Keywords: premeditated destructive events, terrorist attack, bombings, fuzzy logic, fault tree analysis, concrete slab.

Premeditated Destructive Events (PDEs) have reshaped our thinking on building design, construction and maintenance. Little is known about building performance when subjected to PDEs. However, the consequence of not addressing PDEs could be catastrophic both to human life and property. Building structural performance is directly associated with the safety of human lives. Recent PDEs have shown that state-of-the-art buildings do not offer sufficient protection for their occupants from catastrophic failure.

PDEs have become of concern to all nations. According to the Emergency Response And Research Institute (ERRI) Annual Report of International Terrorist Activity 2001, there were total numbers of 1106 cases of terrorist attacks, killing 2494 civilians and wounding 3648; and 864 cases of terrorist attack in Year 2001, with 4739 people killed and 3386 injured. These statistics exclude the number of injured people in the September 11 terrorist attacks.

This paper assesses the performance of constructed facilities subject to PDEs without destructive testing. Fault tree analysis and fuzzy logic are employed to enable engineers to learn from similar events by studying failure mechanisms of constructed facilities. The performance of structures depends on many relevant events that make it difficult for assessors to control and judge quantitatively. Instead, imprecise linguistic terms are often used. For instance, the performance of the building's walls on a floor may be assessed as "fairly good."

The authors use typical design and supporting structural components of a ten-story reinforced concrete frame structure to develop a fault tree analysis model. They present a method that simulates the occurrence of PDEs thereby facilitating the work of engineers and assessors in evaluating the building performance throughout the design and construction process. This method employs experts' subjective judgment of the structure performance that could be used in different stages of the construction process. The fault tree model for the structure is used to show possible relevant events that contribute to a predetermined top event. In this paper, the predetermined top event is reinforced concrete slab failure.

Fault tree analysis as a diagnostic and prognostic tool can be used to evaluate a structural component through deductive analysis by asking, "How a component can fail or become unserviceable?" Then contributing events are categorized into enabling events, triggering events, and loss of support events. In this study, enabling events (events due to internal problems) are designated as inadequate performance of designs, material problems, and poor workmanship. Triggering events (events due to external problems) are considered as results of PDEs such as explosion and fire. Loss of support events are due to the failure of the supporting component, for instance, the slab of a building collapses because the supporting columns have crumpled. These events are also known as basic events. To further analyse the stability of the building, the authors developed a translational model computer program based on fuzzy rules, fuzzy set concepts and fuzzy algebra. The translational fuzzy set model is represented graphically.

This paper presents a method that is sufficiently rigorous so that it can be applied to evaluating PDEs on concrete slabs of a building. It permits experienced based expert opinion to be used to supplement the scant statistical data available on the effect of PDEs on concrete slabs. It also allows engineers to evaluate the relationship among triggering, enabling and loss of support events as they relate to PDEs. Fault tree analysis and fuzzy logic are useful tools to determine the building performance before, during and after construction. By careful and stringent assessment of the detail, corrective action could be performed to enhance the protection of the building's occupants from PDEs.

In follow-on work, the fuzzy logic evaluation of the linguistic terms used in assessing building performance will be applied to fault tree analysis. A computer program will be developed that is expected to facilitate the work of the assessor, particularly the site engineer, supervisor, owner, etc., in assessing the performance of a particular section of a structure or the entire structure with respect to the potential occurrence of a PDE.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £82 +P&P)