Faulty Detection System Based on SPC and Machine Learning Techniques
Bibliographic record
Abstract
Starting from a worrying observation, that companies have difficulties controlling the anomalies of their manufacturing processes, in order to have a better control over them, we have realized a case study on the practical data of the Fertial Complex to analyze the main parameters of the ammonia neutralization by nitric acid process. This article proposes a precise diagnostic of this process to detect dysfunction problems affecting the final product. We start with a general diagnosis of the process using the SPC method, this approach is considered an excellent way to monitor and improve the product quality and provides very useful observations that allowed us to detect the parameters that suffer from problems affecting the quality. After the discovery of the parameters incapable to produce the quality required by the standards, we applies two machine learning technologies dedicated to the type of data of these parameters for detected the anomaly, the first technique called The kernel connectivity-based outlier factor (COF) algorithm consists in recording for each object the degree of being an outlier, the second technique called the Isolation Forest, its principle is to establish a forest to facilitate the calculation and description. The results obtained were compared in order to choose which is the best algorithm to monitor and detect the problems of these parameters, we find that the COF method is more efficient than the isolation forest which leads us to rely on this technology in this kind of process in order to avoid passing a bad quality to the customer in future.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 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) | 0.001 | 0.001 |
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".