$2.6 Million Grant to PRRC for "Fuzzy Expert" System

SOCORRO, N.M., Oct. 13, 1998 -- New Mexico Tech's Petroleum Recovery Research Center (PRRC) has been awarded more than $2.6 million by the U.S. Department of Energy (DOE) to develop a Fuzzy Expert Exploration Tool, or FEE Tool.

The FEE Tool will be an artificial intelligence system that will integrate available information on petroleum reservoirs and basins with advanced computing techniques to improve the chances of locating oil, better assess the risks of prospecting and drilling for oil, and, ultimately, reduce the costs of oil exploration. The tool will use "fuzzy logic," a modern computational approach for problem solving by integrating large amounts of inexact, incomplete information.

The FEE Tool will emulate the decision-making process of a "fuzzy expert." In other words, it will emulate the decision-making processes of oil exploration experts who routinely deal with incomplete information and frequently produce predictions with varying degrees of certainty and confidence.

"Expert systems are not yet widely used in the oil and gas business, although expert system tools currently are being used successfully in fields such as medicine, manufacturing, and mineral exploration," says William W. Weiss, a PRRC field petroleum engineer and the DOE project's principal investigator.

Weiss will be joined in the $2.6 million interdisciplinary research project by Ronald Broadhead, senior petroleum geologist and assistant director of the New Mexico Bureau of Mines and Mineral Resources; Andrew H. Sung, associate professor of computer science at New Mexico Tech; and other researchers under the direction of the PRRC's Reservoir Evaluation and Advanced Computational Technologies (REACT) group.

The FEE Tool will be able to correlate known measurements, such as aeromagnetic surveys and regional structure with sparse data sets, such as mapped oil shows, well logs, oil well production history, and seismic information.

"Except for seismic information, much of the required database is in the public domain," Weiss points out. "Therefore, there is an urgent need to develop a mechanism to promote the exchange of proprietary seismic information among New Mexico oil and gas operators."

Once developed, the FEE Tool will be able to enhance decision-making for both large and small oil and gas operators through custom configurations and will be made available to users on the Internet.