Pitfalls and Protocols in Practice of Manufacturing Data Science

TL;DR

This paper discusses common pitfalls in applying machine learning and data science in manufacturing, highlighting real-world issues and proposing protocols to improve practical implementation and avoid errors.

Contribution

It identifies specific procedural pitfalls in manufacturing data science and offers practical protocols to enhance the reliability and effectiveness of ML/DS applications.

Findings

Identification of key manufacturing data science pitfalls

Proposed protocols to avoid common errors

Guidelines for practical ML/DS implementation in manufacturing

Abstract

The practical application of machine learning and data science (ML/DS) techniques present a range of procedural issues to be examined and resolve including those relating to the data issues, methodologies, assumptions, and applicable conditions. Each of these issues can present difficulties in practice; particularly, associated with the manufacturing characteristics and domain knowledge. The purpose of this paper is to highlight some of the pitfalls that have been identified in real manufacturing application under each of these headings and to suggest protocols to avoid the pitfalls and guide the practical applications of the ML/DS methodologies from predictive analytics to prescriptive analytics.