Operationalizing Machine Learning Using Requirements-Grounded MLOps

Abstract

[Context & Motivation] Machine learning (ML) use has increased significantly, [Question/Problem] however, organizations still struggle with operationalizing ML. [Principle results] In this paper, we explore the intersection between machine learning operations (MLOps) and Requirements engineering (RE) by investigating the current problems and best practices associated with developing an MLOps process. The goal is to create an artifact that would guide MLOps implementation from an RE perspective, aiming for a more systematic approach to managing ML models in production by identifying and documenting the goals and objectives. The study adopted a Design Science Research methodology, examining the difficulties currently faced in creating an MLOps process, identified potential solutions to these difficulties, and assessed the effectiveness of one particular solution, an artifact containing guiding Requirements Questions sorted by ML stages and practitioner roles. [Contribution] By establishing a more thorough understanding of how the two domains interact and by offering practical guidance for implementing MLOps processes from an RE perspective, this study advances both the MLOps and RE fields.