Citation

Duffy VG, Ng PPW, Ramakrishnan A. Int. J. Ind. Ergonomics 2004; 34(4): 335-348.

Copyright

(Copyright © 2004, Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

This paper considers opportunities to improve effectiveness of computer-based industrial training. Modification is made to an intelligent virtual reality-based training system to take advantage of the way people appear to respond to auditory and visual cues. Training scenarios are developed based on the idea that many times human errors in industry are shown to coincide with conditions that are infrequently encountered. Knowledge and auditory cues for different machining conditions for a computer-numerical-control (CNC) milling machine are built into the existing knowledge base. A simulated tool breakage that consisted of an animation, text and sound of glass breaking is integrated into a virtual training sequence and is triggered by the failure to stop operation in the poor cutting condition represented by sub-optimum spindle speed and cutting sound. A test is performed to check the transfer to the real world task. The simulated tool breakage is intended to help users take advantage of their tendency toward 'dual coding'. An analysis is conducted to determine the impact of the simulated accident on decision-making performance in a real machining task. A disincentive is also introduced to assess the impact of potential tradeoffs between safety and performance that exist in the workplace. Results indicate that when a disincentive was introduced during the training task and the real task, mental workload was significantly higher, and operator decision-making performance was significantly improved when a simulated tool breakage was included in the training sequence. Some implications for researchers and practitioners are discussed.Relevance to industryResults obtained from this study indicate that virtual training with simulated accident may improve decision-making performance.