Trusting the Call: Advanced Analysis Techniques for ILI Data in the Age of AI
Course Overview
Pipeline operators and inspection providers can generate detection and sizing results faster than ever — but detecting a signal and trusting a call remain two different problems, and the second is getting harder, not easier, as AI tools spread through the ILI data-evaluation workflow. This course is for analysts, engineers, and operator personnel who work with in-line inspection results — whether producing them or acting on them — and need to move from reading a result to defending a decision.
Why does training still matter when technical information is one AI query away? Starting from this direct question, the course builds on the statistical and procedural foundations: probability of detection and identification, sizing accuracy and confidence, and the three-tier validation framework in API 1163. It connects that foundation to advanced ILI, AI-supported data processing and analysis and to the regulatory and data-governance landscape areas (incl. EU AI Act). Furthermore, geohazards and its monitoring are addressed.
Format: one intensive day, judgment-focused, concluding with a hands-on exercise designed to translate course insights into daily operational decisions.
Agenda
- Introduction and the “WHY”: The shift from information retrieval to human judgment and verification in an AI-driven industry.
- Foundations of Anomaly Interpretation: Physical characteristics of dents, metal loss, and cracks; raw sensor data stream ingestion, data corrections, noise filtering, and an easy introduction to AI.
- Applied Statistics and API Standard 1163: POD, POI, sizing accuracy; applying these metrics directly to validation digs and unity plots.
- Advanced Data Analysis & Processing: Data-alignment methodologies, basic concepts of Geometry/MFL/UT/EMAT processing, HMI workflows (Human-in-the-Loop HITL), and the state-of-the-art in multi-technology data fusion.
- Geohazards & IMU Curvature Mapping: Mitigating geotechnical/hydrotechnical threats; leveraging Inertial Measurement Unit (IMU) bending-strain analysis for active pipeline displacement monitoring.
- Regulatory Compliance & AI Governance: Pipeline data-lineage under the PHMSA Mega Rule and the legal impact of the EU AI Act’s Article 4 “AI Literacy” mandate.
- Practical Capstone Workshop: single hands-on group exercise which helps translating course insights into daily operational decisions.
Approval (for German engineers)
This training course is recognized as a training session in accordance with §6 FuWo. Consulting engineers and members of the German Chambers of Engineers receive 8 training points for participation.
Target Group
Pipeline integrity engineers, ILI data analysts, and technical managers who evaluate, verify, or oversee ILI-based anomaly and integrity decisions — together with pipeline operator personnel who need a working understanding of how ILI data analysis results are produced and how much to trust them. It assumes basic familiarity with inspection technologies and metal-loss/crack/dent threats, as e.g. given by the EITEP Training Courses of M. Beller and K. Reber. Suited to anyone who must trust, or defend, an ILI-based call and who also sees the growing importance of data analysis and AI-supported technologies in the future.
Dr. Dirk Maiwald is an independent consultant and senior technology executive with 35+ years of professional experience in electrical engineering, signal theory, and critical infrastructure safety. He holds a Doctor of Engineering in Electrical Engineering and Signal Theory from Ruhr-University Bochum and completed executive qualifications in Machine Learning in Business at the MIT Sloan School of Management.
During a distinguished 15-year tenure at the ROSEN Group and its technology spin-off Rosenxt, Dr. Maiwald served in key global leadership roles, including Managing Director in Lingen, Germany. In these capacities, he oversaw multidisciplinary R&D for oil and gas inspection technologies, and built dedicated teams focused entirely on AI innovation projects. His prior industrial experience includes pioneering fiber-optic offshore temperature and vibration monitoring applications for offshore power cables and leading major naval sonar technology development programs.
Throughout his career, Dr. Maiwald has focused on translating complex sensor physics and signal streams into reliable, real-world systems. He believes that project and business success is always a fruitful combination of people, teams, and technology. Culture is the main ingredient especially in the age of AI. An overview of his work can be found at www.confidence-solutions.com.
Seminars Contact
