A Multiobjective Coverage-Based Model for Civilian Search and Rescue Yaw Asiedu, Mark Rempel Centre for Operational Research and Analysis, Defence Research and Development Canada, Ottawa, Canada Received 8 April 2008; revised 2 September 2009; accepted 7 September 2009 DOI 10.1002/nav.20387 Published online 28 October 2009 in Wiley Online Library (wileyonlinelibrary.com). Abstract: The Civil Air Search and Rescue Association (CASARA) is a Canada-wide volunteer aviation association that pro- vides air search support services to the Canadian National Search and Rescue (SAR) program. As with any emergency service provider, the locations of CASARA units greatly impact their overall effectiveness. In this article, the optimal location of CASARA units is formulated as a multiobjective maximal covering location problem. The model addresses the objectives of maximizing the coverage, minimizing the number of units, and maximizing the backup coverage of SAR incidents within Canada. A multigender genetic algorithm is proposed to determine a set of nondominated CASARA location configurations. Results are compared with solutions found using commercial integer programming software. It is shown that the nondominated genetic algorithm solutions are near-optimal. These are determined in much less time than comparable solutions using commercial integer programming software. © 2009 Wiley Periodicals, Inc. ∗ Naval Research Logistics 58: 167–179, 2011 Keywords: search and rescue; covering model; multiobjective genetic algorithm 1. INTRODUCTION Responses to Search and Rescue (SAR) incidents within Canada are provided through a combination of assets from the Canadian Forces (CF), Canadian Coast Guard, Royal Cana- dian Mounted Police, Meteorological Service of Canada, Parks Canada, Transport Canada as well as several local authorities and volunteer groups [21]. In the Canadian con- text, the Civil Air Search and Rescue Association (CASARA) is the primary volunteer SAR organization. CASARA com- prises facilities positioned across Canada, with each facility having a group of volunteers and at least one fixed-wing air- craft. Typically, this is a single engine aircraft (e.g., Cessna). The role of CASARA within the Canadian SAR system has expanded since its inception. Today, CASARA is capable of providing fixed-wing aviation searches for locating Emer- gency Locator Transmitters (ELTs) and locating Personal Locator Beacons (PLBs), performing ground based searches, acting as spotters for CF searches, and setting up and running deployed search headquarters. Although CASARA is able to provide this wide range of capabilities, this article focuses on its ability to perform fixed-wing aviation searches. When a SAR incident occurs in Canada, one of three Joint Rescue Coordination Centres is notified, for example, through a satellite indicating an ELT has been activated. It is Correspondence to: M. Rempel (mark.rempel@drdc-rddc.ge.ca) the responsibility of the coordination centre to task assets, if necessary, to perform a fixed-wing search to visually deter- mine the location of the incident. CASARA is often the first responder because it is able to provide a rapid response before CF primary SAR fixed-wing aircraft can be on-site. This is due to the fact that CF primary SAR fixed-wing aircraft are located at only four sites, whereas CASARA consists of over 100 sites across Canada. As such, the location of CASARA bases has a significant impact on the overall performance of the national SAR program. The issue of locating emergency services has been dis- cussed extensively in the scientific literature. Naturally, the focus has been on municipal emergency services such as, fire, ambulance, and police services. Regardless of the service area, the primary objective remains the same; to provide ser- vices to the most people or the widest area at a reasonable cost. Covering models are the class of location models that seek to cover demands (i.e., provide services to demand points) within a predetermined response-time standard. These mod- els can be classified into two main groups: location set cov- ering problem (LSCP) and the maximal covering location problem (MCLP). The LSCP was introduced in the early 1970s by Toregas et al. [23]. The goal of the LSCP is to locate the minimum number of facilities such that all demand points are within a maximum distance of a facility. This may not always be fea- sible in some cases due to factors such as inadequate funds © 2009 Wiley Periodicals, Inc. ∗ © 2009 Government of Canada. Exclusive worldwide publication rights in the article have been transferred to Wiley Periodicals, Inc.