A Deep-Reinforcement-Learning-Based Digital Twin for Manufacturing Process Optimization
Bibliographic record
Abstract
In the context of Industry 4.0 and smart manufacturing, production factories are increasingly focusing on process optimization, high product customization, quality improvement, cost reduction, and energy saving by implementing a new type of digital solutions that are mainly driven by Internet of Things (IoT), artificial intelligence, big data, and cloud computing. By the adoption of the cyber–physical systems (CPSs) concept, today’s factories are gaining in synergy between the physical and the cyber worlds. As a fast-spreading concept, a digital twin is considered today as a robust solution for decision-making support and optimization. Alongside these benefits, sectors are still working to adopt this technology because of the complexity of modeling manufacturing operations as digital twins. In addition, attempting to use a digital twin for fully automatic decision-making adds yet another layer of complexity. This paper presents our framework for the implementation of a full-duplex (data and decisions) specific-purpose digital twin system for autonomous process control, with plastic injection molding as a practical use-case. Our approach is based on a combination of supervised learning and deep reinforcement learning models that allows for an automated updating of the virtual representation of the system, in addition to an intelligent decision-making process for operational metrics optimization. The suggested method allows for improvements in the product quality while lowering costs. The outcomes demonstrate how the suggested structure can produce high-quality output with the least amount of human involvement. This study shows how the digital twin technology can improve the productivity and effectiveness of production processes and advances the use of the technology in the industrial sector.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".