Outgoing selectee rates at hub airports Julie Virta a , Sheldon H. Jacobson a , John E. Kobza b, * a Department of Mechanical and Industrial Engineering, University of Illinois at Urbana-Champaign, 1206 West Green Street MC-244), Urbana, IL 61801-2906, USA b Department of Industrial Engineering, Texas Tech University, MS 3061, Lubbock, TX 79409-3061, USA Received 19 June 2001; accepted 1 December 2001 Abstract The selectee rate at an airport, or the fraction of originating passengers at an airport that have not been cleared by the computer-aided passenger prescreening system, may be useful in determining where to deploy certain baggage screening security devices. At hub airports, this rate alone does not capture the selectee rate of outgoing passengers, which consists of both originating passengers and transferring passengers. The impact of transferring passengers on the outgoing selectee rate is needed to optimally deploy baggage screening security devices at airports. This paper introduces a methodology for computing the outgoing selectee rates for a set of hub and non-hub) airports and to quantify the impact of transferring selectees at airports. Different types of selectee rates are de®ned and relationships between selectee rates for a set of airports and within an airport are discussed. Based on these relationships, two approaches for computing the outgoing selectee rates are presented. The resulting formulations, the exact formulation and the approximate formulation, are illustrated and compared using several examples, including a real-world example with data from the of®cial airline guide. The methodology introduced in this paper can be used to determine which individual airports pose the greatest threat from selectee passengers to the entire system of airports, hence can be used to aid in the deployment of baggage screening security devices at airports. q 2002 Elsevier Science Ltd. All rights reserved. Keywords: Security; Baggage screening; Aviation security; Selectee baggage 1. Introduction and motivation Airport security is an ongoing issue of national impor- tance and concern. With terrorist threats to air transportation and other air disasters on the rise, aviation security has been thrust into the public spotlight [1]. In an attempt to tighten airport security, the Federal Aviation Administration FAA) has encouraged the development of new baggage screening security device technologies. The FAA has been commis- sioned to purchase and deploy these baggage screening security devices in airports throughout the US, as well as deploy these devices such that they have the greatest positive impact on security [2]. Their challenge is to make optimal use of these technologies. One problem faced by the FAA is determining in which airports to deploy the various types of baggage screening security devices [3]. This has prompted the development of procedures that require only a certain fraction of all passen- ger baggage to be screened. These passengers, termed selectees, are individuals who have not been cleared by a system developed by the FAA, called the computer-aided passenger prescreening system CAPPS) [4]. The selectee rate is the fraction of originating passengers on a given ¯ight that are selectees i.e. number of selectees on a given ¯ight divided by the total number of passengers on the ¯ight). The selectee rate on a given ¯ight can be as small as zero, and as large as one, based on various characteristics of the ¯ight origination and destination cities, as well as characteristics of the passengers on the ¯ight. Airports can be either hub airports, de®ned as any airport that contains transferring passengers and originating passen- gers, or non-hub airports, de®ned as any airport that contains originating passengers only. The selectee rate at an airport is de®ned only for passengers originating at that airport. At hub airports, this rate does not capture the selectee rate of outgoing passengers, which consists of both originating passengers and transferring passengers. Therefore, it is necessary to de®ne an alternative selectee rate for each hub airport, termed the outgoing selectee rate, to account for the effect of transferring passengers. Assume that the screening of selectee bags only occurs at origination airports. Therefore, transferring selectee bags that are not screened at origination are also not screened at their transferring hub airport. Furthermore, the impact of transferring selectee bags on the outgoing selectee rate Reliability Engineering and System Safety 76 2002) 155±165 0951-8320/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved. PII: S0951-832002)00004-2 www.elsevier.com/locate/ress * Corresponding author. Fax: 11-806-742-3411. E-mail address: john.kobza@coe.ttu.edu J.E. Kobza).