PROJECT

Mission & Work Plan

See the Work Plan ↓

Mission

Why PRO-ARMOUR

PHYSICAL MODELLING

→

PROBABILISTIC DESIGN

→

PHYSICS-INFORMED AI

→ RESILIENT & RELIABLE COASTAL STRUCTURES

Protecting our coasts

Rubble-mound breakwaters are among the most widely used structures for protecting ports, coastlines and critical infrastructure from wave action. At the heart of their performance is the armour layer - the outer layer of large rock or concrete units that dissipates wave energy and ensures the stability and resilience of the structure.

The challenge

Coastal structures are designed for a future that is inherently uncertain. Yet, current design practice still largely relies on empirical formulae and fixed safety factors, without explicitly quantifying the probability of failure. As climate change and evolving met-ocean conditions increase uncertainty, this can lead to structures that are either over-conservative and unnecessarily costly, or insufficiently robust against future conditions.

The centerpiece

The PRO-ARMOUR approach

PRO-ARMOUR moves from deterministic safety factors to probability-based design. The project combines large-scale physical experiments, probabilistic modelling and physics-informed Artificial Intelligence to explicitly account for the uncertainties governing armour stability. A new experimental dataset provides the physical evidence. Probabilistic models quantify risk. AI surrogate models translate computationally demanding analyses into fast and practical engineering tools.

The goal is to bridge the gap between rigorous uncertainty quantification and everyday coastal engineering practice.

From research to engineering practice

Develop

a validated probabilistic framework for the design of rock-armoured slopes.

Experiment

build a new physical dataset on armour stability and wave–structure interaction.

Accelerate

develop physics-informed AI surrogate models capable of reproducing complex predictions at a fraction of the computational cost.

Translate

turn scientific advances into practical methods and guidance for coastal engineers.

Work Plan

Work Packages

WP1

Model definition and identification of variables and indicators

TasksTask 1.1 System definition and probabilistic model formulationTask 1.2 Development of response-based physical model

MilestoneDefinition of the response-based physical model (M1)

WP2

Experimental campaign and data acquisition

TasksTask 2.1 Experimental designTask 2.2 Experimental setup and test executionTask 2.3 Data processing and extraction of performance metrics

MilestoneCompletion of laboratory investigation (M2)

WP3

M3.1, M3.2

AI integration, probabilistic simulation, and validation of the model

TasksTask 3.1 Calibration of probabilistic model using 3D experimental dataTask 3.2 Development of AI-based surrogate modelTask 3.3 Monte Carlo simulation and sensitivity analysis

MilestoneCalibrated probabilistic model (M3.1);
Development of AI-data-driven surrogate model (M3.2)

WP4

Open-access model design tool and dissemination

TasksTask 4.1 Toolbox implementation and release

MilestoneFinal design tool for rock armour layers (M4)

WP5

Training and Impact activities

Timeline

Time schedule

Scroll to see full timeline →

	Year 1 — 2026/27												Year 2 — 2027/28
	O	N	D	J	F	M	A	M	J	J	A	S	O	N	D	J	F	M	A	M	J	J	A	S
WP1 (6 months)	M1
— T1.1 System definition and probabilistic model formulation
— T1.2 Development of response-based physical model — M1
WP2 (8 months)	M2
— T2.1 Experimental design
— T2.2 Experimental setup and test execution
— T2.3 Data processing & extraction of metrics — M2
WP3 (8 months)	M3.1, M3.2
— T3.1 Calibration of probabilistic model using 3D experimental data
— T3.2 Development of AI-based surrogate model — M3.1
— T3.3 Monte Carlo simulation and sensitivity analysis — M3.2
WP4 (runs throughout)	M4
— T4.1 Toolbox implementation and release — M4
WP5 (runs throughout)

Work package activity Task-level activity