Implementing a relational database for an accelerated-life-test facility

This paper describes the design of a relational database and associated command structure for an accelerated-life-test facility. The primary goals of the design were to: (1) improve data integrity, and (2) provide a complete and consistent historical record of reliability test results. The intended reader is the engineer who wants to understand the issues involved in automating the reliability-data collection and reporting process. The paper: (1) shows (via the story of a particular reliability database design) in a tutorial fashion how to use system requirements and the formalisms of the relational model to create a database system for engineering tests; and (2) serves as a guide for developing relational database applications for other engineering and production activities. The paper makes three primary contributions by showing: (a) a practical method for constructing a relational database. The method is based on an IDEFO representation of the testing process, which is easier for the database novice to understand and implement than the usual approaches of data flow or entity-relationship diagrams. (b) The value of a particular database structure for maintaining engineering test database integrity: a table for test measurements, a table for kinds of test measurements, and a table for test measurement procedures. (c) How to write data entry/reporting application programs that maintain data integrity and ease the user's input burden. While specific issues in reliability testing are discussed, the methodology to construct the database and user interface applies broadly.<
>