INTELLIGENT MAINTENANCE SYSTEMS

The Industry/University Cooperative Research Center (I/UCRC) program is aimed at developing long-term partnerships among industry, academe, and government. These centers are catalyzed by a small investment from the National Science Foundation (NSF) and are primarily supported by center members, with NSF taking a supporting role in their development and evolution. I/UCRCs stimulate highly leveraged industry/university cooperation by focusing on fundamental research recommended by Industrial Advisory Boards (IAB). Each center is established to conduct research that is of interest to both the industry and the university with which it is involved, with the provision that the industry must provide major support to the center at all times. The centers rely primarily on the involvement of graduate students and some full time researchers in their research projects, thus developing students and scientists who are knowledgeable in industrially relevant research. With industrial and other support totaling 10 to 15 times the NSF investment, I/UCRCs are a premier example of "leveraged" funding model. Over the past two decades, the I/UCRCs have led the way to a new era of partnership between universities and industry, featuring high-quality, industrially relevant research, strong industrial support of and collaboration in research and education, and direct transfer of university developed ideas, research results, and technology to U.S. industry to improve its posture in competitive world markets.

Due to rapid advances in sensors, micro-electromechanical systems (MEMS), and with stringent performance requirements and environmental regulations, modern day industrial systems are becoming more and more complex. Intelligent sensor-based decision making agents with wireless communication technologies, when deployed in these complex systems as monitoring/diagnostic/prognostic (M/D/P) tools, will result in better system performance, minimal unscheduled downtime, and reduced maintenance and operating costs. The estimated downtime costs per year for various industries are: Financial $9.4B, Energy $5.7B, Telecommunications $4.2B, Manufacturing $3.3B, and Retail $2.1B. We recently established a University of Missouri-Rolla (Missouri S&T) site of the NSF I/UCRC on Intelligent Maintenance Systems (IMS) to address the issues of such industrial systems.

Complex industrial systems are truly nonlinear, contain deterministic and stochastic terms, and have spatial and temporal characteristics. They may comprise of numerous components but their sum effect is not just an aggregation of individual outputs. They may have component or system failures. At present it is very difficult to diagnose or predict component failures in such systems and the proposed center intends to fill this void. The mission of the proposed center is to provide an opportunity for cooperative research between Missouri S&T researchers and industrial partners to develop appropriate technologies in order to bridge this gap in the areas of M/D/P with industrial applications. Missouri S&T's research expertise includes component health monitoring, diagnostics and prognostics of industrial systems, sensor development, decision making and control, and advanced simulation relevant to M/D/P. Significant funding has been awarded and patents have been granted to Missouri S&T faculty on this topic. Missouri S&T has teamed up with University of Cincinnati (UC) and University of Michigan (UM) to accelerate the pace of becoming a national leader in M/D/P research. Missouri S&T IMS Centerintends to build a quality research program in M/D/P and associated educational activities supported by a well-defined plan for technology transfer to industries.

A critical piece in the economic success of any industry that involves the operation of such complex systems is to minimize the downtime. The availability of such complex systems requires novel sensors, decision making capability, and computer control with suitable communication schemes for various subsystems. A major goal of this Center is to find effective ways to investigate relevant M/D/P tools and technologies for solving important industrial problems by bringing together faculty with expertise in M/D/P and domain-specific knowledge.

Core Competencies: Missouri S&T has been actively working in the area of M/D/P technologies for the past decade. Missouri S&T has been engaged in the area of sensor development, embedded prognostics, agent-based decision support, and wireless communication technologies of industrial machinery/components. Several of these technologies are currently in production use. Several million dollars worth of funding has been awarded from federal and state agencies, and industries. At least 17 patents have been awarded to the Missouri S&T faculty in this area. Some of the success stories relevant to M/D/P include: prognostic agent on a chip, smart wireless sensor network for remote monitoring, oil life monitoring, tire pressure monitoring and prognostic and smart portable service tools. These technologies have resulted in significant cost savings to industries. The proposed site at Missouri S&T will bring value to its members by demonstrating high-impact technologies as well as by harnessing business alliances through test beds. Missouri S&T's core competencies include:

(1) Sensing and Monitoring : sensing techniques (e.g. cable sensors), wireless sensor networking and hardware (Missouri S&T mote), and system on a chip;

(2) Diagnostics and Prognostics : sensor data fusion, neural networks and fuzzy logics based decision making, and agent based decision support; and

(3) Advanced Simulation : virtual reality based simulation and virtual prototyping.

Possible research topics that can be conducted by the Center site at Missouri S&T include:

- Smart Computational Prognostic Agent

o Prognostic algorithms for predicting degradation, life and performance loss

o Decision making and control

- Web-Enabled Training Device to Business

o Optimization of maintenance and production/service scheduling

o Web-based maintenance technology for M/D/P and smart portable service tools

o Degradation modeling for service value chain optimization

o Knowledge acquisition, decision making and data mining

- Embedded Wireless Sensor Networking Technology

o Embedded sensor hardware, prognostics/diagnostics on a chip, remote monitoring

o Wireless-based portable service tools for M/D/P

By serving as a center of excellence for the creation and dissemination of knowledge in intelligence e-maintenance systems, the proposed center intends to impact next-generation product, manufacturing and service systems with six-sigma quality.

Industry Relevance: Demonstration of an advanced on-board M/D/P system is contingent upon developing and integrating intelligent sensor-based agents with these systems. The M/D/P system will provide not only advanced capabilities, such as improved safety, but also it will reduce maintenance and operating costs, minimizes secondary and catastrophic failures and finally it will extend the useful life of industrial components. If the component's life is estimated on-board and ahead, then just-in-time maintenance can be performed. This Center intends to solve various industry-related M/D/P problems.

Leveraging Existing IMS Center: The overall IMS Center (www.imscenter.net) is engaged in industry- relevant, intelligent maintenance research work. The vision of the Center is to enable products to sustain near-zero downtime performance through the advancement of web-enabled predictive infotronics and tether-free technologies, including smart computational prognostic agents, device-to-business platform, as well as self-maintenance design methodologies. The IMS Center at UC and UM is currently supported by 15 industrial members. Every six months, the investigators of the IMS Center present their latest research results and propose plans by engaging the members in a one and half day event called IAB meeting. Day two is a half day program geared towards business aspects and implications of the consortium-based Center membership. The research results from the existing IMS Center will be available to all membership companies joining the proposed Missouri S&T site.

Industrial Memberships: The annual membership rates are$40,000 for Full Members and $11,000 for Associate Members. Companies with less than 500 employees are eligible for associate memberships. All members are allowed to vote on potential projects during the IAB meeting.

Membership Benefits: All members can leverage their internal R&D resources through joining the Center. Missouri S&T will waive all indirect costs associated with the memberships, which in turn will result in significant cost savings. Members can attend workshops conducted by the Center at minimal or no cost. Members have the right to use the Center reports, data and information internally for evaluation and further research. Members shall have an option to a non-exclusive, non-transferable, worldwide royalty-free license on any patentable invention conceived and reduced to practice during the membership period under Center projects conducted by the Missouri S&T, the UC and/or the UM. Members shall have an option to a non-exclusive, non-transferable, royalty free site license to use copyrighted software arising from Center Projects for any of the Center sites agreed upon by the member and the Center.

If you are interested, please contact:

Dr. Jag Sarangapani

Site Director, NSF I/UCRC on Intelligent Maintenance Systems

Room 221, Emerson Electric Hall

University of Missouri-Rolla

1870 Miner Circle, Rolla, MO 65409

Tel: 573-341-6775; Fax: 573-341-4532

sarangap@mst.edu

Current UMS IMS Project

Auto-ID Research Group