Turning Data into Action: Reducing Risks from Offshore Well Piping Growth
Suradech Kongkiatpaiboon, K. Pakdeearporn
Offshore Technology Conference Asia (OTC Asia), Kuala Lumpur, Malaysia, 2026 · OTC-36360-MS
Hot production fluids make offshore well piping grow upward as the well heats, and on several infill wells the movement exceeded design stress limits, creating a loss-of-containment risk. Checking every well by hand offshore was costly, so this work built a data-driven model that predicts which wells are likely to grow too much. Engineers can then focus measurements and mitigation on the high-risk wells instead of over-designing every one.
Key results
- Best models predicted well growth with a mean absolute error of 15–30 mm and RMSE of 22–75 mm
- Back-testing showed a 70% reduction in offshore manual measurement workload
- Teams can prioritise high-risk wells and safely deprioritise low-risk ones, reducing offshore exposure
- Well growth correlated strongly with high-porosity formations and deeper vertical depths, informing future piping design flexibility
Approach
- Compiled elevation (growth) measurements from more than 100 wells
- Normalised surface production data to each well's own conditions
- Tested multiple regression models and selected the best performers for predictive reliability
- Added statistical margins and confidence intervals to turn predictions into a probability-based risk model
- Back-tested against historical operations before deploying into routine monitoring
The problem
Thermal growth of offshore well piping reached up to 12 inches, beyond stress limits, and manual monitoring of every well was resource-intensive.
Tools & methods
Summary written for this site from the published paper. For figures, data and the full method, read the paper on OnePetro.