Quality Assessment of Trustworthiness of AFMC Acquisition Data

Authors: Herman L. Karhoff; MITRE CORP MCLEAN VA

Abstract:

To become a more efficient and effective joint-expeditionary force, the Air Force (AF) and its business partners are adopting an enterprise view that optimizes resources (i.e., people, process, and technology). To achieve this end, the AF has embarked upon an aggressive enterprise Information Technology (IT) modernization strategy. A major challenge with the planning and implementation of transformation/migration strategies is the ability to determine the quality of AF data. A basic premise is that data of unknown quality is inherently untrustworthy. Few would disagree with the premise that good quality data (i.e., timeliness and accuracy) is critical to aiding AF leadership in making the right decisions. How can these decision makers trust the data if they do not have a means to assess and measure their data quality? Is it impossible to measure something that is not understood, and how do you manage something that cannot be measured? These are some of the key questions MITRE seeks to answer in this Mission Oriented Investigation and Experimentation (MOIE) initiative. The purpose of this paper is two-fold: (1) to heighten awareness on the importance and impacts of data quality (DQ); and (2) to document (i.e., DQ assessment methodology, measurement techniques and assessment criteria) the findings and outcomes from this MOIE research. The approach is to apply semantics and heuristics (i.e., utilization of architectures, methodologies, state-of-the art software tools, implementation techniques, and production test data) in exploring capabilities that enhance data quality and thus improve the quality of decisions made with enterprise data. AF production data (e.g., invoice transactions) was used to verify and validate the MOIE hypothesis, assumptions, and findings.

Limitations:

APPROVED FOR PUBLIC RELEASE