Litcius/Paper detail

Qualitative assessment of the impact of manufacturing-specific influences on Machine Learning Operations

Tim Raffin, Tobias Reichenstein, Dennis T. Klier, Alexander Kühl, Jörg Franke

2022Procedia CIRP13 citationsDOIOpen Access PDF

Abstract

Machine Learning Operations (MLOps) enables the streamlining of the development and deployment processes of machine learning models; thus, manufacturers can utilize the inherent flexibility and adaptability of Deep Learning at scale to further optimize their processes. This publication provides insights into the challenges that companies face while striving for the efficient operationalization of machine learning algorithms. Moreover, a mapping of capabilities and requirements is presented to provide a baseline for the qualitative analysis of the current state of the art. In conclusion, this article discusses the shortcomings of the existing literature and provides novel implications for MLOps systems in manufacturing.

Topics & Concepts

OperationalizationAdaptabilityFlexibility (engineering)Software deploymentComputer scienceBaseline (sea)Artificial intelligenceMachine learningIndustrial engineeringScale (ratio)Manufacturing engineeringSystems engineeringEngineeringProcess managementSoftware engineeringBiologyPhysicsGeologyEcologyOceanographyPhilosophyMathematicsQuantum mechanicsEpistemologyStatisticsIndustrial Vision Systems and Defect DetectionBig Data and Business IntelligenceDigital Transformation in Industry