Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mimarlık Fakültesi, Mimarlık Bölümü, Türkiye
Tezin Onay Tarihi: 2016
Öğrenci: NİLÜFER KIZILKAYA
Eş Danışman: ALİ MURAT TANYER, MEHMET KORAY PEKERİÇLİ
Özet:There is a direct relationship between building characteristics and fire spread. Acceptability of fire safety design involves interoperability of fire safety objectives and building design input. In case of unacceptable fire safety design process, modifications are made in proposed building design parameters for which architects are decision makers. Therefore, in order to get an acceptable level of fire safety, vulnerability of building design parameters must be identified during the design process by architects. The vulnerability level provides performance evaluation of the building in terms of hazardous actions and critical building design parameters to get prevention measures during the design process. Conventional design of building parameters regarding fire safety direct architects to regulations. However, for most of the architects, deterministic approaches of regulations are perceived as restrictions for the creative basis of architecture. Regulation-based vulnerability evaluation systems use deterministic and single parameter approach, in which the parameter either follows the rules, or not. On the other hand, the rapid increase in complexity and amount of information on building systems requires quick-response evaluations based on the decision-maker’s intuition, judgement, and experience. Moreover, increased building complexity reveals highly complex decision problems with multi-variables that are stochastic, unknown, and fuzzy. By providing complex interactions between variables, fuzzy logic enables qualitative descriptions of everyday reasoning. Therefore, in order to identify fire safety vulnerabilities of building variables based on expert opinion, this research uses fuzzy expert evaluation model. Fuzzy expert system helps the integration of uncountable, undefined, and uncertain information in the decision-making process. Previously studied fire safety fuzzy models do not cover all critical parameters of building characteristics, and focuses on comprehensive or active fire protection systems. In the proposed fuzzy fire safety vulnerability model, most critical building parameters regarding fire safety are determined through literature analysis. On the other hand, linguistic expert opinion is converted to membership functions through fuzzyTECH fuzzy logic toolbox, and rule-based interrelations of parameters are defined through human reasoning. Performing vulnerability evaluation with fuzzy expert model gives quick response and more accurate results based on human reasoning.