Viewpoints Evolution of the Fire Protection Engineering Profession Over the Last 50+ Years
By Harold E. Nelson, P.E., FSFPE | Fire Protection Engineering
Within the last several decades, there
have been fundamental changes in the understanding and quantification
of fire. The fire protection engineer now has enhanced capabilities,
which enable basing conclusions on an engineering appraisal of the
potential harm or risk. The profession has made a series of important
advances that progressively have added to the engineers ability to
predict the potential impact of fire.
Four different, often overlapping,
approaches have been or are being used. These approaches are referred to
as specification, component performance, scenario performance and risk
appraisal. The first two of these approaches are rule bound; the last
two produce a quantitative understanding of the level of risk or safety
provided. The ongoing transition from design by rule to performance
appraisal has had a major impact on the professional quality and value
of fire protection engineering.
Fire protection engineering in 1946 was a
specialty niche of engineering dedicated for the most part to the needs
of the fire insurance industry. A youngster starting off in the
profession would compare buildings and other facilities to a book (e.g.,
codes or rating schedules) to determine compliance with the prescribed
document. No evaluation of the actual impact of the fire or the facility
was required or normally made.
As time progressed, it became apparent that
the specification approach unnecessarily constrained design.
Element-by-element component performance came into use. Under this
approach, any single element could be replaced by a different element,
if shown to perform at least as well as the specified element. In time, a
statement of required test results often replaced the original
specification, producing the now common form of codes and regulations.
At the same time, designers introduced
many innovative new materials, concepts and architectural treatments.
Many of the tests of fire protection or fire hazard of materials
attempted to imitate a fire exposure. But, the exposures used were at
best only a single point on a broad spectrum of potential fire
exposures. In some cases, the tests were based more on getting
reproducible results than scientific validity. Test results were seldom
expressed in engineering terms usable in an analytical analysis.
However, there now exists a massive collection of test results,
regulations and knowledgeable individuals to support the component
performance approaches. Unfortunately, virtually all of the test data
fail to provide quantitative information on actual fire properties.
Starting about the first decade after
World War II, various research laboratories undertook serious
investigations of fire phenomena and the simultaneous interrelationship
of several factors, such as burning rate and ventilation. Much of the
work addressed understanding and quantification of the development of
fire and fire products in enclosed spaces in buildings. Some of the
important outputs included the identification of the heat zones
occurring in compartment fires; the phenomenon of flashover; the
analytical impact of oxygen availability; fire plumes and entrainment
and other quantifiable aspects of fire development.
The investigations identified a number of
the physical properties important to an engineering analysis. These
include values such as temperature changes in strengths of heated
materials and the rate of heat release from a combustible material as a
function of incident flux. In some cases, the data could be determined
by using existing apparatus. In others, new apparatus such as the cone
calorimeter were needed. Key in all cases is that the measurements
return values of physical properties under the conditions of the
exposures of interest.
Individual solution methods were assembled in a form
(such as a fire model) where applying the fire property data could
produce an output showing expected level of safety. The working products
appear in the form of equations, models and other information needed by
the engineer to describe both the facility and the initiating
conditions. This approach is normally deterministic and tends to assume
that all descriptions and conditions are either constant or vary in a
described manner. Running multiple scenarios may be necessary to cover
the full scope of the fire potential.
The scenario approach can often,
depending on the selection of scenarios, provide the information needed
to draw valid conclusions on the probable course and outcome of fire in a
A more comprehensive set of conclusions involving both probability
and potential can be derived where the input data is expressed as a
distribution in terms of the environment faced. A Monte Carlo approach
is often used to reduce the number of runs of the computer code needed.
The modern practicing fire protection
engineer needs to have a comprehensive understanding of the
phenomena-based approaches to fire hazard analysis and use the most
accurate and suitable for the needs at hand. When using established
models, the engineer needs to understand the phenomena being modeled and
the sensitivity of the model to the data entered. The days of simple
cookbook code compliance are passing.
Further information can be found in The History of Fire Protection Engineering.1
Richardson, K. (ed.), The History of Fire Protection Engineering, National Fire Protection Association, Quincy, MA, 2002