A Digital Twin Can Reproduce the Asset and Still Change Nothing.
15 MINUTE READING TIME
Industrial organisations are often asked how much of a plant they want to replicate in a digital twin. It is usually the wrong first question. The better starting point is the decision.
This Executive Guide sets out how to design industrial digital twins around decisions, workflow and evidence, and what that means in aluminium reduction: where to start, how to validate a pilot, and why each step towards autonomy has to be earned.
“Fidelity follows the decision, rather than becoming an objective in its own right.”
The design principle at the centre of this guide, and the reason scope matters more than replication.
The guide is for smelter leaders and reduction specialists deciding where a digital twin earns its place.
An idoba Executive Guide
A Digital Twin is Only as Good as the Decision it Changes.
idoba is part of Perenti. The contact information you provide to idoba will be used to email you this guide and related idoba.sim material. Every email has one-click unsubscribe. We do not sell or share your details. For information about our privacy practices and commitment to protecting your privacy, check out our Privacy Policy.
INSIDE THE GUIDE
What You’ll Learn.
Four Sources of Operational Knowledge.
What physics-based models, machine learning and operator expertise each contribute, and where each one can fail when treated as a substitute for the others.
Build the Minimum Viable Digital Twin.
The smallest representation that can be tested against a valuable decision, and eight validation tests a pilot should pass beyond model accuracy.
What This Means in Aluminium.
Why cathode and sidewall risks are modelled separately, and how earlier insight widens the response window for an individual cell.
Start With the Decision Boundary.
Six questions that expose whether a proposed digital twin has a credible route to value, before any modelling approach is chosen.
Autonomy is a Transfer of Decision Rights.
A staged pathway from predictive decision support towards automation, and the evidence each step must earn before authority moves to the system.
Five Tests Before Funding a Digital Twin.
Decision, representation, action, validation and governance: the five questions every proposal should answer before significant investment.
Meet the Authors.
Smelting experience, data science, and enterprise delivery, brought together by the idoba.predict.
Aaron Carter
Chief Development Officer, idoba,
Leads strategic growth, governance and enterprise development, with a focus on how advanced models turn into real decisions on site.
Grant Pretorius
Head of Technical Solutions, idoba.
More than 15 years in technical, commercial and operational smelting roles across the primary electrolysis value chain, in three countries.
Dr Nicolas Small
Data Science and Modelling Lead, idoba.
More than ten years in data science, robotic control, industrial IoT, and applied machine learning and optimisation.
ABOUT IDOBA
The Experience of Operators, Encoded in Software.
We build purpose-built industrial AI and simulation products for mining and metals. idoba combines operational experience with site data and specialist models to bring critical operating decisions forward, giving site teams more time to investigate, compare, and act while they can still change the outcome.