PR-271-143716-R02 Bayesian Belief Network (BBN) Decision Support for Pipeline Third Party Interference
Bibliographic record
Abstract
Satellite monitoring offers unique advantages to the industry in meeting the objectives of managing third-party encroachment to mitigate the potential of mechanical damage. Satellite monitoring of third-party encroachment involves a persistent acquisition of satellite imagery over a pipeline right-of-way (ROW), combined with computerized change detection to identify potentially hazardous activities. Monitoring using satellite synthetic aperture radar (SAR) provides an all-weather day or night monitoring of a specific geographic location. This monitoring service can be enhanced with third-party information to increase the confidence in targets detected within satellite imagery. This information can also be used to reduce false positives. A simplistic example of this would be to use road location information overlaid with target information. A target found on a road, such as a tractor-trailer rolling down a highway, represents a small risk to a pipeline and subsequently can be given a lower risk or even be removed as a threat altogether. On the other hand, a large vehicle in a field near a pipeline and not on a road may represent a higher risk to a pipeline. The higher confidence data in-turn allows pipeline integrity operations staff to focus on the higher probability targets, saving time and resources, while maintaining safety standards. This concept has been implemented in the form of a Bayesian Belief Network (BBN) Decision Support System (DSS) that integrates with CalPoly's Representational State Transfer Access for Pipeline Integrity Database (RAPID). RAPID houses multiple data sources such as roads, utilities, agriculture, and construction information to increase target confidence. Both the BBN-DSS and RAPID were developed under the same DOT Cooperative Agreement (OASRTRS-14-H-CAL).
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 | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.188 | 0.074 |
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".