XTRACTIS FOR predictive maintenance

Prediction of the Rupture of a Flexible Underwater Pipe

Benchmark vs. Logistic Regression, Random Forests, Boosted Trees & Neural Networks

Design an AI-based decision system that accurately predicts the upcoming risk of underwater pipes rupture considering the apparent complexity of the phenomenon, to plan rational maintenance operations.

Goals & benefits

Identify the predictors involved in the rupture of a pipe and enhance technical knowledge by helping petroleum industry engineers understand the causal relationships between these predictors, their combination, and the rupture.

Find the truly influential parameters for assessing the state of the pipe and thus reduce measurement and maintenance costs.

Carry out maintenance action specific for each pipe in order to avoid critical damage, thanks to rapid and transparent decisions.

  • The top-model is a decision system composed of 27 gradual rules without chaining, aggregated into 2 disjonctive fuzzy rules.
  • Each rule uses from 1 to 14 predictors among the 20 variables that XTRACTIS identified as significant (out of the 74 ones characterizing each pipe).
  • Only a few rules are triggered at a time to compute the decision.

It has a good Real Performance (on unknown data).

It computes real-time predictions up to 70,000 decisions/second, offline or online (API).

UC13 scores graph
LoR=Logistic Regression
RFo=Random Forests
BT=Boosted Trees
NN=Neural Networks

Detailed results and explanations in full document

Use Case 2024/02 (v3.0)

Powered by XTRACTIS® REVEAL v12.2.43841 (2023/01)


  1. Problem Definition
  2. XTRACTIS-induced Decision System
  3. XTRACTIS Process
  4. Top-Model Induction
  5. Explained Predictions for 2 unkown cases
  6. Top-Models Benchmark
  7. Appendices