Proceedings of the 2021 Winter Simulation Conference
S. Kim, B. Feng, K. Smith, S. Masoud, Z. Zheng, C. Szabo and M. Loper, eds.
BUIDING A DIGITAL TWIN FOR ROBOT WORKCELL PROGNOSTICS AND HEALTH
MANAGEMENT
Deogratias Kibira
Guodong Shao
Brian A. Weiss
Engineering Laboratory
National Institute of Standards and Technology
100 Bureau Drive
Gaithersburg, MD 20899, USA
ABSTRACT
The application of robot workcells increases the efficiency and cost effectiveness of manufacturing systems.
However, during operation, robots naturally degrade leading to performance deterioration. Monitoring,
diagnostics, and prognostics (collectively known as prognostics and health management (PHM))
capabilities enable required maintenance actions to be performed in a timely manner. Noting the importance
of data-based decisions in many current systems, effective PHM should be based on the analysis of data.
The main challenges with robot PHM are the difficulties of relating data to healthy and unhealthy states,
and lack of models to fuse and analyze up-to-date data to predict the future state of the robot. This paper
describes concepts of digital twin development to overcome the above challenges. A use case of a digital
twin modeling robot tool center point accuracy is provided. The proposed procedure for this digital twin
will be applicable to different use cases such as reduced repeatability or increased power consumption.
1 INTRODUCTION
1.1 Industry challenge
Industrial robots advance manufacturing by performing intricate, repetitive, or dangerous tasks such as
material handling, welding, assembly, and painting. Because of their flexibility, robot workcells can
respond to demands for customized products, changes in orders, and changes in equipment status. Once put
into operation, robots begin to degrade. If the degradation leads to a failure, the result can be expensive
repair costs and significant production interruption. To minimize failure instances and enhance their
decision-making with respect to maintenance practices, manufacturers turn to monitoring, diagnostic, and
prognostic technologies (collectively known as prognostics and health management (PHM)).
Invoking any effective PHM capability involves equipment (or process) monitoring along with
corresponding data collection. Analytics are applied to evaluate the status of robots, and if there are
degradation problems, provide a diagnosis. In addition to diagnosis, prognostics can also predict the future
status of the robot components and estimate the remaining useful life (RUL) (Lee et al. 2014).
There are a large number of diagnostics techniques and methods (Borgi et al. 2017; Izagirre et al. 2021).
Many reviewed methods for diagnostics and prediction are for specific performance deterioration types
while prognostics is relatively lacking in many case studies (Peng et al. 2010). A data-driven prognostics
approach involves developing a fault “model” that must be trained with data representing anticipated faults.
This data may be difficult to obtain or validate (Vogl et al. 2019). To be more effective, PHM methods
should fill the gaps in physical sensor data and modeling virtual sensors to obtain data that cannot be
978-1-6654-3311-2/21/$31.00 ©2021 IEEE