From the Technical Director New Zealand - Vanguard of FPE
By Morgan J. Hurley, P.E. | Fire Protection Engineering
New Zealand was one of the first countries to adopt a performance-based regulatory system. New Zealand published a performance-based building code in 1992. A set of calculation methods that could be used to demonstrate compliance with the code was published as the Fire Engineering Design Guide in 1994. While these documents were not the first performance-based codes or design methods, their publication established New Zealand as a leader in the practice of fire protection engineering.
Fire protection engineering was not the only building system where a performance basis was established. Other areas of building design that were previously regulated by prescriptive means were permitted to be designed according to a performance basis beginning in 1994.
However, although the movement towards a performance-based regulatory infrastructure was a boon to the fire protection engineering profession in New Zealand, costreduction pressures from building owners and prime design professionals have created new challenges for the local fire protection engineering practitioners as well.
Performance-based design allows more flexibility than prescriptive design does. While the construction costs of a building that is designed according to a performance basis will generally be similar to those for a building that is designed in accordance with prescriptive codes, design fees are typically greater. This is because the development and evaluation of a performance-based design usually take more time than that required for a prescriptive design. Tools such as computer fire models, which are key elements of performance based designs, are generally not used in the development of a prescriptive design.
These design tools typically require many input variables, most of which will not be known with complete certainty. Therefore, when using these tools to help make decisions that will impact public safety, it is necessary to make sure that reasonable variations in input variables could not lead to a different conclusion. This can be done by selecting bounding values; however, setting all variables to bounding values can represent a set of conditions that could never occur and hence a design that is needlessly expensive.
Another method of evaluating uncertainty is to conduct sensitivity or switchover analyses. These analyses require exercising the tool multiple times, where the values selected for input variables are changed and the effect on output is observed. However, these analyses can take a considerable effort, as sensitivity and switchover analyses require using tools repeatedly with different input variables. Also, as the complexity of tools increases, the number of variables typically increases as well. Whatever method is used, uncertainty analysis is critical to provide the designer confidence in the results obtained.
Unfortunately, some clients of engineering services in New Zealand will not pay fees that enable their fire protection service providers to perform a comprehensive uncertainty analysis. This puts designers in a difficult position.
A possible solution to this problem is for designers in New Zealand to insist on responsible fees for responsible work. One designer could not do this alone. To be effective, this approach would have to be taken by all designers.
Fire Protection Engineering welcomes letters to the editor. Please send correspondence to firstname.lastname@example.org or by mail to Fire Protection Engineering, 7315 Wisconsin Ave., #620E, Bethesda, MD 20814.