A quick decision-making process in response and management of epidemics has been the most common approach, as accurate and relevant decisions have been demonstrated to have beneficial impacts on life preservation as well as on global and local economies. However, any disaster or epidemic is rarely represented by a set of single and linear parameters, as they often exhibit highly complex and chaotic behaviours, where interconnected unknowns rapidly evolve. As a consequence, any such decision-making approach must be computationally robust and able to process large amounts of data, whilst evaluating the potential outcomes based on specific decisions in real time.
TopIntroduction
The occurrence of epidemics has a long-lasting impact, usually affecting a large number of individuals (UNDRR, https://www.unisdr.org/). Epidemics are often regarded as a type of disaster, similar to ‘man-made’ disasters but with increased impact (O'Brien, et al, 2010). In this chapter, we argue that the increasing digitalisation of the management of disasters is directly linked with epidemics, with existing and novel AI and deep learning applications having a direct impact on assessing and monitoring epidemic occurrences.
A quick decision-making process in response and management of epidemics has been the most common approach, as accurate and relevant decisions have been demonstrated to have beneficial impacts on life preservation as well as on global and local economies (Thompson, et al. 2006).
However, any disaster or epidemic is rarely represented by a set of single and linear parameters, as they often exhibit highly complex and chaotic behaviours, where interconnected unknowns rapidly evolve. As a consequence, any such decision-making approach must be computationally robust and able to process large amounts of data, whilst evaluating the potential outcomes based on specific decisions in real time. There are a variety of software products to aid practitioners during the decision-making process, including Emergency Information System (EIS), SoftRisk, EM 2000, and E-Team, which provide a range of emergency management decision support, resource management, and incident documentation functions to emergency managers (Green, et al, 2001). The emergency management and operations research literature also contains a number of examples of Decision Support Systems (DSS) designed for specific scenarios or objectives.
Table 1 depicts relevant support tools, which are widely used in disaster management scenarios. These are also applied to epidemic scenarios especially in the management of resources and infrastructures.
Table 1. Decision support tools widely used in disaster management and applicable to epidemic scenarios
(Tompson, et al, 2006).
Any epidemic and disaster management approach includes four distinct components: ‘mitigation’; ‘preparedness’; ‘response’; and ‘recovery’. The aim of the mitigation component is to identify and minimise the effects of a hazardous event, while the preparedness component aims to identify the best response. Response activities follow the above phases, and they focus on plans as well as suitable emergency activities to mitigate life and economy losses. Finally, recovery identifies the best procedures to allow a return to pre-disaster levels (Yu, et al., 2018).