What a digital twin does
A digital twin is a virtual replica of a physical asset, kept in sync with live sensor data and models. NASA's definition: a multiphysics, multiscale, probabilistic simulation that mirrors the life of its physical counterpart. It tracks thousands of variables at high frequency, and it tells you what is happening.
It does not explain why. And it cannot say what would happen under a change the system has never seen.
What the causal layer adds
A causal digital twin adds the cause-and-effect structure, so it also tells you why, and it can simulate interventions.
That structure is a structural causal model: a directed acyclic graph showing which variables cause which, plus structural equations describing each link. The equations are one-way; they cannot be inverted like ordinary algebra.
Correlation has no direction; causation does. “X causes Y” means that, once confounders are accounted for, changing X changes Y, but changing Y does not change X.
Three levels of questions
- Association. “What if I see this sensor reading?” Standard digital twins and normal machine learning stop here.
- Intervention. “What if I do this?” Change an operating setting, perform a maintenance action.
- Counterfactual. “What if I had done something differently?” “Why did this specific failure happen?”
The causal digital twin is what reaches the second and third levels.
Two ways to build one
- With a digital twin: real system leads to digital twin (thousands of variables) leads to causal digital twin (only the decision-relevant variables).
- Without a digital twin: real system leads to causal digital twin directly, from the decision-relevant sensor data.
The two are complementary. The digital twin tracks thousands of variables at high frequency. The causal digital twin keeps only the decision-relevant causal links needed to simulate interventions.
So you do not need an existing digital twin to get value. If you have one, the causal twin sits on top of it.
Digital twins tell you what. Rhizant tells you why.