Cyber-enabled distributed machine learning for smart manufacturing systems

Yaser Banadaki; Safura Sharifi

doi:10.1117/12.2514911

18 March 2019 Cyber-enabled distributed machine learning for smart manufacturing systems

Yaser Banadaki, Safura Sharifi

Proceedings Volume 10973, Smart Structures and NDE for Energy Systems and Industry 4.0; 109730Z (2019) https://doi.org/10.1117/12.2514911
Event: SPIE Smart Structures + Nondestructive Evaluation, 2019, Denver, Colorado, United States

Abstract

In this paper, we propose a distributed machine learning (DML) algorithm to fulfill the requirements of the smart factory (or Industry 4.0) including self-organization, a distributed control function, communication between the smart components, and real-time decision-making capability. We show the proposed DML algorithm not only enables the smart factory to adjust the components for new demands and circumstances, but also each component of the system acts smart and communicate with each other, either request or offer functions. The DML is an interactive learning mechanism among smart components and a natural way of scaling up learning algorithms. The different machines can have the best learning algorithms of their own data while the communication between different learning processes is an integration of different learning biases that compensate one another for their inefficient characteristics. As such, the size of the smart factory is scalable and the growing amount of data from additional machines has a minor effect on the communication overheat. We will elaborate on the DML model that overcomes the problems of centralized systems and increases the possibility of achieving higher accuracy, especially on a large-size domain.

Citation Download Citation

Yaser Banadaki and Safura Sharifi "Cyber-enabled distributed machine learning for smart manufacturing systems", Proc. SPIE 10973, Smart Structures and NDE for Energy Systems and Industry 4.0, 109730Z (18 March 2019); https://doi.org/10.1117/12.2514911

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available