A "Smart Component" Data Model in PLM

Published

October 28, 2015

Author(s)

Yunpeng Li, Utpal Roy, Seungjun Shin, Yung-Tsun Lee

Abstract

Today's physical products are becoming smarter not only because of their increasingly complex functionalities, but also for their superior capabilities of adaptation to dynamic working environment based on embedded software and real-time information processing. Data analytics, particularly predictive analytics for modeling causal relations between input and output parameters by statistical or machine-learning techniques, is a key technology for implementing the intelligence of these "Smart Products". Many literatures have shown the creation of predictive models in the manufacturing domain; however, they are limited to describe how to effectively employ predictive models throughout a product's lifecycle. We address that integrating a product's physical components and its associated predictive model components is essential to make use of predictive models, and this can be implemented uniformly in Product Lifecycle Management (PLM) systems, due to much commonality found in these two heterogeneous components. We propose a "Smart Component" data model to incorporate predictive models as another type of "parts" or "services" of the product in its master records in PLM, such that individual model components can be modularized, composed, reused, traced, maintained, and replaced on demand. We implement a prototype system to demonstrate the feasibility of the proposed data model using an open-source PLM solution.

Proceedings Title

2015 IEEE International Conference on Big Data

Conference Dates

October 29-November 1, 2015

Conference Location

Santa Clara, CA, US

Pub Type

Conferences

Download Paper

Local Download

Keywords

Smart Product, Smart Component, Predictive Analytics, PLM, PMML, CRISP-DM

Citation

Li, Y. , Roy, U. , Shin, S. and Lee, Y. (2015), A "Smart Component" Data Model in PLM, 2015 IEEE International Conference on Big Data, Santa Clara, CA, US, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=919024 (Accessed April 26, 2024)