An AI-augmented cross-platform application for enhanced operator training and process optimization in thermoforming
Bibliographic record
Abstract
Thermoforming is a manufacturing process in which thermoplastic sheets are heated and shaped over a mould to produce desired geometries. Traditionally, given a mould shape, determining the optimal heaters configurations to achieve accurate temperature distributions has relied heavily on trial-and-error methods, resulting in inefficiencies in both time and resource usage. This thesis presents an interactive operator training application that integrates predictive neural networks with a series of feedback-providing tools in order to assist in determining an optimal heaters setting in thermoforming based on the expert-specified temperature distribution target. Users and learners can interactively sketch their desired thermal distribution on a virtual canvas, and the application computes and visualizes both the corresponding heater configuration and the resulting temperature profile in real-time. This system combines two neural network architectures: a Backward Convolutional Neural Network (CNN) that determines proper heater power settings based on the thermal targets (R² up to 0.9650) and a Forward Fully Connected Neural Network (FCNN) that predicts temperature distributions from power input configurations with high predictive accuracy (R² > 0.97). Both models were trained on synthetic datasets generated from a calibrated lab-scale simulation environment. These models are embedded within a user-friendly, Unity-based application that runs across multiple platforms, such as tablet or mobile, enabling continuous, interactive exploration of heat transfer behaviour in a digital twin environment. To evaluate the educational effectiveness of the developed application, we conducted a user study with participants from engineering and computer science backgrounds to evaluate the educational effectiveness of the developed application. The accuracy of task performance increased significantly when interacting with the predictive module (from 80.6% to 92.10%, p < 0.001, Cohen’s d = 1.27), indicating the application’s potential for conceptual learning and skill transfer. Furthermore, no differences in performance were found according to academic background, prior thermoforming skill, or immersive tool experience, indicating that the application ensures equitable learning for a variety of user groups. By equipping operators with intelligent tools that reduce reliance on costly physical prototyping and accelerate decision-making, this work contributes to the goals of Industry 5.0, promoting sustainable, human-centric manufacturing through emerging digital innovations.
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) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.023 | 0.004 |
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".