From fragmented data to actionable design: Physics-calibrated learning for plastic upcycling
2026-08-03 • Machine Learning
Machine LearningArtificial Intelligence
AI summaryⓘ
The authors created a new computer model called PC-MG-MoE to better understand how to recycle plastic waste through heat-based processes, even when some experimental data is missing or incomplete. Unlike other methods that ignore or guess missing data, their model learns from partial information and matches physical laws. It works well despite different labs using varied methods and offers explanations rather than just predictions. They tested it against real experiments and made it easy to use online for designing better recycling experiments and processes.
Thermochemical upgradingPlastic waste recyclingMissing dataMixture-of-Experts modelPhysics-based modelingCross-laboratory heterogeneityTarget imputationModel interpretabilityExperimental designData-driven modeling
Authors
Jingyang Bai, Zijia Wang, Xiangyi Long, Marcos Millan, Binjian Nie, Mingyue Ding
Abstract
Thermochemical upgrading of plastic waste is a key upcycling pathway, yet the experimental literature is fragmented by heterogeneous conditions and incomplete reporting. Complete-case learning would retain only 10.99% of the curated experiments, while target imputation can introduce biased supervision. Here we develop a Physics-Calibrated, Missingness-Gated, and Load-Balanced Mixture-of-Experts (PC-MG-MoE) framework that converts structured missingness into an informative learning signal. PC-MG-MoE learns directly from partially observed experiments without target imputation, reconstructs physically consistent product distributions, accommodates cross-laboratory heterogeneity, and provides interpretable model behaviour rather than black-box prediction alone. Under stringent source-grouped validation, it achieved the lowest aggregate absolute error among the evaluated models, supporting engineering screening under cross-laboratory heterogeneity. Wet-lab experiments provide an external comparison, showing key composition-dependent trends. Implemented as an interactive web-based workflow, PC-MG-MoE enables forward screening, physics-grounded constrained inverse design, targeted experimental planning that supports reduced experimental workload and trial-and-error, and laboratory-specific adaptation with new platform-specific data. This work establishes a transferable framework for converting fragmented literature data into experimentally actionable guidance for model-guided plastic upcycling and broader thermochemical systems.