Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Industrial and Managements Systems Engineering

Committee Chair

Wafik Iskander

Committee Co-Chair

Bhaskaran Gopalakrishnan

Committee Member

Majid Jaridi

Committee Member

Powsiri Klinkhachorn

Committee Member

Feng Yang


Inefficient facility location decisions result in extra cost throughout the lifetime of the facilities. Hence, facility location decisions are considered as critical and difficult decisions in today's highly competitive business environment. There are two relatively new areas of research in facility location modeling approaches that have recently received much attention. The first is the incorporation of short-term inventory decisions into long-term location decisions. It has been shown in the literature (e.g. Daskin et al. 2002 or Shen et al. 2003) that failing to account for inventory may result in fundamental changes in the location decisions and thus suboptimality. The second considers the relaxation of the assumption in classical facility location models that facilities are always available and perfectly reliable. In reality, facilities are subject to disruptions due to different unexpected events such as natural disasters or terrorism. The interest in this avenue of research has been increasing in the past few years, triggered by recent high-profile disastrous events such as those of September 11, 2001 and hurricane Katrina in 2005. While the importance of the integration of the two aforementioned areas of research has been clarified in the literature, the integration studies are scarce due to the difficulty and challenges that are presented by the incorporation of the nonlinear inventory costs into a reliability location framework. This dissertation integrates the two areas of research and proposes a location-inventory model that also considers facility fortification as a proactive strategy to hedge against disruptions. Two global search methodologies are developed to solve the problem, and their computational efficiencies are investigated under different parameter settings.