Automatic validation and failure diagnosis of human-device interfaces using task analytic models and model checking

  • Authors:
  • Matthew L. Bolton

  • Affiliations:
  • Department of Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, USA 60607

  • Venue:
  • Computational & Mathematical Organization Theory
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

When evaluating designs of human-device interfaces for safety critical systems, it is very important that they support the goal-directed tasks they were designed to facilitate. This paper describes a novel method that generates task-related temporal logic properties from task analytic models created early in the system design process. This allows analysts to use model checking (a means of performing exhaustive mathematical proofs) to automatically validate that formal models of human-device interfaces will let human operators successfully perform the necessary tasks with the system. This paper also presents an algorithm that uses the method to diagnose why a particular task is not valid for a given design. The application of both the method and algorithm are illustrated with a patient-controlled analgesia pump programming example. The method and algorithm are discussed and avenues for future work are described.