Issue 20

P. Rezakhani, Frattura ed Integrità Strutturale, 20 (2012) 17-21; DOI: 10.3221/IGF-ESIS.20.02 18 Assessing risks is the step which prioritizes the risks for further analysis by quantifying their occurrence rates. Risk assessment method is an essential component for this step. The existing methods are classified into (1) simple classical methods, and (2) advanced mathematical models [3]. The existing risk assessment methods are either qualitative or quantitative which require different information and the level of detail [22]. The simple classical methods integrate deterministic risk modeling and analysis into CPM scheduling. The deterministic methods include sensitivity analysis [23], critical path method [24], fault tree analysis [25], event tree analysis [26], failure mode, and effects and criticality analysis, etc [27]. Other advanced approaches were proposed as follows; a Monte Carlo Simulation [23] for stochastic quantitative modeling and analysis; scenario analysis [28], and fuzzy set theory for qualitative judgment [28]. There are many factors which should be considered when a project risk manager selects a risk assessment method as follows; i.e., the cost of employing the technique, the level of external party`s approval, organizational structure, agreement, adoptability, complexity, completeness, level of risk, organizational size, organizational security philosophy, consistency, usability, feasibility, validity, and credibility and automation [29]. It is essential for the risk manager to have high quality data in order to effectively apply the quantitative methods, even if it is not easy to obtain such high quality data relative to risk items in the construction industry. The difficulty is attributed to address the uncertainties and subjectivities associated with construction activities [30]. Beside the lack of collectability, the uniqueness, and non-repetitive nature of construction projects impedes using probabilistic risk quantification approaches [31]. Responding risks is involved in developing options and/or actions to enhance opportunities to achieve the project objectives. Finally, monitoring and reviewing risks is to implement a risk response plan, to keep tracking of the risks identified, to monitor residual risks, to identify new risks, and to evaluates the effectiveness of the project risk management process [15]. For this step, each engineering expertise should use specialized risk management tool as shown in Table 1 for risk analysis depending on project phase. Discipline Planning/ Programming Preliminary Engineering Final Design Construction Planning Environmental Funding Approval Project Management Engineering Civil, Structural, Systems Cost Estimating Scheduling Budgeting Controls Real Estate/Right of Way Construction Management/Oversight Constructability/Contractor Other Technical (e.g. Legal, Permitting, Procurement) Risk Facilitation Table 1 : Key expertise for risk analysis by project phase (Adapted from [32]). Highly desirable; Desirable but optional depending upon circumstances. F UZZY R ISK A SSESSMENT fter Zadeh [33] introduced the concept of Fuzzy sets, and Fuzzy set theory, several researchers such as Kangari [34], Kangari and Riggs [35], Peak et al. [36], Tah and McCaffer [20], Wirba et al. [21], , Cho et al. [38], Choi et al. [39], Lyons and Skitmore [40], Baker and Zeng [41], Dikmen et al. [42], Zeng et al. [30], Wang and Elang [43], Karimiazar et al. [3], and Nieto et al. [15] -introduced fuzzy set theory(FST)-based risk modeling and analysis methods that deal with ill-defined, vague, imprecise, and complex risk analysis problems. For example, Kangari [34] proposes the application of fuzzy theory in risk analysis method using linguistic terms. The fuzzy theory-based risk analysis method was implemented as a part of construction project risk management system which consists of five steps (i.e., risk identification, policy definition, risk sharing and allocation, risk analysis, and risk minimization and response planning, etc). The fourth component, risk analysis, consists of three steps as follows; natural language computation, fuzzy set risk evaluation, and linguistic approximation. A