Electronic Thesis/Dissertation


Textual Analysis of Defects in Engineering Documentation via a Systematic Execution Framework Open Access

Downloadable Content

Download PDF

This dissertation focuses on the Requirements Engineering (RE) vital role in systems development; it is the preparatory point and critical foundation of the overall engineering actives and processes. This praxis explores the understanding of the RE process, and its influence across the System Life Cycle (SLC) stages, utilizing a quantitative empirical methodology to demonstrate the effectiveness of stakeholder decision value in the overall success of Systems Engineering (SE). This praxis proposes a framework for structuring decisions, based upon the INCOSE Decision Management Process, which improves stakeholder confidence during system developmental stages. This praxis also introduces and applies the Nexus Engineering Execution Generation (NEX GEN) framework, a novel method for systematic textual analysis of requirements documentation as a prediction of system performance with an iterative decision-making process.Application of this framework defines requirements in a quantitative manner to achieve mutual stakeholder agreement of the requirements. Regression analysis indicated a statistically significant relationship between the predictor variable of ambiguity (linguistic attribute) and the response variable of system performance (usability metric). This study adds value to the engineering discipline by providing a framework, metrics and decision structure as a complementary method for development of requirements in a theoretical manner for the reflective practitioner. This framework is applied through federal technology systems program case studies. This capability provides the organization an opportunity to improve the quality of requirements by enabling a prediction of their impact on our expected systems’ performance.

Author Language Keyword Date created Type of Work Rights statement GW Unit Degree Advisor Committee Member(s) Persistent URL