Nonparametric, real-time detection of process deteriorations in manufacturing with parsimonious smoothing

Abstract

Machine faults and systematic failures are resulted from manufacturing process deterioration. With early recognition of patterns closely related to process deterioration, e.g., trends, preventative maintenance can be conducted to avoid severe loss of productivity. Change-point detection identifies the time when abnormal patterns occur, thus it is ideal for this purpose. However, trend detection is not extensively explored in existing studies about change-point detection – the widely adopted approaches mainly target abrupt mean shifts and offline monitoring. Practical considerations in manufacturing cast additional challenges to the methodology development: data complexity and real-time detection. Data complexity in manufacturing restricts the utilization of parametric statistical modeling; the industrial demand for online decision-making requires real-time detection. In this article, we develop an innovative change-point detection method based on Parsimonious Smoothing that targets trend detection in nonparametric, online settings. The proposed method is demonstrated to outperform benchmark approaches in capturing trends within complex data. A case study validates the feasibility and performance of the proposed method on real data from automotive manufacturing.

Keywords:

• Trend detection
• real-time
• nonparametric analysis
• parsimonious smoothing

Acknowledgments

The data in the case study were provided by research staff at Global Data Insights & Analytics and engineers from Powertrain Manufacturing Engineering at Ford Motor Company. The authors would like to express their appreciation to the aforementioned organizations for their help with domain knowledge and data collection.

Additional information

Funding

This research was supported by Ford Motor Company under the Ford University Research Program Award. The work was partially supported by Rutgers Research Council Grant Award and Rutgers University Big Data Pilot Initiative Grant Award.

Notes on contributors

Shenghan Guo

Shenghan Guo is a PhD candidate in the Department of Industrial & Systems Engineering at Rutgers University. She received a BS degree in financial engineering from Jilin University, an MS degree in financial mathematics from Johns Hopkins University and an MS degree in engineering sciences & applied mathematics from Northwestern University. Her research interests include statistical process control, Big Data analytics and financial mathematics. Her current research focuses on developing innovative data mining approaches to exploit the value of manufacturing big data in guiding in-situ process monitoring and product quality control. She is the recipient of the 2019 Tayfur Altiok Scholarship at the Department of Industrial & Systems Engineering at Rutgers, a finalist in the University and Louis Bevier Dissertation Completion Fellowship, and the winner of IISE Quality Control and Reliability Engineering (QCRE) Division’s Data Challenge at the 2019 IISE Annual Conference.

Weihong “Grace” Guo

Weihong “Grace” Guo is an assistant professor in the Department of Industrial & Systems Engineering at Rutgers University. She received a BS degree in industrial engineering from Tsinghua University, an MS degree and a PhD in industrial and operations engineering from the University of Michigan, Ann Arbor. Her research focuses on developing novel methodologies for extracting and analyzing massive and complex data to facilitate effective monitoring of operational quality, early detection of system anomalies, quick diagnosis of fault root causes, and intelligent system design and control. She received the Barbara M. Fossum Outstanding Young Manufacturing Engineer Award from the Society of Manufacturing Engineers in 2019. She served as the President of the IISE Process Industries Division in 2016-2018.

Amir Abolhassani

Amir Abolhassani is a data scientist at Ford Motor Company’s Global Data, Insight, and Analytics (GDI & A) organization. GDI & A drives evidence-based decision making by providing timely, actionable and forward-looking insights across all the business domains. As a member of the advanced manufacturing and IIoT (Industrial Internet of Things) team, Amir is working on developing real-time process monitoring and prognostic methods to enhance the productivity on a day-to-day basis. He has held various positions in the area of productivity, energy efficiency and quality improvement in the automotive industry and academia prior joining to Ford Motor Company. He received his PhD in industrial and management systems engineering from West Virginia University.

Rajeev Kalamdani

Rajeev Kalamdani obtained MS and Doctor of Engineering degrees in mechanical engineering from Texas A&M University. He also holds a Graduate Certificate in Data Science from Harvard Extension School. Dr. Rajeev Kalamdani has 25+ years of experience with Ford Motor Company in multiple capacities. His current role is Manager of Industrial Internet of Things (IIoT) Analytics at Ford Motor Company’s Global Data, Insight, and Analytics (GDI & A) organization. Prior to this role, he held numerous positions, including: Supervisor of Manufacturing and IIoT Analytics; Senior Analytics Scientist, supervising a team of data scientists working on analytics projects supporting over 100 manufacturing plants; Virtual Manufacturing Supervisor in Powertrain Manufacturing Engineering, supervising a team of engineers engaged in casting and machining for powertrain components; 6 Sigma Master Black Belt; and Manufacturing Engineer.