Skip to Content

See what you missed at the Causal AI conference

View site

Root Cause Analysis

Causal AI enables a complete root cause analysis of your business to provide organizations and their leaders with a clear understanding of what causes delays and recommends actions to improve on-time and in-full service levels.

Overview

Find and fix root causes rapidly.

As the name suggests, causal methods are perfectly placed to enable you to identify the true root cause drivers in your business challenges:

  • Unique causal approaches ensure that true root causes are identified so you can focus on diagnosis without distraction from spurious correlations.
  • Seeing is believing. Causal graphs allow visual representation enabling you to tackle complex real world challenges with large numbers of dependencies.
  • Built to scale to meet the demands of enterprise workloads from factory floors to global supply chains.
from root_cause_analysis import CounterfactualRCA

rca = CounterfactualRCA(causal_model=causal_model)
rca_result = rca.get_root_causes(data=data, target='Y', event=event)
Unique Causal Approach

Cut through the noise, and get to the root cause.

Traditional, correlation-based approaches to root cause analysis fail to provide accurate and robust results. Correlation-based approaches cannot capture the fact that even small changes in the input can result in significant changes to the results.

Root cause analysis (RCA) with decisionOS uses two causality powered approaches to overcome these challenges:

  • Interventional RCA: This method forces the value of variables to change and observes how this propagates to other variables, allowing outlying events to be detected and its causes identified.
  • Counterfactual RCA: Take a given outlying event and alter the variable values independently of the others to determine which variables have the largest influence on the outcome, thus finding the root causes.
Causal Graphs

See root causes instantly and intuitively.

Unlike other approaches decisionOS’ root cause analysis builds atop of causal graphs. Complex challenges are readily visualised and reasoned about. Root cause paths can be highlighted directly in the causal graph making further analysis straightforward.

Built for Scale

Battle hardened root cause identification.

decisionOS’ root cause analysis has been applied to a variety of diverse use cases from manufacturing optimization to supply chain management.

The methods have been scaled to real world data with hundreds of features meaning you can seamlessly apply them to your toughest challenges.

partner