ABSTRACT
Agent-based models, which are able to represent directly the behaviour and locations of individuals, are ideally placed to address these kinds of questions. Early examples were very abstract (Epstein and Axtell 1996; Wilensky 1998), but the range of different specific diseases that have now been considered using multi-agent models has increased over recent years, although a major focus has been influenza (Bian 2004; Ferguson et al. 2005; Ferguson et al. 2006; Germann et al. 2006; Stroud et al. 2007; Amouroux et al. 2008; Cooley et al. 2008; Halloran et al. 2008; Lee et al. 2008; Yang and Atkinson 2008; Epstein 2009; Brouwers et al. 2009; Rao et al. 2009; Ajelli et al. 2010; Bian et al. 2012; Duan et al. 2013), particularly with regard to the possibilities of a pandemic outbreak. Similar levels of attention have been given to smallpox (Chen et al. 2004; Eidelson and Lustick 2004; Epstein et al. 2004; Eubank et al. 2004; Burke et al. 2006; Brouwers et al. 2010), where an accidental or terrorist engineered epidemic might have large-scale consequences (Halloran et al. 2002; Longini et al. 2007).
