Gainesville, FL 32611
Matthew Bolton, Ph.D.
Assistant Professor of Industrial & Systems Engineering for the University at Buffalo
Title: Using Formal Methods & Human Performance Modeling to Engineer Safe & Effective Systems
Breakdowns in complex systems often occur as a result of system elements interacting in unanticipated ways. In systems with human operators, unexpected human interactions (i.e. human errors) are frequently cited as contributors to such failures. Conventional human factors engineering practices are not exhaustive and thus
miss interactions critical to system safety and performance. Formal methods are tools and techniques from computer science that enable analysts to mathematically prove whether models of computer systems
satisfy safety properties. Dr. Bolton’s research investigates how formal methods can be used with human performance modeling to discover and correct problems caused by potentially unexpected human automation
In this talk, he will describe two efforts performed in pursuit of this objective. For the first topic, Dr. Bolton will describe a novel method for generating erroneous human behaviors as divergences from task analytic human behavior models. He will show how this method can be used with formal methods to prove whether or not potentially unanticipated erroneous behavior can contribute to complex system failures using an apache helicopter example. For the second topic, Dr. Bolton will describe how his lab has been using formal methods and psychoacoustics to find and correct issues where interactions between medical alarms can make one or more of them unhearable. He is currently using this method to evaluate the alarm sounds from the international medical alarm standard.
Department of Industrial and Systems Engineering at the University of Florida