|
On Friday, June
20th, distinguished university faculty from across the country
will present significant aspects of their research programs.
This portion of
the Symposium program has been  designed to allow university researchers
and Symposium participants to explore mutually beneficial
avenues of research that are of interest and relevant to today’s
practitioner. Presentations will be delivered from 8:00 a.m. to
2:30 p.m. and the program concludes with a round table
discussion at 3:00 p.m.
The showcase of current university research is organized by the
ReliaSoft Risk, Reliability, and Maintainability Research
Alliance, which has been established through the Department
of Industrial Engineering at the University of Arkansas and
receives support and cooperation from ReliaSoft Corporation. The
goal of this research alliance is to foster university and
industry partnerships for meeting research challenges and needs
identified by practicing engineers.
Session U-1
8:00 to 8:45 a.m. Friday June
20, 2008
Reliability:
Beyond the Failure Rate, Applications to Homeland Security and
Service Systems
Jose Emmanuel Ramirez-Marquez
School of Systems & Enterprises, Stevens Institute of Technology
Traditionally, the
concept of reliability has been associated with applications in
very well identified industries such as avionic, automotive,
manufacturing and telecommunications. The general belief is that
few applications exist outside these industries. The focus of
this presentation is to present reliability as a dynamic concept
that can be applied in numerous fields. For illustration
purposes, specific application examples of reliability in
non-traditional fields, such as Homeland Security and Service
Systems, will be discussed. In particular, developing optimal
container inspection strategies under constrained resources that
address concerns regarding potential attacks via container
cargo. Also, the presentation will discuss applications of
reliability to supply chain systems and how traditional
reliability techniques can be extended to analyze these systems.
Finally, the new concept of "system resiliency" will be
discussed together with its relationship to system reliability.
Session U-2
8:45 to 9:30 a.m. Friday June
20, 2008
A Real-Time
Degradation-Based Prognostic Methodology for Improving
Reliability Assessment
Nagi Gebraeel
School of Industrial and Systems Engineering, Georgia Institute
of Technology
Unexpected
failures of engineering systems are major contributors to human
fatalities and result in astounding costs. Many failure
mechanisms can be traced to an underlying degradation process
that inevitably leads to failure. The high degree of uncertainty
associated with such degradation processes coupled with our
limited understanding of the physics-of-failure is a major
obstacle in accurately predicting unexpected failures and
assessing reliability measures.
This presentation will
discuss a novel degradation-based prognostic methodology
that utilizes sensor-based condition/health monitoring
information to compute and continuously update, in
real-time, residual lifetime probability distributions of
partially degraded systems and their components. The idea of
the proposed methodology rests on developing a stochastic
framework that combines: 1) in-situ condition-based
degradation signals (generally correlated with the
underlying physics-of-failure) unique to the system being
monitored and 2) general reliability and degradation
attributes of the system's population. In contrast to
conventional reliability techniques where the mean
time-to-failure (MTTF) is a fixed (or time-based) value,
this methodology leads to the evaluation of dynamically
evolving MTTFs that are based on the latest degradation
state of the system. The dynamically evolving residual life
distributions ultimately enable the development of
sensor-driven replacement and spare parts logistics decision
policies, also known as Autonomous Logistics.
Session U-3
10:00 to 10:45 a.m. Friday June
20, 2008
Integration of
Component, Subsystem and System Test Data
David G. Robinson
Risk and Reliability Analysis Department, Sandia National
Laboratories
Engineering
design is dominated by evolutionary design changes rather
than revolutionary changes. A new system might involve the
change-out of a component with next generation technology.
In addition, with the increasing expense of system level
testing, test data may be available from various levels of
indenture within the system: component testing, subsystem
testing and system level testing. Traditional methods can be
used to combine point estimates resulting from the various
test packages. However, if we are interested in
characterizing the uncertainty or confidence in the system
reliability estimates, then traditional reliability methods
are inadequate. This presentation outlines an objective
approach for combining the test data from whatever level of
indenture test data is accumulated. A specific example is
presented where significant errors in reliability estimates
can result if traditional methods are employed. An
alternative approach is discussed along with a number of
variations that may arise.
Session U-4
10:45 to 11:30 a.m. Friday June
20, 2008
Reliability
Importance Measures Considering User-Specific Applications and
Uncertainty
David W. Coit
Department of Industrial & Systems Engineering, Rutgers
University
New reliability
importance metrics are being developed to explicitly consider
reliability prediction uncertainty and to reflect user needs.
Reliability importance or criticality metrics provide a
quantitative basis for comparing the relative influence of
individual components. These metrics generally depend on the
system structure function, the component reliability, and the
impact of component failure on system reliability. These metrics
are useful to prioritize design improvement initiatives, select
aging components to replace or restore, prioritize inspections
and other reasons. There are many different importance metrics
that have been proposed in the literature, and selection of
which metric to use depends on a user's particular needs. These
metrics always assume that the component reliability values are
known or can be predicted very accurately. However, in practice,
reliability prediction has uncertainty, and often, unacceptable
levels of uncertainty. The individual components with more
reliability prediction uncertainty have more risk associated
with their usage, and this should be reflected in the importance
metrics. Furthermore, existing reliability importance metrics
are based on system reliability for a specified mission time.
For many companies, other baseline metrics would be more
appropriate (e.g., availability, MTTF, cost) and this is
part of these research efforts.
Session U-5
1:00 to 1:45 p.m. Friday June
20, 2008
Research Topics
in Scheduled Maintenance of Repairable Systems
C. Richard Cassady
Department of Industrial Engineering, University of Arkansas
Equipment
prognostics and condition-based maintenance are rightfully the
hottest topics in research related to equipment maintenance.
However, the traditional scheduled maintenance paradigm still
provides tremendous opportunities for improving the
effectiveness of equipment operation. In addition, there are
still many open and interesting scheduled maintenance research
problems. In this presentation, the body of knowledge on
scheduled maintenance planning will be summarized and several
avenues of ongoing research in this area will be described.
These current research avenues include maintenance resource
allocation (prioritizing maintenance actions, cannibalization,
etc.), repairable equipment performance modeling (imperfect
maintenance modeling, maintenance optimization, etc.),
productivity-based maintenance (incorporating maintenance plans
into operations planning) and technological change
(obsolescence, upgrades).
Session U-6
1:45 to 2:30 p.m. Friday June
20, 2008
Reliability
Prediction and Equivalency of Accelerated Testing Plans
E. A. Elsayed
Department of Industrial & Systems Engineering, Rutgers
University
John English
College of Engineering, Kansas State University
Reliability
prediction of new components, products and systems is a
difficult task due to the lack of well designed test plans that
yield "useful" information during the test and due to the
stochastic nature of the normal operating conditions. The
accuracy of the reliability prediction has a major effect on the
warranty cost and repair and maintenance strategies. Therefore,
it is important to design efficient test plans. Traditionally,
ALT is conducted under constant stresses during the entire test
duration. In practice, constant-stress tests are easier to carry
out but need more test units and a long time at low stress
levels to yield sufficient degradation or failure data. However,
in many cases the available number of test units and test
duration are extremely limited. There could be other choices in
stress loadings where each stress loading has some advantages
and drawbacks.
This presentation will address many practical questions raised
by industry such as: Can accelerating test plans involving
different stress loadings be equivalent? What are the measures
of equivalency? Can such test plans and their equivalency be
developed for multiple stresses, especially in the setting of
step-stress tests and other profiled stress tests? When and in
which order should we change the stress levels in multi-stress
multi-step tests?
Round Table Discussion
3:00 to 4:30 p.m. Friday June
20, 2008
The program will
conclude with an interactive round table discussion designed to
foster open communication among the participating university
researchers and Symposium attendees. By bringing together
knowledgeable and influential members of the academic community
with reliability practitioners from government and industry who
are intimately familiar with the specific reliability and
quality needs of their organizations, this portion of the
Symposium provides a powerful opportunity to share information
about the ways in which current university research can be
applied to meet real-world challenges and to target specific
areas in which additional solutions are needed.
The discussion will also address the specific role that the
ReliaSoft Risk, Reliability, and Maintainability Research
Alliance can play in focusing the expertise available in
academic reliability programs on specific corporate initiatives.
|
