Adaptive Conformal Prediction for Full-Scale Plant pH Forecasting under Process Drift

Jongyeop Lee, Jong Min Lee
Proceedings of the Fifteenth Symposium on Conformal and Probabilistic Prediction with Applications, PMLR 329:1043-1045, 2026.

Abstract

Soft sensors in full-scale wastewater treatment face operating drift; overconfident point forecasts can mask compliance risk or force overly conservative chemical dosing. We evaluate conformal prediction intervals for 10-minute-ahead reactor pH forecasting on historical data from a semiconductor wastewater plant: a residual MLP trained on one source unit is evaluated offline across four anonymized units over multiple operating months. Shift diagnostics relative to a source reference month show strong covariate separability (domain-classifier AUC) and large increases in relative MAE (RMAE), indicating model-error-scale drift under real operation. Comparing split, weighted, and online adaptive (ACI) conformal methods and an EnbPI-style rolling residual interval at nominal 90% coverage, we find static and covariate-weighted intervals undercover shifted regimes while online methods recover near-nominal coverage at moderate width cost.

Cite this Paper

BibTeX
@InProceedings{pmlr-v329-lee26a, title = {Adaptive Conformal Prediction for Full-Scale Plant pH Forecasting under Process Drift}, author = {Lee, Jongyeop and Min Lee, Jong}, booktitle = {Proceedings of the Fifteenth Symposium on Conformal and Probabilistic Prediction with Applications}, pages = {1043--1045}, year = {2026}, editor = {Ahlberg, Ernst and Johansson, Ulf and Boström, Henrik and Carlevaro, Alberto and Hallberg Szabadváry, Johan and Carlsson, Lars}, volume = {329}, series = {Proceedings of Machine Learning Research}, month = {02--04 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v329/main/assets/lee26a/lee26a.pdf}, url = {https://proceedings.mlr.press/v329/lee26a.html}, abstract = {Soft sensors in full-scale wastewater treatment face operating drift; overconfident point forecasts can mask compliance risk or force overly conservative chemical dosing. We evaluate conformal prediction intervals for 10-minute-ahead reactor pH forecasting on historical data from a semiconductor wastewater plant: a residual MLP trained on one source unit is evaluated offline across four anonymized units over multiple operating months. Shift diagnostics relative to a source reference month show strong covariate separability (domain-classifier AUC) and large increases in relative MAE (RMAE), indicating model-error-scale drift under real operation. Comparing split, weighted, and online adaptive (ACI) conformal methods and an EnbPI-style rolling residual interval at nominal 90% coverage, we find static and covariate-weighted intervals undercover shifted regimes while online methods recover near-nominal coverage at moderate width cost.} }
Endnote
%0 Conference Paper %T Adaptive Conformal Prediction for Full-Scale Plant pH Forecasting under Process Drift %A Jongyeop Lee %A Jong Min Lee %B Proceedings of the Fifteenth Symposium on Conformal and Probabilistic Prediction with Applications %C Proceedings of Machine Learning Research %D 2026 %E Ernst Ahlberg %E Ulf Johansson %E Henrik Boström %E Alberto Carlevaro %E Johan Hallberg Szabadváry %E Lars Carlsson %F pmlr-v329-lee26a %I PMLR %P 1043--1045 %U https://proceedings.mlr.press/v329/lee26a.html %V 329 %X Soft sensors in full-scale wastewater treatment face operating drift; overconfident point forecasts can mask compliance risk or force overly conservative chemical dosing. We evaluate conformal prediction intervals for 10-minute-ahead reactor pH forecasting on historical data from a semiconductor wastewater plant: a residual MLP trained on one source unit is evaluated offline across four anonymized units over multiple operating months. Shift diagnostics relative to a source reference month show strong covariate separability (domain-classifier AUC) and large increases in relative MAE (RMAE), indicating model-error-scale drift under real operation. Comparing split, weighted, and online adaptive (ACI) conformal methods and an EnbPI-style rolling residual interval at nominal 90% coverage, we find static and covariate-weighted intervals undercover shifted regimes while online methods recover near-nominal coverage at moderate width cost.
APA
Lee, J. & Min Lee, J.. (2026). Adaptive Conformal Prediction for Full-Scale Plant pH Forecasting under Process Drift. Proceedings of the Fifteenth Symposium on Conformal and Probabilistic Prediction with Applications, in Proceedings of Machine Learning Research 329:1043-1045 Available from https://proceedings.mlr.press/v329/lee26a.html.

Related Material