Performance-Based & Data-Driven Design
Performance-based, life-cycle design — with machine learning to model, predict, and optimize structural performance.
Overview
Modern structural engineering generates rich data — from experiments, simulations, and monitoring. ALAMS applies performance-based design and machine learning to turn that data into faster, better decisions: predicting how structures behave, designing complex systems efficiently, and weighing performance against cost across a structure’s whole life.
Focus areas
ML for self-centring buildings
Machine-learning prediction and design of self-centring buildings for reliable, efficient seismic performance.
AI design for complex systems
AI-based design solutions for complex structural systems where conventional methods are costly or slow.
Life-cycle cost of isolated bridges
Life-cycle cost analysis of bridges isolated with SMA–lead-rubber bearings to balance performance and cost.
Climate-resilient multi-hazard design
Designing bridges that stay resilient under multiple, climate-driven hazards across their service life.
How we work
Data
Gather from tests, simulations, and monitoring.
Model
Train machine-learning and mechanistic models.
Predict
Forecast performance under demand.
Optimize
Balance performance, cost, and risk.
Explore more
See how this area connects to our publications, the team, and the rest of the group’s research.