Towards a Domain-Specific Framework for Predictive Analytics in Manufacturing

Published

October 3, 2014

Author(s)

David J. Lechevalier, Anantha Narayanan Narayanan, Sudarsan Rachuri

Abstract

Data analytics is proving to be very useful for achieving productivity gains in manufacturing. Predictive analytics (using advanced machine learning) is particularly valuable in manufacturing, as it leads to production improvement with respect to the cost, quantity, quality and sustainability of manufactured products by anticipating changes to the manufacturing system states. Many small and medium manufacturers do not have the infrastructure, technical capability or financial means to take advantage of predictive analytics. A domain-specific language and framework for performing predictive analytics for manufacturing and production frameworks can counter this deficiency. In this paper, we survey some of the applications of predictive analytics in manufacturing and we discuss the challenges that need to be addressed. Then, we propose a core set of abstractions and a domain-specific framework for applying predictive analytics on manufacturing applications. Such a framework will allow manufacturers to take advantage of predictive analytics to improve their production.

Proceedings Title

ACM/IEEE 17th International Conference on Model Driven Engineering Languages and Systems

Conference Dates

September 28-October 3, 2014

Conference Location

Valencia

Pub Type

Conferences

Download Paper

Local Download

Keywords

domain specific modeling, data analytic, machine learning, manufacturing

Citation

Lechevalier, D. , , A. and Rachuri, S. (2014), Towards a Domain-Specific Framework for Predictive Analytics in Manufacturing, ACM/IEEE 17th International Conference on Model Driven Engineering Languages and Systems, Valencia, -1, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=916274 (Accessed December 2, 2024)