Engineering Considerations and Selection Criteria for Proximity Warning Systems for Mining Operations

Introduction

The first step in making mining machines safer is to define and analyze the hazards. Risk assessments can be used to establish priorities so that the most dangerous situations are addressed first and those least likely to cause major problems can be considered later [Brnich and Mallett 2003]. Ergonomic risk factors that can affect the safety of miners who need to work in proximity to machines also need to be considered [Steiner et al. 1994]. Task analysis techniques are used to evaluate the interactions between worker and machine. The results can be used to compare the capabilities and limitations of the operator with requirements of the system. The resulting information is not only useful for designing new machines, but can also be used to improve existing designs [NIOSH 2008a]. System design must consider the worker as a part of the system to ensure efficient and safe operation. Task analysis can also be used to identify and facilitate the need for proper implementation of worker safety devices.

The work area around underground mining machines is very dark and confined (Figure 1). It is normal for equipment to be operated close to the walls of mine entries and for the orientation of the walls to be used to help steer the vehicles. In addition, vehicles commonly bump into one another when loading and unloading ore or coal. Operators are not always aware of the position of their helpers and/or any other person who may be in the work area. A Proximity Warning System (PWS) marker on each worker that is activated as he or she approaches a dangerous area could alert the operator and prevent a potential injury or fatality.

Mining machines continue to grow in sophistication and complexity. Regulatory requirements also have had to be incorporated into designs, which adds to the complexity of the machine. In many cases, little effort has been put into how the introduction of new technology to the machines affects the human operators of this equipment. Human considerations should be addressed and incorporated during the conceptual design phase. Appropriate countermeasures such as PWSs should be considered if an optimal human factors solution is not possible [NIOSH 2008b].

There can be no substitute for proper training, especially in inherently unsafe jobs. Many companies require adhering to workplace protocols to ensure worker safety. Even the best PWS can fail to protect a worker if the worker ignores safe work practices. Therefore, unsafe job tasks should be studied, safe workplace protocols should be developed, and training should be provided for any situation that requires the implementation of a PWS.

Sight Lines, Fields of View, and Blind Areas

Many mining machines, such as large haul trucks, shuttle cars, and continuous mining machines, have operator visibility and illumination issues [NIOSH 2008c]. These problems have contributed to injuries and fatalities to both the machine operators and nearby workers. NIOSH developed a software package called the Crewstation Analysis Program to determine whether mining machines provide adequate visibility and illumination to operate the machines safely [Wiehagen et al. 1994].

Most vehicles have blind spots that obscure the driver's vision (Figure 2). This is particularly true in large construction vehicles and haul trucks [Ruff 2007], but it is also true in much smaller vehicles. Mirrors and cameras have been used to minimize the problem, but accidents still occur - the vehicle operator must first see the problem and then react accordingly. Dirty mirrors and lenses, poor lighting conditions, fog, and driver fatigue can cause the driver not to see other vehicles and pedestrian workers (workers on foot). The addition of a PWS could be beneficial.

The advent of large mining trucks has accentuated the ongoing problem of collisions from poor external vision and limited maneuverability. Determining the blind zones around a vehicle is one important step toward defining which PWS can best be used for a specific situation. Several standardized methods for determining blind zones have been developed: Rear Visibility Index [Paine et al. 2003]; "Earth-Moving Machinery, Operator's Field of View" - ISO 5006 [ISO 1991]; and "H-Point Machine and Design Tool Procedures and Specifications" - Society of Automotive Engineers standard SAE J826 [SAE 2002]. A more simplified version was developed by the NIOSH Spokane Research Laboratory [Ruff 2001] (Figure 3).

Operator Attention and Fatigue

Like many experienced workers, remote-control continuous miner operators perform their everyday jobs making choices and decisions that they may not consciously think about. While performing tasks, they are continually processing feedback and cues that guide them for the next move. Unfortunately, operators sometimes step alongside a moving continuous miner or beyond the supported roof for a better view while coal cutting or tramming. The restricted workspace and reduced visibility in the mine environment cause the continuous miner operator and helpers to assume awkward work postures, and their tasks require quick reactions to avoid being struck by moving machinery [Bartels et al. 2008].

Worker fatigue certainly has been a factor in many fatalities. The worker may see the impending hazard, but it may not register in his or her mind. A "wake-up" call from a PWS may help. When workers perform a task over and over again, it becomes so automatic that the work no longer requires concentration. Their minds can drift, leaving them unresponsive to dangerous events that could occur. Fatigue is a problem in all 24-hr operations. Our biological clocks switch the brain automatically to low levels of alertness after lunch and during the night to induce sleepiness. As a result, mining personnel are not well equipped to sustain alertness and performance during nocturnal work hours or to gain adequate sleep during daytime [Sirois 2003].

Vehicle drivers cannot look directly through the rear window and simultaneously observe the view in the side and rear-view mirrors. An audible alarm provided when an object is in the vehicle's path can improve the situation. Rear-looking video cameras with a cab-mounted monitor can be helpful.

Surface Versus Underground Considerations

It is very important that a PWS selected for a job can work reliably in the given job environment. Radar-based sensors have some unique installation and placement issues. The shape of the signal provided by the sensor in most cases determines where it will be placed. Falling debris, tire motion, flying stones, and mud, rain, snow, and heavy dust concentrations can also dictate placement. The position of the machine operator is particularly important. A radar-based system used for a surface haul truck simply would not be appropriate for a remote-controlled continuous mining machine. The operator of a continuous mining machine spends much time close to the mine walls, large power cables, shuttle cars, and other objects would continuously trigger a radar system. Likewise, a camera and video monitor system typically used on surface haul trucks may not be appropriate for continuous mining machines since the mining machine operator generally carries the remote-control pendant. Further, in an underground coal mine, miners and machines are always close to one another. Any PWS must be very accurate and must work in extreme environmental conditions. With dust being generated, water sprays, electrical noise sources, and flying debris, underground coal mining is probably the toughest industrial environment to which a PWS could be exposed.

The required range of a sensor used on an underground continuous mining machine is generally very short, and the distance measurement must be very accurate. Many underground mining machines run relatively slowly compared to surface haul trucks. Thus, the range of PWSs used in surface mining operations must be proportionally greater.

The operational environment should be considered when selecting alarms to alert vehicle operators and workers on foot. In an underground coal mine face area, the noise can be intense and visibility for most workers can be quite obscured. Placement of the alarms requires careful consideration. Large haul trucks have different problems, such as sun glare, as well as audible noise problems. Some creative alarm methods include seat belt tensioners and steering wheel vibrators.

Areas being monitored by PWSs or cameras need to be determined from blind area diagrams and from a risk analysis. Obstacle detection, particularly for haul trucks, should not be limited just to the rear; front and right-side detection are also needed. The PWS should detect a person or vehicle immediately in front of the truck and off the right front corner where there are huge blind spots.

A suggested "ideal" detection zone for a radar system on the rear of a large dump truck would extend out from the radar antenna to cover the area between the rear tires and immediately behind the tires for the entire width of the truck. The width of the zone could extend slightly beyond the width of the truck, but no more than 5° as measured from the side of the truck. The length of the zone would vary according to the size of the dump truck and the truck's maximum reverse speed. Radar detection to the sides would be important, especially on the right side and rear corners - anywhere where something could be in the truck's path if the wheels were turned. Cameras are essential for adding redundancy to the radar system so that false alarms can be verified without the operator needing to leave the cab of the truck. The detection zone of the radar should not extend beyond the field of view of the camera [Ruff 2000].

Table 1 highlights a sampling of vehicle types and a brief description of selection criteria that can be used as a guide.

Table 1. Basic types of mining equipment and associated PWS selection criteria.

Vehicle type	PWS selection criteria
Continuous miner	Must be an active system; work in close proximity to other objects; alarm only when a person is in danger; be MSHA-approved; be accurate over short distances; operate through coal, rock, dust, water sprays, and steel; work in high electrical noise environments; activate quickly with no time lag to continuous miner shutdown; work with multiple workers and machines. Alarms should be visual, audible, and possibly tactile.
Shuttle car	Must be an active system, work in close proximity to other objects, alarm only when a person is in danger, be MSHA-approved, be very accurate over short ranges, activate quickly with no time lag to vehicle shutdown, work in high electrical noise environments, work with multiple workers and machines. Zone must not be altered near large metal objects. Alarms should be visual, audible, and possibly tactile.
Haulage system	Must be an active system, work in close proximity to other objects, alarm only when a person is in danger, be MSHA-approved, be very accurate over short distances, work in high electrical noise environments, work with multiple workers and machines. Zone must not be altered near large metal objects. Alarms should be visual, audible, and possibly tactile.
Highwall miner	Must stop haulage belt quickly, alarm only when a person is in danger, be MSHA-approved, be very accurate over short distances, work in high electrical noise environments, work with multiple workers and machines.
Load-haul-dump	Must alarm only when a person is in danger, be MSHA-approved, be very accurate over short ranges, activate quickly with no time lag to vehicle shutdown. While many load-haul-dump proximity systems only activate in reverse (e.g., ultrasonic, radar), many accidents occur while going forward. An active proximity system may be appropriate. Must work with multiple workers and machines. Must provide coverage in all blind spots.
Conveyor belt	Must stop belt quickly, work in high electrical noise environments.
Forklift	While many forklift proximity systems only activate in reverse (ultrasonic, radar), many accidents occur while going forward. An active proximity system that works in both forward and reverse may be appropriate. Must work with multiple workers and machines. Must work in all blind spots.
Front-end loader	While many front-end loader proximity systems only activate in reverse (ultrasonic, radar), many accidents occur while going forward. An active proximity system that works in both forward and reverse may be appropriate. Must work with multiple workers and machines. Must work in all blind spots.
Haul truck	Must be capable of monitoring the areas behind the rear tires, the front side, and the right side of the truck. The length of the zones will vary based on the truck size and speed. The rear-zone PWS should only activate when vehicle is in reverse. Must work with multiple workers and machines. Multiple types of systems may be needed (i.e., visual and radar).

Warning System Effectiveness

Warning system tests have been conducted using commercial devices and in-house developments. Specific sensor types investigated include infrared, capacitive, electromagnetic, ultrasonic, lasers, radar, radio frequency identification (RFID), global positioning systems (GPSs), and video cameras.

Very little data are available that show the performance of PWSs when mounted on actual mining equipment. NIOSH engineers have undertaken an evaluation of many available technologies in order to verify their effectiveness on large off-highway dump trucks. The technology types tested included radar, sonar, infrared, GPS, magnetic, tag-based, and video systems. Haul trucks were chosen because of the severity and number of accidents that involve this type of equipment and the extensive blind areas that are typical around these trucks. A simple test procedure was used to evaluate the ability of a system to detect a person and a smaller vehicle, such as a pickup truck [Ruff 2000]. Objects, manikins, or vehicles were moved into the zone of detection, and distance measurements were documented. The test area consisted of an open space on flat terrain with an asphalt base.

It is important to evaluate a PWS on the equipment on which it will be used. Each system should be mounted according to manufacturer's recommendations. Tests should be conducted on the front, back, and sides of the vehicle. Items to be noted and recorded should include at least: false alarms, whether the person was moving, whether the vehicle was moving and in what direction, the speed of the vehicle, and the shape of the detection zone. Vehicle vibration can affect the performance of a PWS. Consequently, it is important to measure the detection zone with the test vehicle's engine running. Radars, ultrasonic, and infrared laser-based warning systems have been evaluated by others for underground mines [Hurteau 1994].

Proximity Sensor Types Investigated

PWSs employing a variety of sensing schemes have been evaluated by NIOSH and others. A few are discussed here, and a sampling of the individual characteristics is provided. Extensive testing of an infrared system was performed by NIOSH. Although it was being used as a guidance sensor, the test results are generally applicable for use as a PWS [Sammarco 1998]. An array of infrared emitters and detectors were placed on two pieces of machinery. The distance and angles between the emitters and detectors were provided by the system. A performance analysis in dust and water sprays was conducted with acceptable results. However, this type of sensor would be prone to nuisance alarms since a small object such as a small piece of coal could trigger the alarm by interrupting the infrared beam between the emitters and detectors. A capacitive type of sensor detects change in the electric field near a vehicle. In a clean, uncluttered environment this can be a very effective sensor; however, a mining environment is not generally clean and uncluttered. Any introduction of objects or debris into the sensing area would trigger an alarm, even when a person is not in danger. Ultrasonic sensors are commonly used for a wide variety of noncontact presence, proximity, or distance-measuring applications. These devices typically transmit a short burst of ultrasonic sound toward a target that reflects the sound back to the sensor. The system then measures the time for the echo to return to the sensor and computes the distance to the target using the speed of sound through the medium. Any object in the detection area of the beam triggers an alarm. Variations occur in speed of sound as a function of temperature, humidity, target size, and target distances. Target surface smoothness, target size, and angle of incidence affect sensor accuracy [Massa 1999a,b]. External ultrasonic noise sources (any nearby operating machinery) can also trigger such devices. Infrared laser-ranging sensors are used in many industrial settings, not only for proximity warning, but also for position determination. These devices determine distance to an object by measuring the time of flight of the laser beam bouncing off of objects in the beam's path. Environmental concerns affecting accuracy include temperature variations, dust, and water sprays. Accuracies are also affected by the target size, surface roughness, and geometric orientations [Anderson 1989].

The most commonly used principle of microwave radar technology that makes motion, direction, velocity, and range sensing possible is the Doppler effect (an apparent change in frequency that occurs when the source and the observer are in motion relative to each other). The distance between each cycle of signal decreases in the direction of motion and broadens in the opposite direction. Factors that affect operational capability are distance to the target, the target's radar cross-section, and the medium through which the wave must propagate [Schultz 1993]. The placement of the unit with respect to the object being sensed is strongly dependent on the object's shape and orientation. A greater amount of energy is returned to the receiver if the radar is aimed perpendicular to the side of a tractor trailer versus a cylindrical tank truck or a person. Special care must be taken to mount the sensor in such a way as to eliminate buildup of thick film deposits, e.g., dust, oil, mud, ice, or snow. Antennas should be mounted in a manner that will prevent water from sheeting or accumulating on the antenna [Hebeisen 1993].

RFID tags are used in many products sold today. They provide a unique, remotely readable identification number and other information using wireless radio technology. These systems employ small tags that are attached to products and a reader that can be attached to anything else. These devices have been adapted to mining in a number of ways, including vehicle tracking, personnel tracking, and inventory control. As the tag enters the detection zone of the reader, the information is conveyed across the link. There are several tag-based PWSs available today. They determine distance based on signal strength, which can be somewhat inaccurate; however, there are other methods to determine distance, such as time of flight. Installation considerations include placement of readers, size of tags, operational life (if self-powered), range, intervening material penetration properties, orientation sensitivity, performance near metallic structures (radio reflective environments), susceptibility to interference (conducted and radiated), response times, multiple tag interaction, multiple read capability, and wearer exposure levels.

Each year an average of about four fatalities occur in surface mines that involve a collision between a piece of mining equipment and a smaller vehicle (Figure 4) or worker on foot, or involve a piece of mining equipment going over the edge of a dump point. The NIOSH Spokane Research Laboratory, in cooperation with Trimble Navigation, Inc. (Sunnyvale, CA), developed a PWS and edge-detection system-based GPS technology and wireless network communications [Holden and Ruff 2001]. The system was demonstrated at the Phelps Dodge Morenci Mine (now FMI) in 2002 and showed an ability to warn equipment operators when in proximity to other mining equipment, smaller vehicles, and static objects like dump points and utility poles. This cooperative work has been completed, but Trimble decided not to commercialize it. Since then, other companies (e.g., Safemine and Acumine) are now selling GPS-based proximity warning systems with deployments at several surface mining operations worldwide.

Large mining trucks have an ongoing problem with collisions resulting from poor vision and limited maneuverability. This has resulted in a very high incidence of accidents where large mining trucks collide with other vehicles and occasionally workers on foot. Additional mirrors or wide-angle Fresnel lenses can give vehicle drivers improved rearward field of view. Ideally, additional sensing could be implemented to give the driver 360° unlimited vision around the truck. However, even this would not be adequate as some trucks are so large that the driver would only be able to look in a limited number of directions at once [Mark and Verhoef 1999]. Proponents of camera-based detection systems believe that, although radar is excellent for accurate range estimates, it is not as reliable as a video camera in determining potential false alarms, such as detected road signs. The most challenging problem for camera-based systems is depth measurement. This could be handled using a stereo system, but such a system is not yet available for rugged applications. For this reason, a camera-based system could also use radar to provide the distance measurement. Problems with using cameras can include the following: the driver must turn his/her head to see a monitor, the field of view can be limited, lighting has to be correct to see properly, and sun glare on monitors and other reflections can be troublesome. Also, a monitor in the driver's cab can affect visibility at night. Dust, dirt, mud, fog, snow, and ice are other issues to address.

Table 2 identifies the various types of proximity sensor systems, the sensing method used by each system, and some advantages and disadvantages of each type.

Table 2. Advantages and disadvantages of various types of PWSs.

Type	Advantages	Disadvantages	Sensing method
Infrared: passive	Good for long distance in fog.	Accuracy issues with heavy snow and rain.	Detects object or person presence by heat energy radiation.
Infrared: active	Good for long distance in fog. Measures vehicle speed.	Environmental concerns affecting accuracy include temperature, dust, and water sprays.	Emits laser beam to ground. Detects reduced time of reflection by objects in path.
Capacitive	Compact and easy to install.	Needs clean environment.	Detects change in capacitance due to object in detection zone.
Ultrasonic - pulse	Compact and easy to install.	All objects trigger alarm. Temperature, humidity, air turbulence, target surface smoothness, target size, angle of incidence, and external noise sources cause accuracy problems.	Detects change in time-of-flight reflection due to object in detection zone.
Radar: pulsed	Compact and easy to install.	All objects trigger the alarm. Snow and ice buildup and angle of incidence accuracy issue.	Measures time-of-flight of a pulse that is transmitted and then reflected off of objects in detection zone.
Radar: Doppler	Compact and easy to install. Measures vehicle speed.	Cannot detect stopped objects. Snow and ice buildup issues.	Detects a frequency shift in generated signal due to object in detection zone.
RFID: passive	Inexpensive and easy to install.	Generally short range. No range information. Orientation sensitivity.	A nonpowered tag detects generated radio signal.
RFID: active	Longer range than passive RFID.	Requires battery in tag. Orientation sensitivity.	A battery-powered tag detects generated radio signal.
GPS	Accurate; covers wide areas.	Only works on the surface.	A receiver detects satellite signals and triangulates position, transmits location to other vehicles/personnel via radio.
Video cameras	Simplicity.	Operator must observe monitor. Limited field of view.	Vehicle operator monitors objects in blind spots on cab-mounted monitor.
Magnetic: passive	Compact and easy to install.	Accuracy issues when metallic objects in field.	Detects change in Earth's magnetic field when objects enter detection zone.
Magnetic: active	Great accuracy over short distances.	Only receiver in detection zone triggers alarms.	A transmitter provides a marker signal. A receiver measures signal strength and provides alarms.

System Design Considerations

PWS design should consider alarm types, nuisance alarms, environmental effects, safety assessments, fail-safe operation, electrical interference, operating range, sensor orientation, activation latencies, vehicle speeds, worker exposures, multiple device activation, explosives ignition hazards, and intrinsic safety issues.

Alarms used to alert vehicle drivers and workers on foot to dangerous situations must stand out from the normal working environment. Ambient noise levels, lighting, and placement of alarm devices should be carefully considered. Creative alarming devices such as a tugging seat belt, vibrating steering wheels, and pager-type vibrating devices may be helpful. False (nuisance) alarms could cause workers to ignore the alarms over time. Maintenance and testing of PWSs should be done regularly to ensure confidence in the safety system. Both surface and underground mining operations are very demanding work environments. A delicate or high-maintenance PWS would not be appropriate.

A safety assessment of the implementation of a PWS should be conducted specific to the vehicle and/or situation to which it is applied. Questions to ask include:

• Does the PWS cause the operator or helper to spend more time in a position of safety?
• Are the warning devices discernable and appropriate (color, signal, and timing)?
• Does the PWS cause operators to react adversely to the warning signals (in terms of operator positioning)?
• Are there situations where the PWS may cause additional safety or health concerns?
• Would it be appropriate or necessary to shut down one or more machine functions in the risk zone?
• Does it appear that the operator can rely on the PWS?
• Will the PWS free up some of the operator's mental workload?

These and other questions should be carefully answered before installing a PWS.

Any safety system must include fail-safe features to ensure that if the safety device fails, the worker is warned of the failure. If a worker gains confidence in a safety system and it fails, the worker's and coworkers' lives can be jeopardized. Warnings and other alerting features must accentuate system failure.

Most electronic systems today, including PWSs, include a microcomputer or software-operated controller as a key component. Software failures can also cause safety system failures. NIOSH researchers have developed several documents that can be used as guides on the proper methods to develop software for safety-based systems intended for the mining environment [Sammarco et al. 2001].

Depending on the PWS used, an evaluation of the electrical environment should be performed. The proximity of a PWS to large motors, high-current switching devices, and generators can wreak havoc with any electronic device. A preliminary assessment of the installation area using an electrical spectrum analyzer could help identify sources of electromagnetic interference that can cause potential system malfunctions and false alarms.

Sensor-specific factors need to be considered, including detection zone shape (Figure 5), operational ranges, orientation, orientation effect on zone, and resolution. Vehicle-specific factors include vehicle braking distance [Williams 1999], vehicle speed, driver reaction time, direction of travel, and delays before an alarm sounds or the vehicle stops. Good design practice dictates that a proper analysis of worker exposures in a given work environment should be conducted before an intervention, such as installation of a PWS, is implemented [Schiffbauer and Mowrey 2003].

Depending on the local environment, it may be also prudent to consider whether the PWS may pose the danger of setting off electrically activated blasting caps, which could trigger an explosion. The Institute of Makers of Explosives publishes guidelines that should be consulted [IME 2001]. Underground mining requires that any electrical system meet Mine Safety and Health Administration guidelines [MSHA 2000] for safe operation in potentially hazardous areas where ignition of methane-air or coal dust is a safety concern.

HASARD: An Electromagnetic-Based PWS

The Hazardous Area Signaling and Ranging Device (HASARD) was the first NIOSH-developed PWS. HASARD was created to protect operators and workers around continuous mining machines [Schiffbauer 2002] (Figure 6).

A number of different types of commercially available PWSs were considered and tested for the job. They included radar, ultrasonic, capacitive, and visual types of sensors and systems. Testing indicated that, for underground mining applications, these devices were probably not appropriate for the following reasons. Any object that got into the detection zone would cause the alarm output to trip. This included walls of the mine and other vehicles that normally come in close proximity to, or in contact with, the machine. Falling coal, rock, and even the water sprays would set off the alarms. HASARD is an active system requiring a receiver on any object that needs to be protected. The active feature of HASARD is very important and is somewhat unique since most PWSs are of the passive type [Jurgen 1998]. Passive types of systems are triggered by all objects detected within their range. Active systems can minimize or even eliminate nuisance alarms.

The HASARD transmitter signal feeds into a wire loop that projects a uniform magnetic field about the dangerous area (Figure 7). The HASARD receiver decodes the magnetic field and calculates the distance to the magnetic field source. A microprocessor in the receiver determines when a local alarm should be activated and when data need to be conveyed over a short-range wireless link. This wireless link provides remote alerts and/or machine shutdown functions.

At surface mine sites, a HASARD transmitter on a large haul truck and a receiver on a worker could effectively prevent an accident. Large haul trucks frequently run over large pieces of earth, rock, and other debris. Rocks, thick dust, water sprays, mud, and debris are always being propelled off the tire treads. If the system triggered with every large piece of material it detected, it would quickly become a nuisance to the driver of the truck, causing him/her to tend to ignore it. A transmitter and receiver combination could be put on a wide range of objects such as people, edges of roads, poles, and other trucks. Since its development, the prototype HASARD system has been experimentally applied to continuous mining machines, haul trucks, and conveyor haulage systems, and it should be easily adaptable to shuttle cars, forklifts, and other vehicles.

The fundamental concepts used in HASARD are covered by two NIOSH patents:

• U.S. patent No. 6,810,353, "A Non-Directional Magnet Field Based Proximity Receiver With Multiple Warning and Machine Shutdown Capability"; and
• U.S. patent No. 5,939,986, "Mobile Machine Hazardous Working Zone Warning System."

The concepts behind the HASARD system have been implemented and refined in a number of systems available to the industry including the HazardAvert system from Strata Safety Products, the Matrix M3 system from Matrix Design Group, and the Coal Buddy system from Nautilus International.

NIOSH researchers are now developing an intelligent Proximity Detection (iPD) system based on the hardware and concepts developed with HASARD. The iPD system uses signals from multiple magnetic field generators to determine a two- or three- dimensional position of a wearable sensor. Using this calculated position, the iPD system determines which machine movements could potentially result in a collision between the machine and a person. All machine movements that could cause a collision are automatically disabled by the iPD system, but safe machine movements are allowed. In this way, safe mining is allowed to continue uninterrupted, and the miners are permitted to position themselves to best avoid other hazards such as unsupported roof and ribs or other pieces of equipment.

Other Issues

Maintenance repair and adjustments for a PWS can become a potentially significant issue for mine operators due to the sensitive and technologically advanced nature of these systems. First, the equipment design should be robust enough to handle the harsh mine environment. The PWS should be installed with protection from water sprays, roof falls, rib bumps, machine collisions, and any other expected potential consequence of the mine production environment. Regularly scheduled performance evaluations of the PWS must be done to ensure that the areas to be avoided are accurately identified and maintained throughout the work shift. The PWS should be designed to perform in a fail-safe manner; it should not allow the machine to operate should the protection zone, worker-worn device, or machine controller not perform according to specifications. The provision of fail-safe features, however, should not negate the need for regular testing and verification of detection zones and safety-critical features of all system components. Calibration of the zones around the machine should be done regularly. A calibration and maintenance log should be kept. Zone sizes, shapes, and areas, once established, should never be changed. It is recommended that installation, protection, calibration, and maintenance of any candidate system be thoroughly discussed with the manufacturer prior to purchase.

The personnel working near a machine employing a PWS should be regularly trained in all of the features of the system. The zones of protection and the resulting system responses to intrusions into warning and danger zones should be demonstrated as part of the training.

Fatalities at a mine site can cost companies millions of dollars in medical services, work disruption, legal fees, and lawsuit awards [Ruff 2007]. The tragic loss of lives and the cost benefit are two of many good reasons to employ a PWS on mining equipment.

Another important issue to be considered is equipment liability. It is recommended that any potential liability and other associated legal issues be addressed (e.g., installation of third-party systems on OEM equipment) with both the PWS and OEM vendors. Equipment manufacturers that distribute and sell products to the mining industry have a legal obligation to make products that are safe for their intended uses and advise people of any limitation or risks associated with using them.

Summary

NIOSH has compiled a collection of knowledge that provides insight into the proper selection, application, installation, and operation of PWSs on mining equipment. The information, when properly applied, increases the likelihood that the system will perform as desired and ultimately minimize injuries and fatalities of mine workers. The considerations have been summarized briefly in this document. This report should be useful as a repository of information, with references describing studies conducted by various researchers related to PWSs.

References

1. Anderson DL [1989]. Position and heading determination of a continuous mining machine using an angular position-sensing system. Pittsburgh, PA: U.S. Department of the Interior, Bureau of Mines, IC 9222. NTIS No. PB 90-265588.
2. Bartels JR, Ambrose DH, Gallagher S [2008]. Analyzing factors influencing struck-by accidents of a moving mining machine by using motion capture and DHM simulations. Presented at the 2008 SAE Digital Human Modeling for Design and Engineering Conference and Exhibition (Pittsburgh, PA, June 17-19, 2008). Warrendale, PA: SAE International. Document No. 2008-01-1911, 6 pp.
3. Brnich MJ Jr., Mallett LG [2003]. Focus on prevention: conducting a hazard risk assessment. Pittsburgh, PA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH) Publication No. 2003-139.
4. Burgess-Limerick R, Steiner LJ [2006]. Preventing injuries: analysis of injuries highlights high-priority hazards associated with underground coal mining equipment. Am Longwall Mag Aug:19-20.
5. Hebeisen S [1993]. Microwave proximity sensing. Sensors Jun:22-27.
6. Holden T, Ruff TM [2001]. GPS-based proximity warning system for mining and construction equipment. In: Proceedings of the 14th International Technical Meeting of the Satellite Division of the Institute of Navigation (ION GPS) (Salt Lake City, September 11-14, 2001). Fairfax, VA: The Institute of Navigation, Inc., pp. 517-525.
7. Hurteau R [1994]. Analysis of detection and obstacle-avoidance systems for vehicle operation in underground mines. In: IEA 94, Vol. 4, Ergonomie et Design (Toronto, Ontario, Canada, August 15-19, 1994). International Ergonomics Association, pp. 271-273.
8. IME [2001]. Safety guide for the prevention of radio frequency radiation hazards in the use of commercial electric detonators (blasting caps). Washington, DC: Institute of Makers of Explosives, Safety Library Publication (SLP) 20.
9. ISO [1991]. Earth-moving machinery, operator's field of view. Part 1: test method. ISO 5006-1. Geneva, Switzerland: International Organization for Standardization.
10. Johnson GA, Griffin RE, Laage LW [1986]. Improved backup alarm technology for mobile mining equipment. Washington, DC: U.S. Department of the Interior, Bureau of Mines. IC 9079.
11. Jurgen RK, ed. [1998]. Object detection, collision warning and avoidance systems. Warrendale, PA: Society of Automotive Engineers.
12. Mark R, Verhoef H [1999]. Collision avoidance system for large mining trucks. In: Proceedings of the Queensland Mining Industry Health and Safety Conference (Yeppoon, Queensland, Australia, August 22-25, 1999). Brisbane, Queensland, Australia: Queensland Resource Council, pp. 177-181.
13. Massa DP [1999a]. Choosing an ultrasonic sensor for proximity or distance measurement. Part 1: Acoustic considerations. Sensors Feb.
14. Massa DP [1999b]. Choosing an ultrasonic sensor for proximity or distance measurement. Part 2: Optimizing sensor selection. Sensors Mar.
15. MSHA [2000]. What you can't see could kill someone! Beaver, WV: Mine Safety and Health Administration, National Mine Health and Safety Academy, DVD 534-S.
16. NIOSH [2008a]. Systems approach to design. Date accessed: September 2008.
17. NIOSH [2008b]. Task analysis. Date accessed: September 2008.
18. NIOSH [2008c]. Visibility. Date accessed: September 2008.
19. Paine M, Macbeth A, Henderson M [2003]. The danger to young pedestrians from reversing motor vehicles. In: Proceedings of the 18th International Technical Conference on the Enhanced Safety of Vehicles (Nagoya, Japan, May 19-22, 2003). Paper No. 466. Washington, DC: U.S. Department of Transportation, National Highway Traffic Safety Administration, 11 pp.
20. Ruff TM [2000]. Test results of collision warning systems on off-highway dump trucks: phase 2. Spokane, WA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH) Publication No. 2001-100, RI 9654.
21. Ruff TM [2001]. Monitoring blind spots: a major concern for haul trucks. Eng & Min J 202(12):17-26.
22. Ruff TM [2007]. Recommendations for evaluating and implementing proximity warning systems on surface mining equipment. Spokane, WA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH) Publication No. 2007-146, RI 9672.
23. SAE [2002]. H-point machine and design tool procedures and specifications. SAE Standard J826. Warrendale, PA: Society of Automotive Engineers.
24. Sammarco JJ [1998]. Concluding evaluation of a continuous haulage guidance sensor. Pittsburgh, PA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH) Publication No. 98-161, RI 9646.
25. Sammarco JJ, Fisher TJ, Welsh JH, Pazuchanics MJ [2001]. Programmable electronic mining systems: best practice recommendations (in nine parts). Part 1: 1.0 Introduction. Pittsburgh, PA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH) Publication No. 2001-132, IC 9456.
26. Schiffbauer WH [2002]. Active proximity warning system for surface and underground mining applications. Min Eng 54(12):40-48.
27. Schiffbauer WH, Mowrey GL [2003]. The work zone analysis system: a tool for quantifying worker interaction with mobile equipment in dangerous work zones [Abstract]. In: NORA Symposium 2003 Book of Abstracts (Arlington, VA, June 23-24, 2003). Cincinnati, OH: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, p. 117.
28. Schultz S [1993]. Traffic and vehicle control using microwave sensors. Sensors Oct:34-49.
29. Sirois W [2003]. Asleep at the wheel. World Min Equip Mar:26-31.
30. Steiner LJ, Turin FC, Hamrick CA [1994]. Ergonomic and statistical assessment of safety in deep-cut mining. In: Peters RH, ed. Improving safety at small underground mines. Pittsburgh, PA: U.S. Department of the Interior, Bureau of Mines, Special Publication 18-94, pp. 124-132.
31. Wiehagen WJ, Lineberry GT, Lacefield WE, Brnich MJ Jr., Rethi LL [1994]. Work crew performance model: a method for evaluating training and performance in the mining industry. Pittsburgh, PA: U.S. Department of the Interior, Bureau of Mines, IC 9394.
32. Williams HS [1999]. Microwave motion sensors for off-road vehicle velocity data and collision avoidance. Sensors Dec.

---