Robust and optimal design of experiments for models with possible misspecifications

点击翻译按钮获取中文摘要

Experiments are performed in all fields of study to investigate processes or systems. Results from the experiments can be analyzed and used to optimize processes or systems. There are many applications in engineering and sciences. For example, experiments play an important role in designing new products, improving product quality and reliability, developing manufacturing processes, and optimizing systems or service operations. Optimal design of experiments is an important area of research which focuses on effective and efficient designs for experiments. In the next five years, I plan to investigate robust estimation methods and develop robust and optimal design theories for experiments with various model misspecifications. In practice, a proposed model is only approximately true, and there are various kinds of departures from the model assumptions which lead to model misspecifications. For a regression model, misspecifications can occur in the response function, the error variance, the auto-correlation structure of the errors and the distribution of the errors. I will study robust and optimal design criteria, develop design theories, explore new numerical algorithms, and construct robust and optimal designs for linear and nonlinear regression models. The results from optimal designs and experiments will help us make good decisions on products, processes, service operations, government policies and so on. This research may result in huge savings in valuable resources and generate significant social and economic benefits.