A graph cellular automaton with relation-based neighbourhood describing the impact of peer influence on the consumption of marijuana among college-aged youths (Q825258)

The authors employ the relation-based graph-cellular automata (r-GCA) approach to develop a model for the drug use on a college campus. The work builds on authors' earlier research reported in [\textit{Y. B. Ruhomally} and \textit{M. Z. Dauhoo}, Comput. Appl. Math. 39, No. 4, Paper No. 327, 29 p. (2020; Zbl 1462.37012)]. The drug consumption is viewed as a contagious epidemics where the spread of the drug use is influenced primarily by the peer influence. A relational graph-cellular automation is set up by local transition rules that describe social interactions between the four states: nonusers (N), experimental users (E), recreational users (R) and addicts (A). Simulations results for an r-GCA model are discussed; they agree reasonably well with the data from the Ontario Student and Drug Use Health Survey (OS-DUHS). The effect of targeted campaigns on the prevention of the drug use has been modeled with three different scenarios indicating a significant consecutive decline in the number of drug users. However, \textit{V. Dabbahian} et al. [``The social impact in a high-risk community: a cellular automata model'', J. Comput. Sci. 2, No. 3, 238--246 (2011; \url{doi:10.1016/j.jocs.2011.05.008})] warned on significant difficulties with the validation and certification of cellular automation due to restricted access to the data on drug-related crimes suggesting that one has to be cautious with the conclusions based on such models.