19.07.2013 Views

SKF Reliability Systems - Library

SKF Reliability Systems - Library

SKF Reliability Systems - Library

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Intelligent Maintenance <strong>Systems</strong>:<br />

The Next Five Years and Beyond<br />

Transforming Condition-based Maintenance to Productivity and Service Innovation<br />

Jay Lee & Mohammed AbuAli University of Cincinnati (USA)<br />

University Cooperative Research Center on Intelligent Maintenance <strong>Systems</strong> (IMS)<br />

(A Paper presented at COMADEM 2009)<br />

The evolution of maintenance has gone through different stages of transformation for the past decades.<br />

The traditional approach to maintaining components, equipment, and processes, has been one that<br />

is purely reactive. Mitigating procedures are performed on the equipment once it breaks down. Many<br />

manufacturing companies have been adopting condition-based maintenance (CBM) or predictive<br />

maintenance techniques (PdM) as the best practices.<br />

To transform maintenance to become a truly proactive and value-added productivity improvement,<br />

major innovation is needed to elevate the value to a new level. This paper will present recent advances<br />

of intelligent maintenance systems as well as its systematic methodology and tools that have been<br />

effectively utilized to transform maintenance into innovative and productive service systems in a<br />

diverse set of industries. Two brief case studies are provided to illustrate the lessons learned with<br />

discussions for future service innovation impacts.<br />

1. INTRODUCTION<br />

Maintenance practices have evolved from reactive to condition-based maintenance, but several issues have<br />

not yet been addressed: (a) an intelligent system that is capable of transforming machine data to machine<br />

health information, (b) a seamless communication mechanism so that health information can be disseminated<br />

to appropriate channels; and (c) a decision-making module to efficiently schedule maintenance and production<br />

to achieve a near-zero downtime manufacturing operational performance.<br />

The Center for Intelligent Maintenance <strong>Systems</strong> (IMS) [1] has been in the pursuit of addressing these maintenance<br />

gaps and needs. The Center for Intelligent Maintenance <strong>Systems</strong> (IMS) is an established NSF Industry/<br />

University Cooperative Research Center (I/UCRC). IMS serves as a center for excellence for formulating,<br />

planning, and conducting research projects with industry in the field of prognostics and health management.<br />

The vision of the center is to research and effectively develop the necessary tools and techniques that are<br />

required to transform today’s industry from a traditional maintenance approach to a predictive and preventive<br />

maintenance practice for value-added services. The center brings value to its members by validating highimpact<br />

emerging technologies as well as by harnessing business alliances through collaborative testbeds.<br />

INDUSTRY COMPANY PROJECT<br />

General Motors Prognostics of Vehicle Components<br />

Automotive<br />

Toyota Industrial Robots Health and Asset Management<br />

Nissan Automotive Robot Torque Monitoring and Prognostics<br />

Harley Davidson Spindle Bearing Machine Monitoring<br />

Heavy Machinery<br />

Caterpillar<br />

Komatsu<br />

Machine Tool Health Monitoring<br />

Heavy Machinery Lifecycle Knowledge Management<br />

Proctor and Gamble Quality-centric Process Health Management<br />

Process<br />

Parker-Hannifin Hydraulic Hose Prognostics and Safety Services<br />

Omron Precision Energy Management <strong>Systems</strong><br />

Consulting<br />

Techsolve<br />

Siemens<br />

Smart Machine Platform Initiative (SMPI)<br />

Reconfigurable Plug-n-Prognose Watchdog Agent®<br />

Table 1 Sample Portfolio of Recent Research Activities at the Center for IMS<br />

By leading research in the development of intelligent learning agents like the Watchdog Agent®, the Center for<br />

IMS has developed mature and enabling technology that will allow the implementation of intelligent prognostics<br />

for the transformation of machine data into machine and process health information. It is a toolbox of algorithms<br />

for different purposes: signal processing and feature extraction, quantitative health assessment, machine<br />

failure prediction, and machine health diagnosis. It is based on a reconfigurable software platform, meaning<br />

that appropriate algorithms can be selected depending on the monitored task. Because of this pervasiveness<br />

feature, the Watchdog Agent ® has been effectively utilized for different applications such as industrial robots,<br />

machine tools, compressors and chillers, production lines, motors, and other components. Table 1 shows a<br />

sample portfolio of recent IMS projects in a diverse number of industries.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!