טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentGoldberg Ohad
SubjectStaffing Counter-Terrorism Agency's Teams using Data
Evelopment Analysis (DEA)
DepartmentDepartment of Industrial Engineering and Management
Supervisor Professor Boaz Golany
Full Thesis textFull thesis text - English Version


Abstract

Counter-terrorism agencies (CTAs) such as Israel's General Security Service, Shabak, deal with monitoring, controlling and interdicting terror. CTAs operate in a complex and inherently uncertain operational environment. In order to obtain their objectives, CTAs require many resources. Some of those are agents that gather intelligence and analyze it. CTAs' performance depends, among other things, on the number and types of employed agents and their allocation to geographical areas.

Even though staffing issues have a significant effect on CTAs' performance, we had little success in finding literature references regarding them. Most of the published material focuses on developing theoretical models that are based on simplifying assumptions. In this study we address CTAs' staffing issues by developing methods that will overcome the disadvantages of current ones. In order to do so, we use the Data Envelopment Analysis (DEA) methodology. 

DEA is a data-oriented methodology for performance evaluation of a set of peer entities that convert multiple inputs into multiple outputs. In the last few decades, a variety of DEA applications were used in different environments such as universities, hospitals and banks. DEA applications provide useful insights that can be applicable in real-life cases.

In this study, we develop several DEA-based methods for assisting with staffing issues in CTAs. Based on those we present several applications that can assist in: performance evaluation of agents' teams in a CTA; changes in the teams' composition effects analysis; and agents' teams' assembly.

This study also provides several expansions to the DEA models academic literature. The main ones are: a DEA model that incorporates dependency between inputs and a DEA model for dealing with uncertain data.