(119a) Moving Towards the Smart Factory Leveraging Big Data Advancements

James Moyne, Michael Armacost, Mingwei Li and Amos Dor

Applied
Materials—Applied Global Services

Santa Clara, CA, USA

Key Words—Big Data, Smart
Manufacturing, Industry 4.0, Analytics, Prediction Methods, Predictive
Maintenance, Virtual Metrology, Digital Twin, Cyber-Physical Systems Abstract Introduction

Smart manufacturing
(SM) is a term “generally applied to a movement in manufacturing practices
towards integration up and down the supply chain, integration of physical and
cyber capabilities, and taking advantage of advanced information for increased
flexibility and adaptability” [1,2,3]. It is often equated with “Industry 4.0”
(I4.0), a term that originated from a project in the German government that
promotes a 4^th generation of manufacturing that uses concepts such
as cyber-physical systems, virtual copies of real equipment and processes, and
decentralized decision making to create a smart factory [4,5].

As “big data” requirements and capabilities in data volumes, velocity
(rates), veracity (quality), variety (merging) and value (analytics) increase rapidly,
manufacturers across industries are beginning to see opportunities to leverage
big data advancements and enable the many facets of SM to improve quality,
capability and throughput, and reduce cost [6,7]. The facets of SM that are
aggressively being pursued include (1) leveraging advanced analytics especially
for predictive technologies; (2) tightening integration up and down the supply
chain network; (3) improving the use of cyber-physical system (a term used to
refer to the “tight conjoining of and coordination between computational and
physical resources” [8]) to improve qualities such as adaptability, reconfigurability, functionality, and reliability; (4)
improving the use of real-time simulation-like tools to begin to realize the
“digital twin”, which refers to a digital replica of the physical system at
multiple levels creating the “virtual factory” [9]; and (5) relying more heavily
on a knowledge network to incorporate subject matter expertise (SME) in
analytical solutions [10]. Determining the State-of-the-Art of Smart Manufacturing

While advancements are being made across these facets of SM, the
progress varies from industry-to-industry and from facet-to-facet, based in
large part on the pace of evolving needs in the industry and the progress of research
and development in a particular facet. With respect to industry progress, the microelectronics
manufacturing industry (which includes semiconductor and display manufacturing)
has seen significant progress on the SM front because it is needed to address
device production requirements coming from the constant innovation in areas
such as mobile devices, Internet of Things (IoT) and
even big data, as well as the pressures to produce high volumes, flexibly at
low cost. During microelectronics manufacturing precision processes are used to
create “chips” or “panels with multiple layers of complex circuitry.  Each layer contains features that must be
controlled to the nanometer or even molecular level. A typical microelectronics
manufacturing process has hundreds of these precision control steps.

Figure 1 illustrates the SM vision for the microelectronics
industry, incorporating the previously mentioned SM facets [11]. One SM area
where the microelectronics manufacturing industry has excelled is in the
application of analytics. As an example the dynamic quality control of these
steps in this industry is facilitated with a set of capabilities collectively
termed “Advanced Process Control” or APC, which includes dynamic model-based
process control, Fault Detection and Classification or “FDC”, and more recently
predictive technologies including Predictive Maintenance (PdM), Virtual
Metrology (VM) [1]. As an example of the use of cyber-physical systems,
high-mix product production requiring flexibility and high throughput is
addressed with real-time scheduling and dispatch capabilities. These
capabilities incorporate characteristics such as product priority, production
process capability, transport times and queue length, and maintenance schedules
in scheduling and dispatch decisions. These and other SM capabilities make
semiconductor manufacturing an idea arena with which to benchmark the current
state of the art in SM.

Figure 1: A Smart Factory vision for the microelectronics
industry.

In determining the progress in SM facets across the manufacturing
base, it is important to understand that some SM capabilities can be
implemented today, while others are still in the research phase. The latter might
best be implemented today with existing leading-edge capabilities that are
effective forerunners to the ultimate SM vision, rather than focusing on the
ultimate solution. As an example of SM capabilities that can be implemented
today, PdM is a technology that has been shown to be effective in semiconductor
manufacturing. While the challenges of wide-spread PdM deployment are
significant, interim solutions for PdM such as extrapolation on diagnostics
data can often be effective as a complement to more complete and complex
solutions, as long as prediction horizon and prediction confidence are still
conveyed with the maintenance prediction event.  As an example of capabilities that are
effective precursors to the ultimate SM vision, the concept of the digital twin
across the enterprise is a vision of factory integration in semiconductor
manufacturing [6], however it is generally accepted that achieving the final vision
in the industry won’t happen for at least five years. In the meantime the
aforementioned rules-based real-time scheduling and dispatch is an effective
precursor to the digital twin. It utilizes configurable rules and workflows to
capture processing and product behavior, thereby providing real-time reconfigurability decisions in order to optimize throughput
in the face of quality, production schedule and product priority constraints.
Further the capability is structured such at the rule base can be supported by
or replaced with real-time simulation/emulation capabilities as these
capabilities evolve [12].

Summary

Effective adoption of SM technology requires an understanding of
the current state-of-the-art of the tenants of SM as well as the path towards
the ultimate SM vision. In this tutorial the SM vision will be described
including a detailed description of the individual SM tenants. The current
across-industry state-of-the art of each tenant will then be detailed,
identifying current solution approaches that can provide a path to the ultimate
vision. State-of-the-art solution approaches and case studies in
microelectronics manufacturing will be used to illustrate the SM evolution
approach.

References:

[1]    J. Moyne and J.
Iskandar, “Big Data Analytics for Smart Manufacturing: Case Studies in
Semiconductor Manufacturing,” Processes Journal, Vol. 5, No. 3, July
2017. Available on-line: http://www.mdpi.com/2227-9717/5/3/39/htm.

[2]    Wikipedia: Smart
Manufacturing. Available online:
https://en.wikipedia.org/wiki/Smart_manufacturing.

[3]    Davis, J., Edgar,
T, Porter, J., Bernaden, J., and Sarli,
M. “Smart manufacturing, manufacturing intelligence and demand-dynamic
performance,” Computers & Chemical Engineering, 2012, vol. 47, pp.
145–156.

[4]    Project of the
Future: Industry 4.0, Germany Ministry of Education and Research, Available
online: http://www.bmbf.de/en/19955.php.

[5]    Kagermann, H.; Wahlster, “W. INDUSTRIE 4.0 Smart Manufacturing for the
Future”, Germany Trade and Invest, 2016.

[6]    International Roadmap for Devices and Systems
(IRDS): Factory Integration White Paper, 2016 edition. Available
online: http://irds.ieee.org/images/files/pdf/2016_FI.pdf.

[7]    Moyne,
J., Samantaray, J., and Armacost,
M., “Big Data Capabilities Applied to Semiconductor Manufacturing Advanced
Process Control”, IEEE Transactions on
Semiconductor Manufacturing, 2016, Vol. 29, No. 4, pp. 283-291.

[8]    Cyber-Physical Systems (CPS) Program
Solicitation NSF 10-515, Available online:
https://www.nsf.gov/pubs/2010/nsf10515/nsf10515.htm.

[9]    Wikipedia: Digital
Twin. Available online: https://en.wikipedia.org/wiki/Digital_twin.

[10]Armacost,
M. and Moyne, J. “Moving towards the ‘smart factory’ in microelectronics manufacturing”,
Nanochip,
vol. 12, N. 2, 2017.

[11]Hasserjian, K. “Emerging
Trends in IC Manufacturing Analytics and Decision Making”, Keynote, Advanced
Process Control Conference XXVIII, October 2016. Available online via: http://apcconference.com.

[12]Lopez, F., Moyne,
J., Barton, K., and Tilbury, D., “Process-capability-aware
scheduling/dispatching in wafer fabs”, Advanced Process
Control Conference XXIX, October 2017. Available online via: http://apcconference.com.