Embedding the MLOps Lifecycle into OT Reference Models

TL;DR

The paper addresses the challenge of integrating MLOps into OT environments that must satisfy real-time, safety and regulatory constraints. It proposes embedding the MLOps lifecycle into OT reference models, with a detailed examination of RAMI 4.0 and ISA-95 and a focus on RAMI 4.0 for precise mapping. A comprehensive lifecycle-to-reference-model mapping is developed and validated in an ICS testbed using a data pipeline (process data, Kafka streaming, GPU-accelerated training, MLFlow deployment) to demonstrate feasible integration while preserving operational stability. The results show that direct transplantation of IT MLOps practices is infeasible; structured adaptation via RAMI 4.0 supports interoperability, traceability, and safe deployment in OT contexts, offering a blueprint for industrial uptake.

Abstract

Machine Learning Operations (MLOps) practices are increas- ingly adopted in industrial settings, yet their integration with Opera- tional Technology (OT) systems presents significant challenges. This pa- per analyzes the fundamental obstacles in combining MLOps with OT en- vironments and proposes a systematic approach to embed MLOps prac- tices into established OT reference models. We evaluate the suitability of the Reference Architectural Model for Industry 4.0 (RAMI 4.0) and the International Society of Automation Standard 95 (ISA-95) for MLOps integration and present a detailed mapping of MLOps lifecycle compo- nents to RAMI 4.0 exemplified by a real-world use case. Our findings demonstrate that while standard MLOps practices cannot be directly transplanted to OT environments, structured adaptation using existing reference models can provide a pathway for successful integration.

Figures (9)

• Figure 1: The lifecycle as described in Salama2021practitioners showing its key components and their interactions.
• Figure 2: The Model
• Figure 3: The Model
• Figure 4: Model Development
• Figure 5: Training Operationalization