2010 Joint Research Conference Plenary Presentations
Reliability Inference Based on Multistate and Degradation Models
Dr. Vijay Nair
University of Michigan
Reliability or survival analysis is traditionally based on time-to-failure data. In high-reliability applications, there is usually a high degree of censoring, which causes difficulties in making reasonable inference. There are a number of alternatives to increasing the efficiency of reliability inference in such cases: accelerated testing, collection and use of extensive covariate information, and the use of multistate and degradation data when available. This talk will focus on the last topic. We will describe different multistate models that arise in applications and discuss inference for semi-Markov multistate models with panel data (interval censoring), a common type of data collection scheme. The second part of the talk deals with degradation data. We will review some common models for analyzing degradation data and then describe a class of models based on non-homogeneous Gaussian processes. Properties of the models and methods for inference will be discussed. The talk is based on joint work with Yang Yang, Yves Atchade, and Xiao Wang.
Think Big, Think Different: Designing Experiments for National Security and Public Policy
Dr. Stephen Fienberg
Carnegie Mellon University
The field of experimental design and its use in quality improvement has often looked narrowly at relatively small, albeit important, problems arising in agriculture, clinical trials, and industrial processes. I will describe a series of issues facing the nation that to me demand new statistical thinking about the design and analysis of statistical experiments. They arise in contexts such as medical innovations, national security, social policy, and the utilization of network data.
Designed Experiments that Changed the World
Dr. Bradley Jones
JMP
This talk tells the story of several experiments. Each of these experiments made such a major addition to our understanding or capabilities that in a very real sense we can say that they changed the world. The goal of this talk is to promote more standard usage of designed experiments.
Kaizen at Google Scale
Dr. Diane Lambert
Loosely speaking, Kaizen means continuous improvement of industrial, engineering or management processes through small local changes. At its core lie measurement, experimentation, and learning followed by implementation. This talk will show how Google uses these well-established quality principles (along with huge amounts of data) to improve seach and ads for users, advertisers, and publishers.