Ayesa – Optimisation in the process industry
Ayesa
Sector: Industry
Business Case
How to reduce energy consumption and CO2 emissions in those industries whose manufacturing processes last hours/days? How to take a common approach to a cognitive plant prediction/optimisation that works for 4 different productive plants?
Objectives
Identify, collect and homogenise relevant information in the process. Development of prediction and simulation models. Development of a digital twin to optimise key processes as realistically as possible.
Use case
Architecture for data capture in plant, sent to a central platform where it is stored. Generation of different prediction models that allow the generation of optimisation models of a key value of the process. Visualisation of relevant information to the operator to facilitate decision-making.
Infrastructure
On Premise and Cloud.
Technology
Deep learning
Data
Private data set of real time process values for each of the 4 use cases (real process industries)
Resources
Process technicians, industrial data capture and processing technicians, data engineers, data scientists, UX specialists.
Difficulties and learning
Complex processes, with variables that interact with one another, with a large component of human decision, very long inertia in the process, dependencies between processes, etc.
KPIs (business impact and metrics of the model)
Reduction of energy consumption, reduction of C02 emissions, reduction in scrap, processes and products out of quality.
Funding
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 869931.
Collaborators, Partners
14 partners from 8 countries, including: Acerálava, Ideko, Ingeteam, Savvy, etc.
