Abstract
The literature discusses data science (DS) as a very promising set of techniques and tools to support lean production (LP) practices. DS could aid manufacturing companies in transforming massive real-time data into meaningful knowledge, increasing process transparency and product quality information and supporting improvement activities through data-driven decision-making. However, no attempt has been made in the literature to formalise the links between DS and LP practices. Thus, this study aims to overcome this gap by clarifying the DS techniques and tools that can support LP practices and how to apply them. This study employs a quantitative bibliometric method – specifically, a keyword co-occurrence network analysis – on a set of papers extracted from Scopus. The results obtained allowed the researchers to identify a set of DS techniques and tools that can support LP practices and to develop a model to guide their implementation based on the typical improvement implementation stages of the plan-do-check-act cycle. The model shows how to use DS techniques and tools in LP for: identifying areas for improvement and subsequent implementation (plan); enabling a better knowledge and process management (do); identifying/predicting potential problems and employing statistical process control (check); providing remedial actions and effectively applying process improvement (act).
