A paper co-authored by College of Computing & Informatics (CCI) Professor
Ellen Bass, PhD, won “Best Paper in Human Performance Modeling” at the
Human Factors and Ergonomics Society 2017 Annual Meeting in Austin, Texas (October 9 to 13).
The award was presented by the Human Factors and Ergonomics Society's Human Performance Modeling Technical Group.
The paper, titled “Simulating Human Performance of Task Sharing: Modeling Task Delay and Delegation of Authority” and co-authored by Douglas W. Lee and Daniel Fitzick, BS computer science ’17 describes a computational human agent model that manages work by either executing or delaying the execution of a task or delegating activities to other agents. Each agent considers its capacity and strategies for delegation to coordinate with other agents.
One example used in this paper was a framework for simulating multiple types of agents to evaluate a hypothetical operation that could distribute work through an air traffic controller that delegates tasks to flight deck crews. Such case studies show how capacity can alter agent utilization and delegation strategies as well as redistribute task work amongst multiple agents while creating teamwork demands. (See full abstract below.)
The paper was published in the Proceedings of the Human Factors and Ergonomics Society 2017 Annual Meeting.
Bass shares joint appointments in both CCI and the
College of Nursing and Health Professions (CNHP). She also holds affiliate status in the
School of Biomedical Engineering, Science and Health Systems. She has over 30 years of human-centered systems engineering research and design experience in air transportation, biomedical informatics, healthcare, process control and weather related applications.
Bass earned a bachelor of science in engineering (BSE) with a major in bioengineering and a bachelor of science in economics (BSE) with a major in finance from the University of Pennsylvania; a master of science in advance technology from the State University of New York at Binghamton; and a doctorate in systems engineering with a major in human-machine systems and a minor in artificial intelligence from the Georgia Institute of Technology.
Abstract
In systems that support dynamic allocation of work across human and autonomous agents, analyzing the implications of task sharing can support operational concept development. Computational tools should address not only the taskwork but also the teamwork emerging from the allocation. This paper describes a computational human agent model that manages work by executing or delaying the execution of the task, or by delegating activities to other agents. The agent considers its capacity and strategies for delegation to coordinate with other agents. Using a framework for simulating multiple types of agents, case studies apply this computational human agent model to the evaluation of a concept of operation that distributes work across an air traffic controller capable of delegating and flight deck crews. The case studies show how capacity changes agent utilization and delegation strategies redistribute taskwork across multiple agents while creating teamwork demands.