Title: Towards Martingale Theory for Multi-Dimensional Survival/Warranty Analysis
: In this paper we present one possible way to develop a martingale theory for the
problems of survival analysis with multi-dimensional time. If proved fruitful, it
could play a role similar to the existing theory in survival analysis with
one-dimensional time. First, we highlight some of the major advantages which a
martingale theory brings to survival analysis. Then we discuss some difficulties
which arise in the multi-dimensional case. Finally, we describe the construction of
the so-called "scanning martingales" which avoids these difficulties, and give some
Title: Constructions and Applications of Ageing Distributions
: Ageing distributions play a fundamental role in reliability.
We present a unified approach in constructing them; and show that
most of the existing distributions may arise from one of these
methods. Generalizations/modifications of the Weibull are often
required to prescribe the non-monotonic nature of the empirical hazard
rates. We also briefly sketch some applications of ageing
distributions in diverse disciplines.
* D.N.P. Murthy (University of Queensland, Australia) - email@example.com
Title: Research in Reliability, Warranty and Maintenance
Please Note: Unfortunately Professor Pra Murthy will not be able to attend APARM 2010. We wish him well.
In place of his keynote presentation on Friday 3 December there will be a Conference Mixer from 9.10-10.25am in AM102.
: Reliability, warranty and maintenance are three closely related topics that have been studied extensively by researchers from many disciplines over the last 60 years. The lecture will briefly review the current status and highlight the need for greater inter-disciplinary approach (and more complex model formulations) to tackle new and challenging problems. It will conclude with a brief discussion of few research topics for study in the future. The lecture should be of interest to engineers, operations researchers, statisticians and managers.
Title: AIS for Self-Maintenance of Wireless Sensor Networks
: Wireless ad-hoc networks form a connected topology using wireless radio decentralized communication without any centralized or fixed infrastructure. Participating devices act as relay stations, thus enabling data transport from a source to a destination through intermediate nodes. Nodes can be subject to several forms of malfunctions which cause a disruption of node availability and reduce the reliability of a network function. Routing protocols can handle some errors but they have problems with violations that are not based on protocol behavior.
In this talk self-maintenance of a sensor (ad-hoc) network deploying autonomous detection and reaction to even previously unknown malfunctions will be discussed.
In our approach we employ principles of the Biological Immune System (BIS) to solve computational problems. The BIS is an adaptive highly distributed learning system containing several mechanisms for defense against pathogenic organisms. The immune system learns through adaptation to distinguish between dangerous foreign antigens and the body's own cells. These very effective mechanisms that are capable to make organisms protect themselves against a wide variety of pathogens seem to us to be a very good metaphor for self-maintenance of sensor- networks. We will highlight how Artificial Immune Systems (AIS) contribute to anomaly detection in sensor networks.
Title: A Non-Obtrusive Method for Uncertainty Propagation in Analytic Dependability Models
Please Note: Due to unforseen circumstances Professor Trivedi will unfortunately not be able to attend APARM 2010. A shorter version of his presentation will be delivered in one of the regular sessions. We regret that he will be unable to attend, and wish him well.
: In this paper, a method for propagating the epistemic uncertainty in the model parameters, through the system dependability model is discussed. This method acts as a wrapper to already existing stochastic models and does not need to manipulate the basic model, giving it a wide range of applicability and ease of use. It is also independent of the solution method of the underlying model and pre-existing model solution methods or tools are relied upon. The applicability of this method is illustrated with some real examples. While our examples discuss confidence intervals for system availability, service reliability and performance, this method can be directly applied to compute uncertainty in the output metrics of other stochastic analytic models of dependability, performance and performability. It is important to note that though this is a sampling based method, no simulation is carried out in our method but actual execution of the underlying analytic model is performed. An adequate number of samples over the parameter space is chosen and the analytic model is solved at each set of sampled parameter values. Statistical analysis of the output vector then yields the distribution and confidence intervals of the model output. Latin Hypercube Sampling (LHS) and random sampling are applied and their robustness is compared.