How one builds, checks, validates and interprets a model depends on its 'purpose'. This is true even if the same model code is used for di erent purposes. This means that a model built for one purpose but then used for another needs to be re-justified for the new purpose and this will probably mean it also has to be rechecked, re-validated and maybe even rebuilt in a di erent way. Here we review some of the di erent purposes for a simulation model of complex social phenomena, focusing on seven in particular: prediction, explanation, description, theoretical exploration, illustration, analogy, and social interaction. The paper looks at some of the implications in terms of the ways in which the intended purpose might fail. This analysis motivates some of the ways in which these 'dangers' might be avoided or mitigated. It also looks at the ways that a confusion of modelling purposes can fatally weaken modelling projects, whilst giving a false sense of their quality. These distinctions clarify some previous debates as to the best modelling strategy (e.g. KISS and KIDS). The paper ends with a plea for modellers to be clear concerning which purpose they are justifying their model against.

For millennia tuberculosis (TB) has shown a successful strategy to survive, making it one of the world’s deadliest infectious diseases. This resilient behavior is based not only on remaining hidden in most of the infected population, but also by showing slow evolution in most sick people. The course of the disease within a population is highly related to its heterogeneity. Thus, classic epidemiological approaches with a top-down perspective have not succeeded in understanding its dynamics. In the past decade a few individual-based models were built, but most of them preserved a top-down view that makes it difficult to study a heterogeneous population. We propose an individual-based model developed with a bottom-up approach to studying the dynamics of pulmonary TB in a certain population, considered constant. Individuals may belong to the following classes: healthy, infected, sick, under treatment, and treated with a probability of relapse. Several variables and parameters account for their age, origin (native or immigrant), immunodeficiency, diabetes, and other risk factors (smoking and alcoholism). The time within each infection state is controlled, and sick individuals may show a cavitated disease or not that conditions infectiousness. It was implemented in NetLogo because it allows non-modelers to perform virtual experiments with a user-friendly interface. The simulation was conducted with data from Ciutat Vella, a district of Barcelona with an incidence of 67 TB cases per 100,000 inhabitants in 2013. Several virtual experiments were performed to relate the disease dynamics with the structure of the infected subpopulation (e.g., the distribution of infected times). Moreover, the short-term effect of health control policies on modifying that structure was studied. Results show that the characteristics of the population are crucial for the local epidemiology of TB. The developed user-friendly tool is ready to test control strategies of disease in any city in the short-term.

No abstract

No abstract