GearXAI

An Explainable Neuro-Symbolic Gearbox Fault Diagnosis Challenge

Bremen, Germany · August 15-21, 2026 · Online competition in advance

Read the abstract View timeline

Task: 9-class fault diagnosis
Input: 8-channel vibration windows
Artifact: ONNX, CPU-only
Ranking: Explainability first

Competition abstract

Transparent diagnostics for rotating machinery

Title: GearXAI: An Explainable Neuro-Symbolic Gearbox Fault Diagnosis Challenge
Website: https://gearxai-ijcai-ecai2026.pages.dev/

GearXAI is a single-track IJCAI-ECAI 2026 competition focused on explainable and neuro-symbolic approaches for multiclass gearbox fault diagnosis from vibration time series. Participants submit ONNX models that are evaluated in CPU-only, containerized runs for reproducible inference. Submissions must meet a minimum predictive-performance requirement, after which ranking is determined by an automated explainability score that quantifies faithfulness under controlled perturbations and stability under small input changes. The challenge will release a realistic drivetrain dataset, an ONNX-exportable baseline, and an official evaluator to enable transparent comparison and follow-up research.

Competition task

Predict the fault. Explain the evidence.

Drivetrain data

Participants work with synchronized multichannel signals from the Drivetrain Dynamics Simulator testbed, sampled at 5120 Hz across multiple speed-load regimes and fault states.

Shared label space

Each window is classified into one of nine classes: healthy operation plus eight fault types, aggregated across planetary and parallel gearbox domains.

Standardized tensors

The official loader outputs windows shaped [N, 8, 100]. Models return class probabilities [N, 9] and relevance maps [N, 8, 100] for XAI scoring.

Evaluation

Accuracy opens the gate. Explanations decide the rank.

XAI

Explainability Score

Faithfulness: 40%
Mechanistic relevance: 40%
Simplicity: 20%

Valid submissions must reach at least 80% macro-F1 on the hidden test set. Models that pass the gate are ranked by an automated Explainability Score combining deletion/insertion faithfulness, physically plausible relevance, stability under small perturbations, and compactness.

Final submissions are single ONNX models evaluated with ONNX Runtime.
Organizer evaluation runs offline in a sealed, CPU-only container with no internet access.
Determinism, runtime, memory, and explanation sanity checks are part of the evaluation process.

Timeline

Competition milestones

April 27, 2026 Official competition announcement submitted to IJCAI-ECAI
Late April / early May 2026 Call for Participation, baseline, and development evaluator release
Through June 30, 2026 Development leaderboard and daily submission window
July 15, 2026 Final ONNX submission deadline, 23:59 AoE
July 25, 2026 Winner notifications
August 15-21, 2026 Competition session at IJCAI-ECAI 2026 in Bremen

Resources

Competition resources

Dataset

Drivetrain vibration windows, metadata, and official preprocessing loader.

Coming soon

Baseline

ONNX-exportable neuro-symbolic baseline with published reference metrics.

Coming soon

Evaluator

Reproducible scoring code for macro-F1, faithfulness, stability, and simplicity.

Coming soon

Leaderboard

Development leaderboard for transparent comparison before final evaluation.

Coming soon

Cite us

Referencing the GearXAI challenge

If GearXAI is useful for your work, please cite the challenge website until the official competition proceedings or archival report is available.

@misc{hogea2026gearxai,
  title        = {GearXAI: An Explainable Neuro-Symbolic Gearbox Fault Diagnosis Challenge},
  author       = {Hogea, Eduard and Onchis, Darian M. and Ivascu, Todor and Yan, Ruqiang},
  year         = {2026},
  note         = {IJCAI-ECAI 2026 Competitions and Challenges Track},
  url          = {https://gearxai-ijcai-ecai2026.pages.dev/}
}

Organizers