Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Industrial and Managements Systems Engineering

Committee Chair

Steven E. Guffey.


It is possible that providing quantitative feedback would help individual workers recognize that their hearing protector device (HPD) is either not properly seated or is not in place when needed. In this study, to test the effectiveness of alerting workers immediately when noise levels in their ears are excessive, noise levels were recorded using noise dosimeters proximal to HPD (SPLear) and at the shoulder (SPLsh) during work. The average noise reduction (NRA) at the ear was calculated at times when a signal light was activated and when it was not. The signal light was positioned below the brim of a hard hat worn by each of twenty-two subjects (16 males and 6 females, average age 42) while they worked normal work shifts at a metal products manufacturing facility.;When activated, the signal light flashed continuously when noise levels recorded proximal to the ear exceeded 80 decibels, A-weighted (dBA). The expectation was that workers would wear their HPD a larger fraction of the time when noise levels in their ears exceeded 80 dBA. Likewise, it was expected that a worker whose HPD was on but not properly seated would be induced to re-position or otherwise re-fit the HPD if the light came on. Hence, changes in behavior due to signaling would produce effectively higher NRA values than would occur when no signaling was activated.;Results of the six tests per subject showed that minute-by-minute NR values fluctuated widely during their tested work shifts. Signaling with the light when noise levels were high had modest effects on NRA, producing a statistically significant (p<0.0023) but very modest difference (8.4 dBA Light Deactivated and 10.6 dBA Light Activated), suggesting that the light modestly increased the average HPD NR for workers when activated.;Note that 7 of the 20 subjects showed values of NRA < 5 dBA, both with and without the light signaling, suggesting that the plug provided poor protection for them at all times, perhaps making it impossible for them to improve their performance by re-adjusting their fits. On the other hand, Subject 6 obtained NRA = 21.6 dBA both with and without light signaling, the highest reduction experienced by any subject. It is possible that Subject 6 could not improve because he already had his best possible fit. It was noted that three other subjects had NRA ≥ 15 dBA both with and without the light. It is possible that both also experienced the best protection that they could achieve with this plug, making it impossible for the light to produce improvement.;Finally, four subjects experienced an improvement of more than 5 dBA, with the highest being 7.9 dBA. Five subjects experienced reductions in NR A, with the greatest being NRA = 13.6 dBA. That subject experienced the only decline greater than 1.1 dBA. Because eleven subjects experienced improvements less than 5 dBA, it is possible that the minor improvements and declines were actually due to variability of fits, not actual reductions or improvements.;Although the average improvement in NRA was less than 2.2 dBA, the light signaling acted as a fit test because the seven subjects who could not obtain even modest protection (i.e., > 5 dBA) all reported being unable to get the light to stop flashing, making it clear to them that the plugs were ineffective. It also produced > 5 dBA improvements for another four subjects.;In conclusion, the results suggest that some of these highly experienced workers were doing their best to protect themselves, yet were not successful. Light signaling had three beneficial effects: (1) it revealed that some workers could not obtain a NRA >5 dBA and needed either training or a different protector, (2) it revealed that only two employees had excellent protection (NRA > 20 dBA), and (3) it lead to an improvement of at least 5 dBA for four of the workers.