[Tutorial 3] Statistical Methods for Information Fusion System Design and Performance Evaluation

Date & Time:

Sunday, October 7, 13:00 – 16:20


2F R02 (Orchard South)


Ali K. Raz, Daniel A. DeLaurentis


In this tutorial, a domain-agnostic framework, based on Design of Experiments, is presented which provides holistic performance evaluation of complex systems. This framework leverages systems engineering principles for identifying design variables in large design spaces which are then investigated by statistical methods (e.g., analysis of variance) for establishing statistical significance and quantifying their impact on end system performance. This tutorial will discuss theoretical foundations for performing design and analysis of experiments, followed by a hands-on application on an information fusion system example which can be easily transferred to domain-specific implementation of a complex system in the
participant’s area of interest (Information fusion systems are widely applicable in defense applications, self-driving cars, and autonomous systems). A refresher on Monte-Carlo simulations and hypothesis testing will also be provided. At the conclusion of the tutorial, participants will be able to formulate an experimental design for complex system performance evaluation, employ hypothesis testing for comparing uncertain data, perform analysis of variance (ANOVA) to establish statistical significance of design variables and interactions, and perform multiple comparison range tests to quantify the impact of design variables and obtain sensitivity analysis of interactions for system performance evaluation.

Short Biography:

Dr. Ali K. Raz is a research scientist at Purdue University’s Center for Integrated Systems in Aerospace. His research interests are in Complex Systems, System-of-Systems Engineering, and Information Fusion System. He was the recipient of Alexander Kossiakoff fellowship awarded jointly by the John Hopkins University and the International Council on Systems Engineering (INCOSE) for developing performance evaluation methods for system-of-systems. He holds BSc and MSc in Electrical Engineering from Iowa State University and PhD in Aeronautics and Astronautics from Purdue University. He is a Certified Systems Engineering Professional (CSEP).

Dr. Daniel A. DeLaurentis is a Professor in Purdue University’s School of Aeronautics and Astronautics. He is the Director for Purdue’s Institute for Global Security and Defense Innovation and leads Purdue’s Center for Integrated Systems in Aerospace (CISA), which is home to 20 faculty affiliates. His research is conducted under grants from NASA, FAA, Navy, the DoD Systems Engineering Research Center UARC, and the Missile Defense Agency. Dr. DeLaurentis is an Associate Fellow of the American Institute of Aeronautics and Astronautics (AIAA), and senior member of the IEEE. He is also Co-Chair of the Systemof- Systems Technical Committee in the IEEE Systems, Man, and Cybernetics (SMC) community and is an Associate Editor of the IEEE Systems Journal.