In this talk we review an ongoing research project on group testing procedures jointly with Wolfgang Stadje (Osnabrück, Germany), Frank van der Duyn Schouten (Tilburg University) and Onno Boxma ( Eindhoven University of Technology). This project has already yielded several research papers. More specifically we review various types of group testing models. The objective is to choose an optimal group size for pooled screening of a contaminated population so as to collect a prespecified number of good items from it with minimum testing expenditures. The tested group that are found contaminated are either discarded or are used as new sampling population in later stages of the testing procedure. Since testing may be time consuming, we also consider deadlines to be met for the testing process. We derive algorithms and obtain exact results for the underlying distributions of the associated stopping times, enabling us to find optimal procedures. We shall briefly review various aspects of group testing procedures such as multinomial group testing; unreliable results (false-positive and false-negative), window periods and expiration dates. Some applications will be given, the most of which are concerned with blood screeningfor detecting viruses and some relevant industrial problems. The talk will not be of any mathematical nature but rather a brief “state of art” on the subject.
Prof. Shaul Bar-Lev is full professor at the Department of Statistics at the University of Haifa in Israel. His research interest lies in statistics and in stochastic operations research. He is an international expert on exponential families. In the past 12 years, he has visited The Netherlands many times, leading to intensive research collaboration with Prof. dr. van der Duyn Schouten (leading to 12 published papers plus one under review) and other researchers.
As a visiting professor Shaul Bar-Lev continues to intensify his research collaboration on Stochastic Operations Research at the Econometrics and OR Department, and will be actively involved in various research activities and events within the Faculty of Economics and Business Administration.
This is not a Tinbergen seminar, please register for the seminar by sending an email to email@example.com before September 26th.