Project Details
Description
One of the innovative steps taken by Cognitive Geology in the area of Oilfield Evaluation area is to create the CDF of economic outcomes weighted by quality. This project proposes to improve the quality metric and in the process innovate the preservation and automatic use of expert Geological/Geophysical opinion.
Assessing model quality is of critical importance to the creation of the cumulative density function that is used by Cognitive Geology to evaluate likely economic outcomes. Beyond the statistical measures of the quality of fit, there are also expert opinions. Expert opinions are very expensive to obtain, often use tacit knowledge and mathematically are often not wholly supportable from the apparent evidence. Capturing and automating the application of the expert opinions is therefore an important objective within the established oil industry and also within the development of the cognitive technology aspects of the product range being developed by Cognitive Geology.
Traditional Oilfield evaluation practises use expert knowledge. The group of Geologists who are skilled in the art are rapidly reaching the end of their careers, and as they leave with them leaves the skill base and the knowledge that took decades to acquire. Evidence based Probabilistic Oilfield evaluation is proving itself increasingly valuable, but is traditionally mathematically derived with little consideration of the Geological expert’s opinions, the automatic inclusion of which has up to now been impossible.
This project proposes the creation of a rigorous method by which relevant knowledge possessed by an expert might be preserved for future use in an automated system.
Assessing model quality is of critical importance to the creation of the cumulative density function that is used by Cognitive Geology to evaluate likely economic outcomes. Beyond the statistical measures of the quality of fit, there are also expert opinions. Expert opinions are very expensive to obtain, often use tacit knowledge and mathematically are often not wholly supportable from the apparent evidence. Capturing and automating the application of the expert opinions is therefore an important objective within the established oil industry and also within the development of the cognitive technology aspects of the product range being developed by Cognitive Geology.
Traditional Oilfield evaluation practises use expert knowledge. The group of Geologists who are skilled in the art are rapidly reaching the end of their careers, and as they leave with them leaves the skill base and the knowledge that took decades to acquire. Evidence based Probabilistic Oilfield evaluation is proving itself increasingly valuable, but is traditionally mathematically derived with little consideration of the Geological expert’s opinions, the automatic inclusion of which has up to now been impossible.
This project proposes the creation of a rigorous method by which relevant knowledge possessed by an expert might be preserved for future use in an automated system.
Key findings
1. Design of elicitation protocol. This activity will be the result of a thorough literature review, secondary data analysis, discussions with industrial partner Cognitive Geology and preliminary testing on experts.
2. Design of a test case on which to apply protocol with experts.
3. Empirical testing of protocol. A pilot test will be conducted as well as a fuller evaluation on a pool of relevant experts.
4. Analysis of test results and revision of process.
2. Design of a test case on which to apply protocol with experts.
3. Empirical testing of protocol. A pilot test will be conducted as well as a fuller evaluation on a pool of relevant experts.
4. Analysis of test results and revision of process.
| Status | Finished |
|---|---|
| Effective start/end date | 1/11/16 → 31/03/17 |
Keywords
- elicitation
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.