An Exploratory Study of V-Model in Building ML-Enabled Software: A Systems Engineering Perspective

TL;DR

This study investigates whether the Systems Engineering V-Model can address interdisciplinary collaboration challenges in ML-enabled software. Through 11 practitioner interviews, the authors derive 8 propositions linking V-Model elements (system decomposition, interfaces, V&V, documentation) to collaboration pain points across requirements, architecture, data, and QA. They find that, despite extra effort, the V-Model’s structured, boundary-aware approach aligns with many real-world needs in ML-enabled systems, and propose leveraging its characteristics to develop future process models and tools. Limitations include reduced speed-to-market and rigidity, suggesting further work to balance rigor with agility. Overall, the paper contributes an exploratory, practitioner-grounded argument for adopting SE-oriented V-Model practices to improve cross-functional collaboration in ML-enabled software development.

Abstract

Machine learning (ML) components are being added to more and more critical and impactful software systems, but the software development process of real-world production systems from prototyped ML models remains challenging with additional complexity and interdisciplinary collaboration challenges. This poses difficulties in using traditional software lifecycle models such as waterfall, spiral, or agile models when building ML-enabled systems. In this research, we apply a Systems Engineering lens to investigate the use of V-Model in addressing the interdisciplinary collaboration challenges when building ML-enabled systems. By interviewing practitioners from software companies, we established a set of 8 propositions for using V-Model to manage interdisciplinary collaborations when building products with ML components. Based on the propositions, we found that despite requiring additional efforts, the characteristics of V-Model align effectively with several collaboration challenges encountered by practitioners when building ML-enabled systems. We recommend future research to investigate new process models, frameworks and tools that leverage the characteristics of V-Model such as the system decomposition, clear system boundary, and consistency of Validation & Verification (V&V) for building ML-enabled systems.

Figures (2)

• Figure 1: V-Model walden2015systems in the context of building ML-enabled systems
• Figure 2: 11 SE functions beale2006systems in the context of building ML-enabled systems