But, there are times when automatic sprinklers are not sufficient to limit property damage or reduce the risk of a process to an acceptable level. Some equipment and processes may warrant additional levels of protection beyond what a sprinkler system alone can provide. In the early years of sprinkler system development, high-risk locations and hazards were identified as a result of a loss. Lessons learned from these events resulted in additional protection being provided on a go-forward basis, or as an immediate retrofit when risk was found to be particularly severe.

Nowadays, the fire protection engineering discipline has established robust techniques available to identify and evaluate hazards and risk. SFPE’s Engineering Guide to Fire Risk Assessment provides the framework for identifying hazards and quantifying risk. Another potential source for guidance is SFPE’s Engineering Guide to Performance-Based Fire Protection.

Hazard identification usually focuses on equipment and processes that represent a heat or ignition source or a fuel source. Once the hazards are identified, various protection options or scenarios can be contemplated. The next step is to estimate the risk of the various scenarios. Both the frequency and severity or consequence of the scenario influence the overall risk. The process of estimating risk can be an involved and time-consuming task. For example, the engineer is faced with obtaining or calculating frequencies and probabilities. And, in many cases, this data is not readily available in the public realm.

In the commercial property insurance industry, a fire hazards assessment is usually conducted in lieu of the fire risk assessment. With this, the uncertainties of whether a fire will occur or a system will operate are not explicitly addressed. Murphy’s Law is generally assumed to hold true – the first assumption is "anything that can go wrong, will go wrong.” Therefore, ignition is always assumed. The fire hazards assessment will typically include identifying the hazard(s) and exploring protection options. And, protection options are not limited to fire suppression. A wide range of improvements from system design changes to reduce the risk to installing safety interlocks to shut down equipment in a safer manner could be recommended.

In some cases, reductions in property damage and/or business interruption may be calculated to show the value of various protection options. Of course, with these cost benefit estimates, the consequences are quantified but not the frequency. The type of extinguishing agent can also have a significant impact on property damage and business interruption. Some equipment may be more susceptible to damage by one extinguishing agent versus another. Also, potential extended clean-up issues from one suppression agent over another could result in a quantifiable increased interruption to business operations.

Looping back to the start of this article, we revisit the question, "When are ceiling-level sprinklers not enough?” Three such examples are included in reply:

• Equipment that is handling a combustible material and represents an inherent ignition source where ceiling sprinklers may be sufficient protection by "Code” but the lack of ignition source control results in an unacceptable risk (so, we do consider frequency after all). A classic example is a pneumatic conveying system requiring proper grounding and bonding to prevent the buildup of static electricity that could ignite the combustible materials being conveyed.
• Equipment containing combustible materials that are shielded from ceiling sprinklers, such as large ovens containing combustible materials.
• Equipment containing combustible materials of a type, quantity, or form that if ignited cannot be controlled by ceiling-level sprinklers, such as dip tanks containing flammable or combustible liquids, processes containing flammable gases or equipment or areas representing explosion hazards. This is where we get into some very serious hazards.

One of the more significant challenges in assessing industrial fire hazards is providing cost-effective solutions with a proven track record. No one wants to reinvent the wheel – it is just not that efficient. NFPA occupancy-based documents are a good source to find this information. And, of course, another source are those commercial property insurers who often have internal or external guidance to offer – much of which was literally vetted by paying the claims on actual fires.

Paul Hart and Bruce Clarke are with AIG.