The mining industry provides energy resources and raw materials that have a direct impact on the US economy, producing $100 billion of the gross domestic product (GDP) and providing 1.7 million direct jobs in 2015.¹ Worldwide from 1900 to 2005, there were about 90 billion tons of materials extracted, with a world GDP of about 45 trillion international dollars.² The US mining industry has made positive strides in reducing the severity and frequency of mine accidents and decreasing the number of mine disasters (defined as incidents resulting in five or more deaths) through research, technology and preventive programmes.³ However, during 2006, mine disasters occurred in the United States at the Darby Mine No. 1 Mine and Sago Mine that together claimed 17 lives. In response, the US Congress passed the Mine Improvement and New Emergency Response (MINER) Act⁴ that includes the following provisions: underground coal mines must have at least two separate, distinct travelable passageways as escapeways that are clearly marked to show route and direction of travel to the surface, and a directional lifeline or equivalent device must be installed the entire length of each escapeway. While the MINER Act was a major piece of legislation to improve mine safety, efforts need to continue to improve miner safety and their ability to self-escape.

A 2013 report by the National Research Council identified the need to empower miners to self-escape during a mine emergency; numerous recommendations were put forth, including the need to accelerate wayfinding technology efforts that enhance situational awareness and self-escape. The report recommended that the National Institute for Occupational Safety and Health (NIOSH) should ‘accelerate efforts to develop technologies that enhance self-escape. These technologies should use human-centered design principles with specific attention to facilitating improved situational awareness and decision making…’’⁵ Currently, passive self-escape technologies include signage, markers designating primary and secondary escapeways, and lifelines. Other self-escape aids have been evaluated that include a safety cane used to tap about the area in order to find the mine rib and a handheld laser pointer used as a flashlight to enable miners to better see objects in the mine. Miners had the most favourable opinion of the safety cane, but did not want to carry it. These devices proved useful, but they did not enable faster escape from a mine.⁶ More sophisticated, active systems include the Mains Fail Operated Evacuation System (MOSES), which consists of a series of sound- and light-emitting units designed to help miners to self-escape in low or no visibility conditions.⁷ MOSES was installed in 76 Australian and South African mine sites from 1994 to 1995. There were encouraging results when miners used the system in zero visibility conditions and they reported it was unlikely they could have reached safety without MOSES.⁷

The IMC Egress Beacon System, developed with the UK Health and Safety Executive and Mines Rescue Service, performs a similar function of sound and visual cues from red and green LEDs to alert miners if they are egressing in the proper direction. Tests were conducted and the system was found to have considerable value.⁸ Another active system is the Miniguide Ultrasonic Mobility Aid developed by GDP Research in Australia, a handheld device that uses ultrasonic sounds to detect objects and then provides tactile and audio feedback through vibrations and sound when an object is near. Data from tests conducted by SIMTARS indicate that the Miniguide is effective in locating tripping hazards, but it did not significantly improve escape time travel in smoke.⁶ Overall, these active systems provide varying degrees of improvement to egress from a smoke-filled mine, but are not without limitations. Depending on audible cues for wayfinding can be problematic because after an explosion a miner’s hearing may be compromised. Furthermore, the absence of tactile feedback for indicating direction and locations can slow or inhibit evacuation, especially in low visibility where a miner’s vision can be very limited.

Lifelines are required for anthracite, bituminous, and lignite mines.⁹ Lifelines are used to aid the self-escape of miners by guiding them along the way out of the mine. They provide a tactile cue, given they are fitted with directional cones that are felt with a gloved hand to guide miners in low-visibility conditions. Additional tactile markers on the lifeline are positioned to identify other important locations in an emergency escape, including doors, branch lines, refuge alternatives, and self-contained self-rescuer (SCSR) caches that contain portable sources of breathable air. There are several types of lifelines that are approved by the Mine Safety and Health Administration (MSHA) for use in underground coal mines. The lifelines are constructed of polypropylene rope, aircraft cable, or two-conductor insulated wire housed in an outer jacket. However, in order for the lifeline to be effectively utilised, it must first be reached, which can be challenging in emergency conditions of limited visibility from smoke. Mines fires can produce either white or black smoke, the latter of which can worsen visibility further. The smoke density can vary greatly up to the point where a miner has zero visibility. During a mine fire, miners don a SCSR that isolates them from the smoke and provides breathable air. Additionally, miners typically experience disorientation after a mining incident that results in poor visibility conditions from an explosion or fire, so they may experience difficulty in locating a lifeline.⁷

The approach to improve miner self-escape was to focus on the lifeline given that it is a very effective passive navigational aid^10,11 and lifelines have been identified to be the most effective self-escape aid in low visibility.¹⁰ Lifelines have an advantage over signage and escapeway markers that can be very hard to see in smoke.¹² Miners who had escaped a smoke-filled mine were interviewed and they reported that they had a difficult time seeing the escapeway markers located on the ceiling because they had to bend over to walk or crawl in the thick smoke.¹³ The lifeline can be very useful even in zero visibility conditions because it can be located by feel. Lifeline visibility can be improved with the use of diffuse fibre optic technologies such as Fibrance, which ‘leaks’ light along its entire length, thus enabling an illuminated visual cue. Visual cues are critical given that 80% of human perception is visual.¹⁴ Visual perception directly affects cognitive, task, and motor performance¹⁵ all of which are essential for empowering a miner to escape or take actions to prevent accidents. NIOSH is developing a lighted lifeline, using the Fibrance diffuse fibre illuminated by red, green or blue lasers. These colours were selected because they are readily available commercially; however, it was unknown which colour would provide the best visual cue. Secondly, a miner’s cap lamp is a primary light source that is critical to aid in self-escape when visibility is good, but the cap lamp’s effect on visual performance in detecting a lighted lifeline is unknown in smoke conditions. Therefore, the objectives of the present study were to: (1) determine the fibre colour that enabled the best visual performance in terms of the detection distance, miss rate and the ability of people to identify the colour, and (2) determine the effect of the cap lamp light source on visual performance.

The results of this study clearly indicate that having a cap lamp on severely limits the ability to detect the fibre lifeline, regardless of colour. The cap lamp produces luminances in smoke ranging from 1.5 to 30 cd/m², which are orders of magnitude greater than the luminance of the lighted lifeline. The contrast between the fibre and surrounding area becomes very limited when the cap lamp is on because the light illuminates the smoke and a veiling of the fibre takes place that inhibits visibility.¹⁵ Other research has shown that ambient light scattering in smoke creates a veil of light that can greatly reduce visibility of emergency exit signs.²² Using a cap lamp resulted in the lifeline not being detected 98.3% of the time compared to 49.2% without the cap lamp on. The latter result is somewhat misleading given it includes false positives from the rope lifeline, which cannot be seen without a light source. Eliminating the rope lifeline reduces the miss rate to 25.8%, of which 90% of those misses occurred with the red-lighted fibre. These descriptive statistics provide compelling evidence that it is better to have the cap lamp off. SDT was used to analyse miss data for the conditions of the cap lamp on and off. The SDT analysis results (Table 3) clearly indicates that detectability d’=0.0 with the cap lamp on was merely by chance, while detectability d’=1.21 with the cap lamp off indicating detectability was most likely not by chance.

The miss rate is of prime importance given detection of a lifeline is more critical for self-escape than recognising the lifeline colour. The colour recognition data do provide some useful information. First, it is important to understand that the average spectral sensitivity of the eye changes depending upon the general light levels. For daylight conditions (photopic) the peak sensitivity is about 555 nm (yellow–green), and for moonlight conditions (scotopic) the peak sensitivity is about 507 nm (blue–green). The mesopic state is between the photopic (>10 cd/m²) and scotopic (<0.001 cd/m²) states, where the peak mesopic sensitively of the eye is between 555 nm (photopic) to 507 nm (scotopic).²³ The spectral sensitivity of the eye shifts from green towards blue as the light level decreases. Although the lighting conditions were mesopic, the conditions were much closer to scotopic than photopic given the fibre luminance in smoke was 0.025 cd/m², 0.047 cd/m² and 0.041 cd/m², for the red, green and blue, respectively. The poor visibility of the red-lighted fibre was likely because the wavelength of 639 nm was further from the peak wavelength of scotopic conditions and because it had a much lower value of luminance in smoke. Further investigation is needed to determine why the luminance of the red fibre lifeline was much lower in smoke even though the luminance was higher than the green or blue fibre lifelines in clear conditions.

Colour recognition in the mesopic state will become more difficult as the luminance decreases and approaches the scotopic state. This is because as luminance decreases, the rod photoreceptors of the eye become increasingly dominant over the cone photoreceptors that enable colour perception.²³ This helps explain the relatively low percentages of correct colour recognition for the green and blue fibre lifelines, these being only 36.7% and 6.7%, respectively. These results indicate that using colour to convey useful information for miners would not be very effective in mesopic luminance conditions, especially in smoke, which would further reduce luminance.

Although the blue fibre lifeline had a lower correct colour recognition (6.7%), it was detected 100% of the time compared to the 90% detection for the green fibre lifeline. From this, one could infer that blue had an advantage. However, the results for visual performance, which were based on the detection distance, indicate that there is not a statistically significant difference between green and blue fibre lifeline detection distances, although the green-lighted fibre lifeline had a slightly better average detection distance of 5.36 m compared to the average detection of 5.43 m for blue.

Overall, the results indicate that the blue fibre lifeline performed slightly better than the green fibre lifeline given it had a 100% detection rate, but there was not a statistically significant difference in visual performance between the blue- and green-fibre lifelines. Therefore, additional measurements were made to determine if one of the colours had an advantage not evident from the tests conducted.

It is important to determine the extent of luminance attenuation as a function of the fibre length. It is desirable to have a smaller luminance decrease as length is increased. A 105 m length of fibre was illuminated with green and blue lasers. The luminance of the green fibre decreased 39% between the 25% to the 50% length location, while the blue fibre luminance decreased 68% for the same locations. Given this additional information, green appears to be the most desirable colour for the lighted lifeline.

The testing was conducted in a clean and controlled laboratory environment that minimised the effects from confounding variables likely to be encountered in a mine environment, such as the accumulation of dust on the lifeline that would partially reduce the luminance. More investigation is needed to determine differences between the laboratory results using artificial smoke described by the present paper and those from testing done in the field using real smoke. The testing was in a somewhat benign situation; thus, results might vary in an actual self-escape emergency where people would be under duress. In addition, the testing scope was limited to the detection of the lighted fibre. It would be very useful to conduct field-testing in smoke to determine if the lighted fibre improves the speed of egress, especially given that some evaluations of devices for self-escape did not significantly improve egress speed.^6,8

The lighted fibres were tested at a constant luminance. Flashing the lighted fibre on and off might improve detectability given that participants would see a somewhat abrupt variation compared to a gradual increase in light intensity that was afforded as the lighted fibre apparatus came closer to the participant. A 4-Hz flashing visual warning system developed for mining machines enabled the fastest detection in comparison to multiple visual modes used for warning participants of machine movements.²⁴ A study of automotive rear warning lights indicated that a light flashing rate of 4 Hz was optimal²⁵ and another study indicated that a flashing brake system significantly improved drivers’ response time.²⁶

Lastly, the study targeted an older age group with normal vision so it is unknown how younger age groups would perform with respect to the detection and recognition of the lighted fibre colour. Age is the most common cause of limited visual capabilities that include declines in visual acuity, contrast sensitivity, colour discrimination and sensitivity to glare.²⁷ We assume that younger age groups would perform better. Similar research that evaluated a person’s ability to detect coloured escapeway markers in a simulated mine of smoke determined that the youngest age group had the best visual performance.¹⁶

A lighted lifeline is being developed in response to the 2013 National Research Council report recommendation that NIOSH should ‘accelerate efforts to develop technologies that enhance self-escape. These technologies should use human-centered design principles.…’⁵ The human factors of lighting drove the selection of a green-lighted fibre for the lighted lifeline based on visual performance data concerning detection distance, miss rate, colour recognition and data that indicated the green-lighted fibre lifeline could allow a longer length to be used given the luminance decrease of 39% versus 68% for the blue fibre lifeline over a distance of 105 m. It should be noted that the green laser tested in the experiment had a peak wavelength of 515 nm, putting it in the blue–green region and close to the peak scotopic sensitivity of 507 nm. The blue laser had a peak wavelength of 450 nm – closer to the violet part of the spectrum and well below the peak sensitivity of scotopic vision, though, very interestingly, the blue lighted fibre lifeline performed slightly better than the green lighted fibre lifeline with respect to miss rate. The rods in the eye, which are the only photoreceptors involved in scotopic vision, are more sensitive to the blue region and are highly sensitive to movement, which may explain the increased detectability of the blue-lighted fibre lifeline as participants could be perceiving the motion of the apparatus travelling towards them. An optimum laser colour for detectability could be possible, though commercial availability of more specific laser wavelengths may be limited and costly.

The research also indicated that visual performance in smoke was much better with the cap lamp off, which might be counterintuitive to miners. The lighted lifeline appears to have merit for improving self-escape from underground mines based on the visual performance data that were far superior for the lighted fibre compared to the traditional rope lifeline. The technology could have other mining applications that include outlining mining machines to improve their visibility to miners and thus empower them to avoid moving machinery hazards. The technology could potentially be used for non-mining applications that include escape from rail and roadway tunnels and buildings.