-optimized layouts show 10–20% higher long-term livability scores compared to traditional designs Adaptive reuse projects like this can cut embodied carbon by 50–70% versus new builds This is what happens when AI meets architecture: Less waste. Better living. Smarter use of what already exists. #AI #Architecture via @alot_design #SpaceOptimization #GenerativeDesign #AdaptiveReuse #SustainableDesign #FutureOfLiving #UrbanInnovation

Traditional surrogate-based design optimization (SBDO) is hitting a wall, especially with high-dimensional, complex designs. In this new paper, Dr. Namwoo Kang presents a next-gen framework using generative AI, integrating three key models: - Generative model (design synthesis) - Predictive model (performance estimation) - Optimization model (iterative or generative) Rather than optimizing directly in a high-dimensional design space (x), the workflow introduces a low-dimensional latent space (z) learned via generative models. ➡️ z → x → y z = latent variables x = CAD geometry y = performance (drag, stress, etc.) This means we’re no longer hand-coding design parameters or doing trial-and-error with simplified surrogate models. 🧠 Why this matters: - Parametric modeling is no longer a bottleneck - Complex shapes are learned directly from CAD - Dynamic and multimodal performance data (1D, 2D, 3D) can be used - Near real-time optimization is possible #AI #GenerativeDesign #CAE #DesignOptimization

AI is starting to challenge how we approach Design of Experiments (DoE) in industrial R&D. Traditional DoE requires you to define everything up front… factors, ranges, interactions. ML-guided optimization learns as it goes. Instead of fixing an experimental plan, the model watches results and decides where to test next, focusing on regions where behaviour is changing or still uncertain. Take Monolith AI's "Next Test Recommender" tool. After each batch, they fit a surrogate model, estimate uncertainty, and select the next points to maximize information gain. Classical DoE is planned exploration. ML-guided optimization is adaptive. But this only works if simulations are cheap. If runs are slow, licence-constrained, or organisationally painful, you don’t get enough data to train a useful surrogate. You fall back to intuition. That’s where infrastructure matters. Containerised FMUs (by the Modelica Association) remove licence bottlenecks. API-driven simulations let you parallelise, store, and replay every run. Tools like the Quix FMU runner are built for exactly this: turning simulations into something you can scale and automate. Now you have enough data for the model to actually learn. At that point, experimentation stops being manual. It becomes software that improves with every run. Less tweaking. More exploration. And once exploration is cheap, engineers stop asking “what should I tweak next?” and start asking “what parts of the design space haven’t we explored yet?”

Thermal Simulation is here! Doable with Newton too, with some tuning! Traditional CFD simulation for a single datacenter configuration: 8-12 hours AI surrogate model prediction: < 1 second Here's what Wistron and NVIDIA just proved is possible: The Old Way: → Design a datacenter hot aisle layout → Wait hours for OpenFOAM simulation → Results show a hotspot → Tweak the design → Wait hours again → Repeat 50+ times to optimize → Weeks of iteration The New Way: → Train a 3D UNet on simulation data once → Test 1000s of configurations in minutes → Real-time temperature/airflow predictions → Instant design optimization → Days instead of weeks Why this matters: Data centers consume 1-2% of global electricity. Poor thermal design = wasted energy + hardware failures + $$$ down the drain. This AI approach using NVIDIA PhysicsNeMo doesn't just predict faster—it enables: ✓ Rapid exploration of design variations ✓ Real-time "what-if" scenarios during planning meetings ✓ Physics-guided learning (works even with limited data) ✓ Digital twin capabilities for existing facilities The Technical Magic: They combined: - 3D UNet architecture for spatial predictions - Signed Distance Fields to capture geometry changes - Sinusoidal embeddings for sharp flow features - Physics-informed loss functions (data + governing equations) The physics-informed variant especially shines when training data is limited. The model learns the underlying physics, not just patterns. Real Impact: Wistron is now using this to transform factory planning and operations with digital twins built on NVIDIA Omniverse + PhysicsNeMo. The future of engineering isn't replacing simulation—it's making it 10,000x faster. Source: https://lnkd.in/gaTgtizh

Digital & AI-Driven Concrete Design: From Art to Data Science Concrete mix design is no longer just experience and trial batches, it’s becoming a digital, data-driven process. With AI and advanced modeling tools, companies can now: 📊 Predict performance before the first batch is poured (strength, durability, shrinkage, and workability) 🧠 Optimize mix designs using historical data, material variability, and performance targets 🌱 Reduce cement content and CO₂ while maintaining or improving performance 🔁 Adapt in real time as aggregates, SCMs, or environmental conditions change Digital mix design platforms and machine-learning models are helping producers move faster, waste less material, and design concrete that is fit-for-purpose, not overbuilt. That being said, do you trust AI to design your mixes without doing the trial batches? #concrete #concreteconstruction #concreteindustry #aiconstruction #ai #nanotech #sustainable #sustainability Jon Belkowitz, PhD, PE, Mallory Westbrook

🔬 Engineering design synthesis is moving from manual iteration to automated, data-driven approaches. MIT researchers map out how deep generative models are enabling this shift, with important technical implications for how we develop products in the reference paper below. The paper provides a systematic analysis across: 🎯 Design Problems: • Topology optimization • Materials & microstructure design • 2D/3D shape synthesis • Multi-component product design 💾 Data Representations: • Voxels & point clouds for 3D • Images for 2D designs • Parametric specs for manufacturing • Graphs for component relationships 🧮 Model Architectures: • GANs with various conditioning approaches • VAEs for latent space exploration • RL for sequential design decisions • Integration with physics-based simulation ⚖️ Loss Functions: • Performance metrics from simulation • Manufacturability constraints • Style transfer for design aesthetics • Multi-objective optimization 📊 Key Datasets: • UIUC airfoil database • ShapeNet/ModelNet for 3D shapes • BIKED bicycle design dataset • Material microstructure collections 📝 Reference: "Deep Generative Models in Engineering Design: A Review" by Regenwetter et al. https://lnkd.in/g_mMR-8y S2 Labs #EngineeringDesign #MachineLearning #TechnicalResearch

🧠 What advantages does Generative AI (Generative Design) have over traditional design optimization? In my last post, I compared deep learning–based surrogate models with traditional ML approaches. This time, from a design optimization perspective, let’s look at how Generative AI (built on deep learning) differs from—and often surpasses—conventional design workflows. Below are five standout advantages. 🔹1. Non-Parametric Design Optimization Traditional optimization starts by parameterizing a 3D shape into a small set of 1D variables, which inevitably restricts the design space. With Generative AI, we learn a latent representation z from the raw 3D geometry x, then search for the optimal z and decode it back to the design x. This enables non-parametric shape generation unconstrained by hand-crafted parameters. The generator can be coupled directly with CAE to evaluate objectives and constraints; when sufficient CAE data exists, a predictive model can be trained and paired with the generator for even faster loops. 🔹2. Real-time Inverse Design Conventional predictive models learn the forward map x → y (design → performance). What designers really want is the inverse: given a target y, find the optimal design x. Generative AI enables this via conditional generation, producing optimal designs that meet target specs in (near) real time. 🔹3. Synthetic Data Generation Building accurate predictive models requires many (x, y) pairs, yet industry often lacks diverse design examples x. Generative AI can create synthetic 3D designs, which can be automatically evaluated with CAE to produce labels y. This jump-starts dataset creation and alleviates the most common bottleneck in training predictive models. 🔹4. Realistic Rendering for Instant Customer Feedback A practical benefit of Generative AI is rapid, photorealistic rendering of optimized concepts from simple text prompts. You can show customers lifelike visuals immediately, collect feedback, and steer the design toward what users actually prefer—bridging aesthetics and performance early. 🔹5. Seamless Design–Engineering Integration Generative AI closes the loop between design and engineering. From a rough concept sketch, it can produce 3D geometry and renderings and estimate engineering performance via predictive models. Instead of waiting for long feedback cycles, designers get live engineering guidance and can iterate toward optimal solutions themselves. We’ve summarized these ideas in the attached visual PDF. 👇 #DesignOptimization #GenerativeAI #GenerativeDesign #DeepLearning #EngineeringDesign

Sustainable design isn’t just a goal, it’s a measurement, and the firms that lead today are the ones that pair creative vision with data-driven decisions. HMC Architects ' recent work on a STEM center shows what this looks like in practice. By using Forma Site Design to evaluate daylight, wind comfort, noise, microclimate, and natural site features, they didn’t just explore ideas. They tested them against real performance goals early, under a tight timeline. What stood out to me is how this approach: • Brings environmental and human factors into design from the very beginning, not as an afterthought • Turns intuition into insight, making decisions easier to explain and stand behind • Helps teams move faster without sacrificing quality, freeing time for higher-value work As we think about the next generation of schools, infrastructure, and communities, this is the shift that matters. When performance data is available early, teams can make more confident decisions and deliver outcomes that truly serve people and place. If you want to go deeper, we’ve shared the full story on how HMC Architects used Forma Site Design to drive these outcomes. Check out the link to it in the comments below. #DesignForGood #DataDrivenDesign #SustainableArchitecture #AEC #AutodeskForma