Technology Driven Safety / Innovations to Improve Safety and Productivity
By Larry Mahan, President & COO, Sky-Trax, Inc.
Everyone wants to be safer - vehicle manufacturers, managers, workers, union leaders, and regulatory bodies. Unfortunately, too many safety programs seem to be in the dark ages, relying primarily on driver training and compliance, and not much on technological solutions that are readily available. Ask anyone familiar with industrial operations, and you’ll hear that when safety interferes with efficiency it often ranks 2nd on the priority list.
Leading businesses know that safety is not incompatible with efficiency – instead, it can enhance productivity. These companies use safety as a competitive advantage. Since the majority of serious accidents involve stability incidents and vehicular collisions with pedestrians, the new safety systems solution will integrate technology addressing both problems with intelligent speed control, vehicle tracking, and pedestrian tracking that can provide both improved safety and increased productivity.
Dangerous Work
Employees working in or around forklift equipment are statistically exposed to one of the most dangerous work environments in industry. While forklift trucks are involved in only 1% of all industrial accidents, these accidents are responsible for 10% of physical injuries. In the US, industrial forklifts kill an average of 100 people per year and seriously injure tens of thousands more. According to US Dept of Labor, OSHA forklift violations (29 CFR 1910.178) are one of the most common OSHA citations, resulting in nearly $1,000,000 in fines in 2005.
Forklift accidents are not a uniquely American problem. Where reliable statistics are collected, similar numbers are seen around the world. In the UK, 25% of occupational health and safety fatalities are attributed to workplace transport accidents including 70 fatalities in 2004/05.
Cost of Accidents
While safety is a goal of its own, the business costs of forklift accidents are increasingly a bottom line issue. The direct cost of accidents with injuries includes medical costs, lost work time, and reduced productivity in the aftermath, ensuing investigation, and implementation of improvements. The U.S. Department of Labor reports the average cost of a recordable injury in the United States is $35,000. In the most serious cases there will be litigation costs, OSHA fines, and repair costs for damaged equipment and facilities. A repeated pattern will lead to higher insurance and workers compensation premiums, adverse publicity, higher recruiting and training costs, low productivity with higher overtime wage expense, and low morale.
Based on data from the National Safety Council and the Bureau of Labor Statistics, Alert Safety Products has calculated the costs of an industrial vehicle accident with injury as follows:
- The median number of lost work days is seven;
- 22.8% of the injuries will result in more than 31 lost work days;
- The direct cost of lost productivity will be $47,000;
- Equipment damage adds another $6,400 in repair costs;
- An OSHA fine will be a minimum of $3,500;
- If a fatality results, costs will exceed $1,000,000.
It is clear that a strong business case can be made for investing in safety systems and technology to eliminate hazards and accidents.
Hazard Control Strategy
To systematically improve safety, safety engineers recommend the following strategy for hazard control: 1) Remove, 2) Guard & 3) Warn.
Strategy 1 strives to remove the hazard by eliminating the danger by designing it out of the system or environment. For example, close or eliminate a blind door opening into a warehouse to eliminate the possibility of a pedestrian walking into forklift traffic.
If the hazard can’t be eliminated, Strategy 2 is to guard the hazard by installing safety apparatus to prevent exposure to the danger. Even if pedestrians and drivers are distracted, guards can protect both from harm. For example, elevated loading docks may not be eliminated, but automatic barrier guards can be installed to prevent fork trucks from dropping off a vacant receiving dock. A good example is the Rite-Hite RHH dock leveler with the Safe-T-Lip barrier, which prevents forklifts from running off an open dock and can stop a 10,000 lb. forklift traveling at up to 4 mph.
Finally, if the hazard can’t be eliminated, or exposure to the danger prevented, Strategy 3 is to warn workers of the hazard. Ideally, specific alerts should be communicated only to those directly involved in the hazard situation and only where and when a danger actually exists. For example, warning lights can be installed at blind corners to warn of oncoming forklifts with a system like the Wickham Fork-Alert™ product.
Safety system designers now have new technologies to consider for hazard control, particularly for detecting collision and speeding hazards.
Collision Avoidance: The Time/Space Problem
Traditionally, collision avoidance strategies have relied on procedures and driver training and compliance. These will always be elements of workplace safety programs, but collision statistics clearly indicate that training, signage, and floor markings for traffic control are not enough. Avoiding collisions between powered vehicles, or between pedestrians and vehicles, is based on a simple principal: pedestrians and vehicles must be kept separated in time and space.
This is the key to understanding the safety versus efficiency dilemma. Time separation and space separation, which have always been effective, are inefficient, while simultaneous, real-time monitoring and control can improve safety and efficiency.
As an example, imagine a warehouse that allows powered vehicles to operate during the day shift, while precluding pedestrians from entering the facility. Then, on the night shift, no IUV’s are operated, while workers have full access to the facility. This plan implements a time-shared facility.
The alternative would have workers and vehicles operating during both shifts, but impervious barriers separating the two – the space shared facility. Neither arrangement proves practical in real life. A combination of time and space separation is needed, and that’s where modern technology steps in.
Collision Avoidance: A Technological Approach
Sensor technology today provides the ability to detect and track the location and proximity of vehicles and pedestrians in industrial facilities. Sensors that detect vehicles and people have the ability to work in localized areas, over large areas, and throughout entire facilities. Whether tracking pedestrians or trucks, the best safety technologies will have the following capabilities:
- Location determination accurate to within a meter or less
- Velocity determination
- Determination of orientation or direction of travel
- Ability to identify vehicles
Technology for Detecting Trucks
A wide spectrum of technologies can be used to detect trucks. These can be functionally categorized as follows:
- Presence Detection
- Presence & Distance Detection
- Presence & Identification
- Location and Tracking
Presence Detection
Presence detection sensors indicate that a vehicle is within the detection distance or zone of the sensor. In most cases, there is some ability to configure or engineer the detection distance. Inductive or capacitive proximity sensors and photoelectric sensors, all of which are familiar to automation engineers, fall into this category.
Proximity sensors detect the presence of a large metal mass like a truck within their detection range – usually less than 1-2 meters. This short detection range makes this type of sensor most applicable for detection at “chokepoints” like dock doors. Photodetector sensors are also used for this purpose. Fork-Alert ™ and Alert Safety Products offer warehouse safety products based on this technology. Fork-Alert employs an invisible infrared light beacon mounted on the top of the vehicle. An infrared receiver can detect the infrared light up to 25 meters away and trigger warning lights or audible alarms for pedestrians and other drivers. Alert Safety Products combines the light source and the detector in a single unit that is mounted on a wall or post. Strips of reflective tape are applied to both sides of forklifts so the vehicles can be detected when traveling by the sensor.
Microwave sensors work similarly and can shape the detection zone to match an area of interest. For example, both Door-Man and Alert Safety Products offer warehouse intersection warning products using microwave sensors. Four sensors and a warning light are hung above an intersection with microwave sensors aimed in all four directions. A vehicle approaching the intersection is detected and triggers the appropriate warning light.
Presence and Distance Detection
The next class of sensors not only detects a target but can accurately measure the distance from the sensor to the object. The principal technologies here are
ultrasonic range sensors and laser time-of-flight sensors. Ultrasonic sensors emit high-frequency sound waves which are too high for the human ear to hear. When waves are reflected back from a solid object, the sensor can determine the distance from a few centimeters up to 10 meters.
Laser systems can measure distances with higher accuracy and longer ranges. Found typically in high-end safety systems on automated guided vehicles (AGVs), these sensors measure distances very accurately with time-of-flight calculations on the reflected laser light. A commercial safety laser scanner from SICK GmbH can be programmed for different scanning areas and distances and configured to have both warning and emergency stop thresholds.
Presence and Identification
RFID technology has received extensive press for inventory tracking applications in warehouses. Typical applications use passive RFID with inexpensive tags that can be read (detected and identified) by an RFID reader, but the read distance is small - usually less than a meter. Longer read distances of up to tens of meters are possible with active RFID systems. These systems detect and identify a tagged entity within the read zone of the RFID reader. This capability has been employed widely for security and access control applications. An instructive paper describing the capabilities and uses of all the RFID technologies can be downloaded from the Axcess International website (A Best-Fit Application Guide to Active RFID System Alternatives).
Location and Tracking Systems
Systems that can track forklifts and know the location at all times will have the greatest impact on creating the next generation of safety systems for warehouse operators. This new technology is known as real time location services (RTLS). Radio frequency RTLS and optical RTLS systems are available today.
The RF RTLS tracks vehicles that carry an RFID tag. This identifying tag can be read simultaneously by multiple RF receivers in the detection region. Using one of several different sensing algorithms, a high speed computer applies triangulation techniques and computes a location estimate for any tag that is read by three or more sensors. Overall accuracy in industrial buildings currently is approximately 2 to 5 meters. Leading suppliers of RF RTLS technology include AeroScout, WhereNet, and Ekahau.
The latest technology for tracking vehicles in warehouses is machine vision for optical RTLS. Machine vision (image processing) has been used widely in industrial automation for high speed package sortation, automated product inspection, and robotic guidance for the past 20 years. Sky-Trax Inc. has adapted this technology to accurate and reliable tracking of forklifts inside buildings. With the Sky-Trax Indoor Position Sensing™ (IPS) technology, vehicles are tracked in real time to accuracy of 10-25 cm. Important to many safety applications, IPS systems also know the speed and orientation (direction) of each tracked vehicle.
The IPS system employs a small image sensor mounted on each vehicle to look up at the ceiling where an array of low-cost printed tags are always visible (see illustration above). The image sensor includes image processing intelligence to capture and analyze pictures of the ceiling several times per second. From ceiling scene analysis, IPS calculates exact X, Y position as well as angular direction of the vehicle. From location changes noted from frame to frame, velocity data is also calculated. This information is transmitted wirelessly to a computer which collects data on the location and status of all vehicles in real time.
Technology for Detecting Pedestrians
There are two types of pedestrians in industrial settings; the first is a person who works on foot in an area with forklift traffic, such as a hand truck operator or an order picker. It is reasonable to expect these workers to wear or carry a safety device if required by work rules. The second type of pedestrian is the visitor who does not work in the area and is less familiar and aware of the dangers. Examples of this type are vendors, front office staff, and contract maintenance workers. These pedestrians are less predictable and controllable, creating a heightened risk level. Visitors are less likely to understand and comply with a requirement to carry a transponder or wear a safety device when entering forklift traffic areas. Safety system designers must address the risk and establish effective measures for both classes of pedestrians.
The Accident Research Centre at Monash University (Victoria, Australia), a leader in evaluating technology for preventing forklift-pedestrian accidents, prototyped a system employing a simple RF-tag placed in safety vests worn by warehouse workers. An RF receiver was installed on each truck alerting drivers to the presence of any workers within the detection radius of the receiver. The researchers found this wearable RF tag prototype to be a low cost solution that they recommend to be used along with other safety measures. (Industrial Forklift Trucks – Dynamic Stability and the Design of Safe Logistics, 2003)
A product from ProxAlert takes the Monash University prototype concept to the next step. ProxAlert operates with an RF transceiver installed on each vehicle. A similar battery-powered transceiver is clipped onto any pedestrian prior to entering the warehouse. As illustrated; the transceiver range creates a virtual protection zone around the vehicle or person. When the zones intersect the transceivers energize a switch closure to enable a warning signal for both the pedestrian and the vehicle operator.
This approach is a viable solution for worker pedestrians; however, it may be difficult to assure that visitor pedestrians will carry or wear a proximity warning device; for example, customers in warehouse stores. Machine vision technology may provide better detection and tracking controls for this since it does not require any compliant action from the pedestrian.
Tracking Pedestrians with Machine Vision
Over the last 10 years, image processing research has made great progress in developing methods for detecting, identifying, and tracking pedestrians in video images. Driven largely by the need for smart surveillance and security systems, the technology has moved beyond military uses and is now used in commercial applications. Brickstream (Atlanta, GA) has marketed a pedestrian tracking system since 2002 that tracks and analyzes the movement of customers in commercial buildings. Processing images from overhead cameras, the system determines the number of customers entering a store and the exact paths taken by customers shopping in the store. In retail and banking applications, the technology is used to track queues of customers and to signal when more check out lanes need to be opened.
While this technology has not yet been applied to collision-avoidance systems, it can be expected in the near future. Because industrial spaces are less diverse and more orderly than public areas like streets and stores, the application of machine vision for pedestrian tracking in warehouses should be very feasible. Location accuracy is likely to be less than a meter, and the pedestrian’s direction of movement and speed will also be provided. These capabilities are needed for detecting pedestrian hazards in areas where forklifts operate.
Speed Control for Safety
Speed is usually a contributory factor to both collisions and stability-induced incidents (tip-overs) which together represent well over half of all serious accidents. Monash University researchers report that 75% of side tip-overs occur when a forklift is empty, leading them to conclude that these incidents are due more to speeding than other causes. Systems to control speed will be a significant way that technology will improve safety. The best solutions will do this without impacting productivity.
A medium size forklift has the same mass, and potential for destruction, as a small dump truck. Moving at 10 mph, a 5000 lb. forklift truck with a 4000 lb. load has the same momentum as a Cadillac Eldorado moving at 20 mph. This highlights the need for speed control to provide safe stopping distances for specific conditions. Like automobiles, forklifts cannot safely stop on a dime; and panic stops create additional hazards with loss of handling capability and unstable loads.
Stopping distance is a function of speed, mass, driver reaction time, driving surface conditions, and braking system performance. Speed and reaction time are the key variables that can be controlled by the driver, and are the two areas where technology can help. Technology can provide advance warning of hazards (earlier reaction time) and can directly limit speed to assure adequate stopping distance based on location, load, vehicle type, and known hazards.
Safety system analysis begins with understanding safe stopping distances. If an empty forklift truck moving at 10 mph requires 40 feet for a safe stopping distance, the driver needs to allow at least 40 feet to react once a pedestrian hazard is recognized. If this is not practical, for instance at blind intersections, speeds need to be reduced in order to allow for proper stopping distance.
Speed limits are established by rule and drivers are expected to recognize and obey the appropriate speed limits. As on the highways, these rules are often violated and difficult to enforce. Many forklifts do not have a speedometer, and speed limits are not usually posted. Complicating the situation, the safe speed changes as mass changes with loads, and driving surface conditions vary at different locations in the facility. Unfortunately drivers often experience more pressure to be efficient (drive fast) than to be safe.
Some believe that installing speed limiters on trucks is the solution to control speeding; however, this creates a dilemma. While reduced speeds are necessary in some areas and conditions, speed reductions are an unnecessary restraint on productivity in other areas and circumstances. Speed control must balance productivity with safety to permit a vehicle to travel at the fastest safe speed for the specific location and conditions. Allowable speed must vary as the vehicle moves from location to location and as conditions change. This can be accomplished with technology that monitors the conditions, location, direction, and speed of the vehicle and also of all the other vehicles and pedestrians in the area. The intelligent safety system (ISS) will include a direct means for alerting drivers and pedestrians when hazards exist and a direct means of automatically limiting speed.
The Intelligent Safety System
The Intelligent Safety System will utilize data collected from on-board sensors, and facility monitoring systems. ISS technology will rely on sensors to:
- Accurately track the location, direction, and speed of all vehicles.
- Accurately track the location and movement of all pedestrians.
- Know the status of each vehicle (driver ID, load, current task, impact events, etc.)
Given an abundance of real-time data, ISS software intelligence will predict collision hazards and initiate action to warn or eliminate hazards. ISS will have communication and control links with drivers for hazard alerting, with trucks for automatic speed limiting, and with facility safety systems for intersection control and other intelligent warning systems.
This image illustrates the benefits of an ISS. Safety zones are defined and configured in software; for example, safe, caution, and danger zones are designated for intersections as illustrated with the green, yellow, and red shading in the illustration. Truck A approaches the intersection from a safe zone. The ISS, which controls the intersection traffic lights, gives a green light to Truck A and pedestrians even though Truck B and Truck C are in the intersection’s caution zones. Because it knows the exact location, direction of travel, and speed of vehicles and pedestrians, ISS determines that Truck C is moving away from the intersection and presents no danger to either vehicle or the pedestrians. Likewise, Truck B’s orientation and speed indicate that it is putting away a load and not entering the intersection.
An unintelligent intersection safety signal system would illuminate caution or stop signals for Truck A based on the proximity of Truck B and Truck C. Importantly, an ISS preserves productivity - Truck A will not be slowed - while establishing a safer workplace.
Safety Wins
Everyone is in favor of increased safety, especially when it enhances productivity and the bottom line. Technology now offers the ability to reduce vehicle stability accidents and vehicle/pedestrian collisions, using new sensors and intelligent, automated safety solutions. As this progress is increasingly recognized by industry leaders, regulatory agencies, and safety researchers, it is expected that industrial vehicle manufacturers will incorporate the new safety technology into their products, and that market-leading companies will take advantage of the new capabilities. For a model of safety progress, look at the systems the automotive industry is designing into cars and trucks.
Technology cannot replace the basics of safety – strong management commitment, good operations design, training, and accountability. But when safety is introduced into a safety conscious culture, technology will provide the tools for transforming safety into a competitive advantage.
For more information or discussion on this topic contact Larry Mahan by phone at (302) 395-9540, by email at Larry.Mahan@sky-trax.com or visit www.sky-trax.com.
Want still more? Subscribe to our Electronic Edition for a year's worth of industry-specific information, six downloadable PDF issues and access to past issues, all for only $42.00 (USD). Click here for more details on how to subscribe to our Electronic Edition.
For back issues of this publication, call 518.329.0067 or visit our back issues page.
