--> Abstract: The EPMI Malay Basin Petroleum Geology Database: Design Philosophy and Keys to Success, by H. E. Low, S. Creaney, L. H. Fairchild, M. H. Feeley, and P. S. Koch; #90982 (1994).

Datapages, Inc.Print this page

Abstract: The EPMI Malay Basin Petroleum Geology Database: Design Philosophy and Keys to Success

Hooi Ean Low, Steve Creaney, Lee H. Fairchild, Missy H. Feeley, Phillip S. Koch

Esso Production Malaysia Inc. (EPMI) developed and populated a database containing information collected in the areas of basic well data: stratigraphy, lithology, facies; pressure, temperature, column/contacts; geochemistry, shows and stains, migration, fluid properties; maturation; seal; structure. Paradox was used as the database engine and query language, with links to ZYCOR ZMAP+ for mapping and SAS for data analysis. Paradox has a query language that is simple enough for users. The ability to link to good analytical packages was deemed more important than having the capability in the package.

Important elements of design philosophy were included: (1) information on data quality had to be rigorously recorded; (2) "raw" and "interpreted" data were kept separate and clearly identified; (3) correlations between rock and chronostratigraphic surfaces were recorded; and (4) queries across technical boundaries had to be seamless.

The key to success included the blending of geologic and computing expertise and having a multidisciplinary dedicated staff to see that the database was fully developed, populated, and used. Selecting the "correct" data to collect was critical. The specialized expertise of the database designers minimized problems. Data selection was driven primarily by the questions to be answered, and the need to minimize significant bias in sample data. Anticipating questions enabled linkages between data tables to be made early. Early pilot tests also tested the design and linkages, leading to reevaluation of data collection strategy. It was crucial to check validity to ensure consistency during data entry so that data was linked and sorted properly. Finally, sizing tables properly improved databa e efficiency.

AAPG Search and Discovery Article #90982©1994 AAPG International Conference and Exhibition, Kuala Lumpur, Malaysia, August 21-24, 1994