Shaping the Future of Industrial Manufacturing with Predictive Maintenance

A team of researchers from PARC and Panasonic Corporation took top honors for the “Best Paper” award at the International Conference on Prognostics and Health Management hosted by the Institute of Electrical and Electronics Engineers (IEEE) on June 19th in San Francisco, California.

Their winning paper was titled “High-Accuracy Unsupervised Fault Detection of Industrial Robots Using Current Signal Analysis.” The collaboration between PARC and Panasonic researchers proposed a framework for unsupervised fault detection that can effectively identify the faults in industrial robots using current signals.

Robots and other automation machines have been widely used in various industries such as the automotive and semiconductor sectors to improve productivity, quality, and safety in manufacturing processes. However, an unforeseen robot shutdown has the potential to cause an interruption in the entire production line, resulting in significant unplanned downtime, economic costs, production losses, and even work injuries. Thus, it is of high interest to detect incipient faults in industrial robots before they totally shut down or otherwise fail.

A challenge for fault detection in industrial robots is the difficulty to obtain enough labeled training data under normal and abnormal health conditions. Thus, unsupervised machine learning algorithms are desired.

Schedule-driven maintenance is the typical practice observed for industrial robots today. This approach can be inefficient for large production lines because unnecessary maintenance can lead to increased downtime. On the other hand, reactive fail-and-fix maintenance can lead to even worse and expensive instances of unplanned downtime. A single hour of unplanned downtime in automotive, semiconductor, petrochemical or other high-volume factories can lead to operational losses greater than $1 million.

However, with the development of low-cost sensing technologies and advances in analytics, there is an emerging opportunity for predictive or condition-based maintenance (CBM) to increase asset availability by reducing unnecessary maintenance and unplanned downtime. A key goal of CBM is to detect faults in the early stages and allow proactive, predictive, and effective planned maintenance actions to be performed during off-peak production hours and thus avoid expensive unplanned downtime.

PARC is addressing this problem with a new platform for Industrial Internet of Things (IIoT) System Analytics known as MOXI™. The MOXI suite combines embedded sensing, complex system models and artificial intelligence technologies to predict adverse system conditions with high accuracy, negligible false alarm rates and near-zero missed detections.

For this paper, experiments were performed on an industrial robot under both normal and abnormal conditions. The results validated that the proposed fault detection framework was effective in detecting gear-wear faults in robots with higher than 96% accuracy.

The award-winning authors of this paper included Fangzhou Cheng, Ajay Raghavan and Deokwoo Jung from PARC, and their colleagues Yukinori Sasaki and Yosuke Tajika from Panasonic. Congratulations to the entire team for their important research work which is helping to shape and advance the state of industrial manufacturing worldwide.