Systems for Today's Threats

Facility managers should start with a thorough assessment, then choose an emergency notification system that cannot be compromised.

• By Jerry Laws
• Feb 01, 2005

Editor's note: Notifying employees and/or the public in an emergency situation is a tougher challenge in the post-9/11 world. Many facilities face an expanded threat profile that cannot be adequately addressed by a siren or some other general alarm signal. MadahCom Inc. (www.madah.com) of Sarasota, Fla., under the direction of CEO Reuben Ben-Arie, is a leading provider of advanced alerting systems to the U.S. Department of Defense and other federal agencies. MadahCom President Dr. Alan Avidan discussed notification technologies, decision-making, and responsibility for industrial safety managers in a Dec. 1, 2004, conversation with Occupational Health & Safety's editor. Excerpts from the conversation follow.

What technologies are available today enabling employers to carry out emergency notification?

Dr. Alan Avidan: There are several means of disseminating emergency information to people, including fire alarms and bells, sirens, phones, computers, and Mass Notification Systems (MNS), which will be detailed later. But to disseminate emergency information you must first become aware of the threat and then be able to offer actions to mitigate the threat. You need to have trained personnel that can accept and process information received from sensors or staff about the perceived threat in order to alert the public and activate the appropriate notification message. Once the threat and response are determined, the message must be delivered. Voice is a very effective means of delivering emergency information, while text shown on an LED sign comes second. Going back to "ancient times," we have used only fire gongs and bells; you still find a lot of these systems around today. It was simple: The gong or bell starts ringing, and it's time to evacuate the building because apparently there's a fire. Today, the realm of threats has evolved to such a higher complexity level that using these devices as the main information dissemination tool can ultimately work against you and can increase the size of the problem you have on hand. I'm thinking from the perspective of the facility security manager or facility owner.

Meaning there is such a broad range of threats that you can't tailor the warnings specifically enough?

Avidan: That's correct if you have only devices that generate basic tones or wails. It used to be that the main threat was fire, and if it wasn't a fire, then somebody would come around and tell you what it was. The natural human instinct with fire is to leave the building as soon as possible. But if you are under a gas attack--I don't mean necessarily by terrorists, although today this is considered a credible threat, but because a tanker truck just overturned on the highway not too far from you and it's spilling large amounts of chlorine or some other gas. That chlorine gas is approaching you as a cloud; should that fire gong go off and your employees consequently leave the building, they may be heading right into harm's way.

That is why, quite a while back, different organizations including OSHA advocated abandoning the simple bell or siren sound in favor of voice notification. Because obviously with voice you can disseminate complex information and tell people not only what the threat is, but also how to respond specifically to that threat in order to minimize the consequences.

You're talking about employees' computers and phones, beepers perhaps? Especially at sites with multiple buildings, where a manager can't stand on a balcony and make an announcement.

Avidan: Time is of the essence. If there's a bomb threat that has just been sounded, or perhaps notification of an aircraft in the air heading your way, you need to respond instantly. For that to happen, a lot of technology needs to be in place.

A telephone, a beeper, or a computer screen can be used to disseminate sound and visual information, but there are some serious limitations to these methods of dissemination. First, you have to be by your phone, someone has to know which one, and someone needs to dial, or you can use a computerized dialing notification system. Second, the phone system has to be operational. It is quite common for cellphone systems to collapse during an emergency due to capacity limitations and or loss of grid power. As we've also seen in the World Trade Center 9/11attack, the system may collapse directly because of the emergency itself.

Cellphones and derivatives are not considered reliable means of reaching people during emergencies. Land lines are a little more reliable perhaps, because they have their own electricity. The problem is that most people are not by their land phones most times.

What is meant by "reliable"? Are you talking about reaching 100 percent of the employees or of the target population?

Avidan: Obviously, 100 is a nice goal to have, but a lot of emergency managers would be quite content to reach the majority of people. Often, 70 or 80 percent reach would be considered a success. When you talk about using devices such as PDAs, beepers, cellphones, etc., we're really talking about success rates in the 10 to 20 percent; that's about the best you can expect.

The most effective way to get emergency information to people is through an MNS alerting system, which utilizes voice messages inside buildings through speakers and outdoors in open areas by use of speaker towers. These are high-output electronic sirens, much like ones used in the Midwest for tornado warning, except that rather than having a single siren tone, they can output high-fidelity voice that can be directed in a specific dispersion pattern.

Could it be delivered within a plant? The speaker might be the safety manager or CEO.

Avidan: In the workplace, say a 30-building campus, if information has been received about a bomb planted in one of the buildings, your emergency coordinator/security officer wants to alert only the right people quickly and guide them to safety. He or she doesn't want to alert everybody in order not to provoke panic. He doesn't even want to stop work in unaffected areas of the campus.

I'll give you an example: At one of our major industrial customer sites, someone activated the site-wide warning. Even though it only affected a single building, everyone got the message and left their building to a designated area. It took a long time to get everyone back to work with all the excitement, and there was a significant loss of productivity.

When you have such fine zoning control like you do in an MNS, you can address specific buildings, even specific floors or areas within or outside of buildings in any combination, depending on your needs.

Often automating the voice message process can have a big payoff and reduce panic even among well-trained operators or security personnel who can lose their cool in an emergency situation. For example, a triggering of a gas sensor can be tied to automatic release of alerting messages in the affected areas and in pre-defined areas (such as first responder hangouts).

Are these systems tied into the community emergency responders' network?

Avidan: Our [wireless system] can tie into local and regional emergency management centers because they are digital and talk common protocols. CAP (Common Alerting Protocol) is one such protocol. However, is not widely used yet. This is why those voice-capable speaker towers may be the most effective answer for wide area coverage to the population, and also because they allow that information to get directly to the local first responders.

When we deploy a system using our wireless technologies, we're not only utilizing large area coverage devices [speaker towers], but also providing localized coverage in individual buildings by placing our wireless transceivers, allowing us to cover any particular areas with great granularity.

So you may have the east wing, second floor, of a building covered by one transceiver driving a few speakers, strobes, and LED signs as needed, while the west area is covered by another transceiver. So you can direct a specific message to the east zone or the west zone, or the second floor, the fourth floor, but not the third floor, which could get a different message. You can have event-specific and/or time-specific zoning schemes combining any areas, indoor and outdoor as needed at that time.

What is the maximum range for these speaker tower devices?

Avidan: These very high-output audio devices can sound for several miles, but intelligible voice normally stops at 1,000 to 2,000 feet, depending on conditions.

That's the best you can do?

Avidan: Voice intelligibility is a complex function that depends on the medium [air] density variations that are caused by temperature and humidity fluctuations. Distortion creeps in as the sound propagates. Under ideal conditions, you can achieve up to half a mile or even a mile. The problem is that you can't plan for ideal conditions.

Do you have to plan for the worst conditions?

Avidan: You can't plan for the worst case scenario because you end up with a system that is not economical. One way for a security manager to compromise is to think of certain probabilities; for example, "I want to reach at least 85 percent of my people 95 percent of the time." This can then be a rough guide to the design requirements.

You indicated this would be a way of also alerting the outside responders. They'd find out the same way everybody else would?

Avidan: Our system is totally modular so you can place a transceiver at the fire house. This way they can hear all the messages given out to the public or employees and also get messages that are directed specifically at them.

What about special populations, such as disabled workers or visitors, non-English speakers, etc.?

Avidan: You hit on a very good point, because there are people in our society who need a little more help or need information in a different way than those of us who are blessed to have all of our senses fully functional.

I think it's in our social weave that we need to provide visual information to people who have hearing difficulties. Ten percent of the population is considered hearing impaired to some degree, and it's our moral duty to make sure that these people get all the proper emergency information needed. So if you provide voice information to those who have a hearing capability, you mustn't forget to also provide text visually. This is where devices like LED scrolling signs play their most important role.

Under duress conditions, all of us who have proper hearing can become hearing impaired. Either the background noise levels are very high or someone is screaming in our ears in panic. The most calming thing you can see at that time is the LED sign saying, "There is an emergency condition. Please leave the building from the west end." Calming because you get a sense that the situation is under control.

Right. And, depending on the facility, you might need quite a number of signs.

Avidan: You would definitely need to place them strategically, primarily in high-traffic areas, by exit areas, inside large conference rooms, etc. Now, if this is a work situation and people are generally there every day, what would probably be wise to do is to have a plan that exercises these devices often. It lets people know how to use them and how to absorb the information coming at them. This is useful because, as with all people, things that happen to us for the first time are usually more difficult to comprehend. If you repeat, you drill, and you use these devices on a regular basis, you teach people how to use them. And that's very, very effective.

How frequently should employers do this?

Avidan: That's a good question. OSHA for instance mandates for workplaces that systems be used every two months if they're not supervised for a drill or practice. For supervised systems--once annually. In my view, this is way too far apart. Systems of an emergency nature, notification systems, would be best served if they are also used for ordinary uses. If you can announce a barbecue on a Friday afternoon using your emergency system, you have done yourself a big favor. Plus, you've tested the system. If there are areas of the system where people didn't get the message, you're going to hear about it and you'll be able to make adjustments. This way, when you need [the system], you're more likely to get effective use out of it.

We really advocate a weekly test of this type. It's probably wise to vary it rather than have it, say, every Friday at noon, though there's a slight advantage to doing that because people then anticipate the Friday noon test. But it's also nice to do it on a varied basis so that people become accustomed to having information thrown at them from these specific notification devices, speakers, LED signs, or sometimes even strobe lights. Having people become accustomed to emergency info presented to them on a non-emergency basis is important. Obviously, it's also important not to make this a trivial event [and thus induce warning fatigue].

Does the process start with an assessment by the facility manager or some other person--an assessment of what the possible threats are, what the target population is, and where the population is spread out?

Avidan: The most important thing the facility manager can do first is to conduct a vulnerability self-assessment of the facility. And anyone who says, "No, I already have it done and I know what my threats are" is not realistic because there are many threats that are not obvious, especially at this time and age. Also, many of the threats can emanate from outside the facilities.

I may be a security manager or the facility manager at a site, and I know that we have propane tanks that can explode. I'm aware of that. I've taken precautions, at least in my plan, so I'm prepared for that. But, as we've said, the threat could come from the outside. It could be a terrorist action or it could be a natural disaster or a man-made disaster of large proportions. It could be just that tanker truck that tipped over on the highway, a common event, or a train accident that discharged some chemical intermediate that you've never heard of, and suddenly a major threat is in the air heading your way.

Who should prepare the assessment? Should it be a team of inside managers and outsiders with expertise in this area?

Avidan: I would say it's probably wise to bring in a professional company to conduct threat and vulnerability assessments. First, to just get an outside expert opinion; second, management is more likely to spend the required funds when an outside expert is brought in. The main objective of the assessment is to determine what else can be done to complete a response plan to meet any one of identified perceived threats. Attempt to think out-of-the-box regarding threats you've never thought of before.

Once you have a response plan, you want to try to automate the responses when you can. And automation means pre-recording messages that can be disseminated by the emergency notification system anywhere on site, wherever people are. But to properly deliver critical information during an emergency, you've got to have a properly designed system that is able to disseminate emergency information under these conditions.

The common fire system is clearly not an effective way to do this, as mentioned earlier--the immediate response is to evacuate the building. In many emergency events of today's threats, the best response may often be to hunker down, shelter-in-place. You need to receive information about the threat: if it's a gas or a biological agent--weapons of mass destruction, commonly called CBRNE (chemical, biological, radiological, nuclear, or explosive) threat.

If you know what the threat is and you're being told in the message reaching you how to respond, you're more likely to stay calm and to respond correctly to it and hence increase your chances of survival without harm. So the fire system should be part of the emergency alerting arsenal but should never be considered as a valid mass notification system.

What standards apply to emergency employee notification?

Avidan: OSHA does have a regulation, specifically OSHA 1910.165, that's titled Employee Alarm Systems. It's a bit archaic, but it does advocate that you have recognizable and distinct ability to tell people what the threat is. But, as we discussed earlier, the varied range of modern threats and their respective appropriate responses are such that anything but intelligible voice and perhaps comparable text on LED signs, coupled with strobe lights to get your initial attention, is less than sufficient.

NFPA is working on a mass notification standard that will be coming out initially as an annex; I'm part of that task force. It will be published in about a year's time. As an annex it will be non-mandatory, informational only, but that annex will have defined a requirement for intelligible voice disseminated via wireless systems that are immune to interference.

Wireless is preferred because adding a new wired infrastructure to an existing facility can be a nightmare, and you can't "cut" the wireless signal. But you have to be real careful not to implement a system based on a single-frequency, FM modulated wireless (such as walkie-talkie or LMR) because of its extreme vulnerability to interference, jamming, and spoofing. You'd be much more secure with a frequency-hopping spread spectrum type of RF of the type the military uses and available commercially.

This article appeared in the February 2005 issue of Occupational Health & Safety.

This article originally appeared in the February 2005 issue of Occupational Health & Safety.