Datamining turbofan engine performance to improve fuel efficiency
Bibliographic record
Abstract
Snecma as major engine manufacturer is often asked by its airline customers to help them improve their efficiency in the operation of the engines. The main concern is fuel consumption but also a long-term expectancy about the engine cost during its full life. For example, this includes maintenance frequency and shop costs. Snecma built a new laboratory for data analysis who's mission is to find numeric models able to take most of the manufacturer knowledge into account when observing large amounts of data coming back from the aircrafts' operations. In the next five years, we need to be able to collect more than one gigabyte of data per flight and per engine. This becomes huge when looking at the big fleet of CFM engines, which already accounts for almost 30000 turbofans, with a new flight beginning every 2.5 seconds. This flow of data continuously increases and should be used to help our customers improve their operations. As manufacturer, we also need to help our maintenance teams improve the reliability of the systems and optimize the shop operations with all information available in the data and even the design of future engines. In this paper, we outline our data laboratory, the way we take care of all sources of information including operations, shops and also production, integration, tests and external observations like weather, airports data, etc. As an example, we present the statistics we get from the analysis of the aircraft fuel consumption during climb. To be able to interpret the flight data we need to get rid of all external conditions that may bias the data. The algorithms we use for this process are the same as the ones we implemented in our prognostic and health-management (PHM) system. They are used on ground stations to compile models that may eventually be reworked to create new embedded health indicators.
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 | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
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".