TY - JOUR
T1 - Performance Evaluation of Production Systems Using Real-Time Machine Degradation Signals
AU - Kang, Yunyi
AU - Yan, Hao
AU - Ju, Feng
N1 - Funding Information:
Manuscript received October 1, 2018; revised January 7, 2019; accepted March 25, 2019. Date of publication June 28, 2019; date of current version January 9, 2020. This work was supported by the National Science Foundation under Grant CMMI-1829238. This paper was recommended for publication by Associate Editor L. Tang and Editor Fan-Tien Cheng upon evaluation of the reviewers’ comments. (Corresponding author: Feng Ju.) The authors are with the School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, Tempe, AZ 85281 USA (e-mail: ykang37@asu.edu; haoyan@asu.edu; feng.ju@asu.edu).
Publisher Copyright:
© 2004-2012 IEEE.
PY - 2020/1
Y1 - 2020/1
N2 - A machine's degradation status directly influences the operational performance of the production system, such as productivity and product quality. For example, machines associated with different health states may have different remaining life before failure, thus impacting the system throughput. Therefore, it is critical to analyze the coupling between the overall system performance and the machine degradation to better production decision-making, such as maintenance and product dispatch decisions. In this paper, we propose a novel model to evaluate the production performance of a two-machine-and-one-buffer line, given the real-time machine degradation signals. Specifically, a phase-type distribution-based continuous-time Markov chain model is formulated to estimate the system throughput by utilizing the remaining life prediction from the degradation signals. A case study is provided to demonstrate the applicability and effectiveness of the proposed method.Note to Practitioners - Machine degradation is commonly observed in many industries, such as automotive, semiconductor, and food production, which gradually deteriorates the machine conditions in different operating processes and affects the production system performance. In practice, sensors are largely deployed on the factory floor to monitor the machine's operating condition. However, a gap still exists between machine operating conditions and system performance. In this paper, we develop an analytical model to predict the machine remaining lifetime and estimate the system performance of a small scale production system, using the machine degradation signals from sensors. Furthermore, a Bayesian updating scheme is provided, which enables online evaluation by utilizing the real-time signals. Such a method provides an effective tool for production engineers to analyze the real-time system performance, and further conduct system improvements and control.
AB - A machine's degradation status directly influences the operational performance of the production system, such as productivity and product quality. For example, machines associated with different health states may have different remaining life before failure, thus impacting the system throughput. Therefore, it is critical to analyze the coupling between the overall system performance and the machine degradation to better production decision-making, such as maintenance and product dispatch decisions. In this paper, we propose a novel model to evaluate the production performance of a two-machine-and-one-buffer line, given the real-time machine degradation signals. Specifically, a phase-type distribution-based continuous-time Markov chain model is formulated to estimate the system throughput by utilizing the remaining life prediction from the degradation signals. A case study is provided to demonstrate the applicability and effectiveness of the proposed method.Note to Practitioners - Machine degradation is commonly observed in many industries, such as automotive, semiconductor, and food production, which gradually deteriorates the machine conditions in different operating processes and affects the production system performance. In practice, sensors are largely deployed on the factory floor to monitor the machine's operating condition. However, a gap still exists between machine operating conditions and system performance. In this paper, we develop an analytical model to predict the machine remaining lifetime and estimate the system performance of a small scale production system, using the machine degradation signals from sensors. Furthermore, a Bayesian updating scheme is provided, which enables online evaluation by utilizing the real-time signals. Such a method provides an effective tool for production engineers to analyze the real-time system performance, and further conduct system improvements and control.
KW - Machine degradation
KW - Markov chain
KW - performance evaluation
KW - remaining life
UR - http://www.scopus.com/inward/record.url?scp=85074867337&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85074867337&partnerID=8YFLogxK
U2 - 10.1109/TASE.2019.2920874
DO - 10.1109/TASE.2019.2920874
M3 - Article
AN - SCOPUS:85074867337
SN - 1545-5955
VL - 17
SP - 273
EP - 283
JO - IEEE Transactions on Automation Science and Engineering
JF - IEEE Transactions on Automation Science and Engineering
IS - 1
M1 - 8751152
ER -