Applicability and interpretability of logical analysis of data in condition based maintenance
Bibliographic record
Abstract
his thesis studies the applicability and adaptability of a data mining artificial intelligence approach called Logical Analysis of Data (LAD) to diagnostic applications in Condition Based Maintenance (CBM). Most of the technologies used so far for decision support in CBM tend to automate the diagnostic process without offering any added knowledge that could be helpful to the maintenance operation and maintenance personnel. LAD possesses two key advantages over other decision making technologies used in CBM: (1) it is a non-statistical approach; as such no statistical assumptions are required for the input data, and (2) it generates interpretable patterns that could help solve maintenance problems. A study on the implementation of LAD in CBM is presented in this research whose objective are to study the applicability of LAD in different CBM situations requiring special considerations regarding the types of input data and maintenance decisions, adapt the LAD methodology to the particular requirements that arise from these applications, and improve the LAD methodology in line with the above two objectives in order to increase diagnosis accuracy and result interpretability. The novelty of the research presented in this thesis is (1) the application of LAD to CBM for the first time in applications that stand to benefit from the advantages that this technology provides; and (2) the innovative modifications to LAD methodology, particularly in the area of pattern generation, in order to improve its performance within the context of CBM. The research conducted in this thesis followed an evolutionary approach in order to achieve the objectives stated in the Introduction. The research applied LAD in three applications: (1) the detection of Rogue components within the spare part inventory of reparable components in a commercial airline company, (2) the detection and identification of faults in power transformers using DGA, and (3) the detection of faults in rotor bearings using vibration signals. This research concludes that LAD is a promising decision making approach that adds important benefits to the implementation of CBM in the industry.
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 | 0.014 | 0.050 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.001 | 0.004 |
| Scholarly communication | 0.006 | 0.005 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| 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".