-
Notifications
You must be signed in to change notification settings - Fork 0
Home
- Design Values
- Random Distributions
- MCS0 Introduction
- MCS0 Problem Definition
- Useful Unicode Characters
The structural fire resistance of buildings is traditionally determined through guidance documents or prescriptive requirements. An alternative approach is Probabilistic Risk Assessment (PRA), which assesses fire resistance in terms of time-equivalence. PRA studies often require building-specific stochastic parameters and involve multiple calculation iterations to determine an output variable distribution in an unbiased manner. Despite its potential, adopting PRA in design is challenging due to limitations in current engineering tools. This paper introduces a Python library of probabilistic functions for estimating fire severity distributions within an enclosure for specific scenarios. The library's application is demonstrated through an 18 m tall office building design case, where the structural fire resistance is computed based on the conditional reliability targets underpinning BS 9999.
The required fire resistance for a structure is commonly specified using contemporary guidance documents, such as BS 9999 in the UK. These documents provide a straightforward approach to determining minimum structural fire resistance. In BS 9999, the ventilation-dependent fire resistance period tables are derived from the time equivalence method, applied in a probabilistic framework based on the work of Kirby et al. However, the limitations of the studies underlying BS 9999 guidance are often overlooked. For example, the recommendations for offices are compatible only with compartment sizes up to 1,000 m². Furthermore, the Eurocode parametric fire models are applicable to fire compartment areas of up to 500 square metres. These limitations necessitate building-specific analyses to fully utilize the rationalization opportunities provided by BS 9999 for buildings outside these constraints.
PRA, as documented in PD 7974-7:2019, offers an alternative framework to assess appropriate fire resistance for a range of building-specific fire scenarios. This involves using stochastic variables for key fire development inputs. Such methods are well-documented in literature and have been employed in real-world engineering projects. PRA, being building-specific and applying fundamental methods, addresses the limitations of prescriptive methods.
Monte Carlo Simulation (MCS) is a technique used in PRA. The general MCS procedure, shown in Figure 1, involves using stochastic parameters defined by historical data probability distributions. Several sets of parameters are randomly generated using these distributions. Then, a time-equivalence calculation is performed for each parameter set.
Despite its advantages, adopting PRA in engineering projects is challenging due to the complexity of the required calculations and time constraints in real-world design scenarios. Building-specific PRA, involving intricate calculations, can make result reproduction difficult for reviewers.