Multimodal LSTM with Data Augmentation for Predicting Rubber Compound Cure Properties in Industrial Batch Processes
Bibliographic record
Abstract
Predicting scorch time ( t 5 ) of rubber compounds is critical in industrial polymer manufacturing to avoid premature vulcanization and ensure product quality. This study presents an LSTM-based multimodal deep learning model to forecast t 5 by integrating online time-series plant sensor data with scalar process parameters from the mixing operation. Key preprocessing steps, including feature normalization and interpolation, were applied to harmonize disparate data sources. To address a highly imbalanced distribution of scorch times in the data set, a synthetic data augmentation strategy was introduced, effectively expanding underrepresented t 5 ranges without additional experiments. The augmented LSTM model achieved high predictive accuracy, outperforming baseline models and maintaining robust performance even for extreme scorch time values. The model was validated using online data from a full-scale industrial rubber compounding plant. Results demonstrate that this multimodal augmented modeling approach substantially improves t 5 prediction, highlighting its potential for batch-end process monitoring and subsequent adjustment of process conditions. Implementing the proposed model in an industrial setting can enable proactive adjustments for future batches during compounding, thereby reducing scrap, enhancing safety, and ensuring consistent product quality in chemical manufacturing processes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".