This document is part of technology development efforts of the Horizon 2020 All Atlantic Ocean Sustainable, Profitable and Resilient Aquaculture (ASTRAL) project. The deliverable describes the functional and technical architecture of the Artificial Intelligence Data Analytics Platform (AIDAP). The goal of this platform is to give access to Artificial Intelligence (AI) analytics to farmers, especially integrated multi-trophic aquaculture (IMTA) farmers. It provides a user-friendly interface to visualize all the important features (for example status of water quality, biomass, health, etc…) of one farm in one place, including AI advice for sustainable and efficient production.

This document describes the main technical components of the platform, for data ingestion, AI prediction and visualizations while giving some context to its usage and outcomes. Data ingestion is a critical part of the platform, which allows us to centralize and standardize the data coming from different sources. Important work has been done to propose a data standard dedicated to aquaculture and to build an efficient way to collect various kinds of data.

AI prediction is the core of the intelligence of the platform. While the data modelling is not done in the platform but offline, in another process, the platform offers technical support to run the predictions in real-time as soon as the data are collected. It requires powerful AI support tools and architecture.

Data visualisation is the front end for the end users. It must be easy to use and must respond to their specific needs while being technically strong enough to support real-time data and AI-generated advice display. It requires a light but powerful tool to reveal the true potential of AIDAP dedicated to aquaculture users.