Empirical research in systems engineering: challenges and opportunities of a new frontier

Abstract

This paper aims to advance the pedagogy of systems engineering by identifying opportunities and challenges in empirical research in the field. After an introduction to how empirical research could be further utilized in systems engineering, this paper discusses challenges faced when conducting empirical research in the field, threats to validity associated with systems engineering data collection, and considerations for empirical mixed-methods research. Two recently completed systems engineering empirical studies are used to illustrate specific examples. Finally, suggestions are given on how a professional society might provide additional support for researchers completing empirical research in systems engineering. The overarching goal of this paper is to describe how the increased use of empirical methods can be used to enrich the quality of research results which will in turn enhance the position of systems engineering as a widely recognized academic field. It is proposed that utilizing well-grounded, valid theory will improve understanding of systems engineering phenomena and advance the maturity of the field.

Worldviews, Research Methods, and their Relationship to Validity in Empirical Software Engineering Research

“…Which world view to adopt is also related to the 'objects of study' in a particular field; whether they are inanimate and cannot interpret themselves (as in natural sciences) or they are human beings who can interpret both themselves and their environment [5]. Therefore, for example, if a study is about people in an organization, the interpretivist world view might dominate, while if the objects of the study is a software product and some hypotheses are to be tested, a positivist world view is likely to dominate.…”

Matching research goals and methods in system safety engineering

“…A simple example of a mixed-methods is given by Valerdi in [23] for systems engineering research. He proposes pilot interviews, leading to point sampling, leading in turn to structured surveys and interviews, followed by data analysis of the results.…”

An Effective Verification Strategy for Testing Distributed Automotive Embedded Software Functions: A Case Study

“…Hence, to mitigate this threat, the terms that posed this threat were defined when and where necessary in the report to ensure a common understanding of the terms is maintained as suggested in [121].…”