How to Automate Building Design: Product-Informed Digitization
Photo by ThisisEngineering on Unsplash
Earlier work carried out within a tier-one property and construction company focused on exploring how to automate complex building design processes. The industry insights shared here are drawn from that experience, and they continue to shape how Exelsiv approaches AEC automation.
Read Part 1: How to Automate Building Design: A Case Study
Read Part 3: How to Automate Building Design: Adapting to Local Codes
Why Product Data Matters More Than Ever in Automated Design
Design automation has advanced rapidly, but much of it still operates in a digital vacuum. The logic and rules that drive automated systems are often based on idealized assumptions, such as perfect materials, unlimited supply, and uniform performance. It is very rarely based on the variability and constraints of real-world products.
As a result, models can optimize theoretical designs that later prove difficult or impossible to deliver.
For example:
A beam that meets all digital load and deflection rules might exceed the supplier’s manufacturable length.
A connection detail may comply structurally but fail certification because the tested product variant differs from the modeled one.
A wall system designed parametrically might rely on a product unavailable in the supply chain.
These gaps between digital logic and material reality limit the industry’s ability to trust and scale automation. The next frontier lies in connecting automated systems with the properties of actual products, namely their physical, measurable, and behavioral characteristics.
The Problem
Most design automation initiatives focus on codifying engineering logic: load paths, boundary conditions and coordination rules. This is essential, but it often stops short of engaging with the physical nature of the materials and products themselves.
The result? Automated systems that function beautifully in theory but lack awareness of how real products behave under stress, moisture, or connection loads.
This creates a crucial gap between design intent and product performance; a gap that limits automation’s accuracy, reliability, and usefulness in real projects.
The way forward lies in embedding verified, up-to-date data about the materials and products we actually build with, directly into automated design systems.
The Approach
The path forward is to evolve from rule-based systems to product-informed systems. That means embedding the measurable properties of actual products (not generic assumptions) directly into design logic.
This approach brings the reality of manufacturing and material behavior into the digital workflow. It captures data that reflects how products actually perform, respond, and interact within a system.
This includes three key dimensions:
Physical properties: strength, stiffness, moisture response, fire behavior, and tolerances.
Performance limits: maximum spans, fixity conditions, deformation ranges, and load capacities.
Material variability: differences introduced by manufacturing methods, fabrication tolerances, and environmental exposure.
(Regional codes and compliance requirements are also critical; a topic we’ll explore in a future case study.)
When this data is linked to automation, design systems no longer just “apply rules”. they simulate real behavior.
A beam isn’t just sized according to equations; it’s selected and adjusted according to the specific properties of the product that will be used to make it. This concept aligns with the evolution toward Digital Twins that incorporate closed-loop real-world data for refinement.
Implementation in Practice
Embedding product data requires collaboration between engineers, manufacturers, and digital product teams. It’s as much a data challenge as it is a cultural one, connecting the knowledge held in physical products to the systems that drive design.
1. Data Connectivity
Manufacturers hold rich datasets about product performance, from mechanical test results to environmental response data. Integrating this data through structured formats and APIs allows design systems to validate geometry and behavior in real time.
2. Shared Data Structures
For automation to interpret product behavior, data must be structured consistently. For example, specifying how stiffness, strength, and moisture expansion are represented and related to geometry.
3. Feedback Loops
Data from testing, prototyping, and field performance should feed back into design logic. Over time, this creates a continuously improving system that refines product understanding with every iteration.
4. Dynamic Parameters
Rather than relying on fixed values, design rules can reference parameter ranges that adjust automatically based on actual product data, allowing designs to flex with different suppliers or evolving specifications.
A Case in Point: Mass Timber
Mass timber demonstrates why product-informed design matters. Two cross-laminated timber (CLT) panels can appear identical in a model but behave differently depending on their manufacturing process, adhesive system, or fiber orientation. These variations affect span capacity, stiffness, and how joints distribute loads. Surface finish, too, influences visual feel and detailing.
When product-specific data is embedded into design rules, automation can instantly adapt geometry and connections to reflect each supplier’s tested characteristics. Engineers receive immediate feedback on feasibility, avoiding rework and improving design certainty.
The Outcome
Integrating product-informed design transforms automation from a digital exercise into a material-aware decision system. This enables:
Accuracy → designs reflect actual product behavior, not theoretical averages.
Efficiency → decisions are faster because material properties are already embedded.
Reliability → fewer surprises during fabrication or installation.
Innovation → new products can be tested virtually, accelerating R&D and design feedback.
Automation no longer ends at the computer screen; it becomes a bridge between engineering intent and material reality.
Exelsiv Insight
At Exelsiv, we see product-informed design as the foundation for the next generation of digital delivery. Our work connects engineering logic with verified product data so that systems reflect the real world.
We help organizations:
Map product and material data into structured digital systems.
Define parameter ranges that reflect actual performance and not just assumptions.
Build feedback loops that continuously refine understanding of product behavior.
By tying design back to the properties of actual products, we create systems that are smarter, more reliable, and ready for the future of construction.
If your organisation is exploring how to make product-informed design part of its future, Exelsiv can help chart the path forward. Get in touch with us today.
