Optimisation of COVID-19 vaccination process using GIS, machine learning, and the multi-layered transportation model

Abstract

COVID-19 has affected the lives and well-being of billions of citizens worldwide. While nondrug interventions have been partially effective in containing the COVID-19 epidemic, vaccination has become the most important factor in maintaining public health and reducing deaths. In this study, a model is proposed to overcome the difficulties in organising vaccination due to heterogeneous population distribution in cities and to optimise the vaccination process considering the available resources. The results of the model are of strategic importance for the control of the COVID-19. Considering the transportation structures, population and vaccine resources in the regions, a different number of clusters is formed for each city. Each cluster consists of several districts that share health resources. A hybrid approach consisting of mathematical modelling and k-means algorithm is proposed, and it reduced the difference between vaccination times of three different vaccination clusters to about 3.5 days. The results also showed that the vaccination process can be reduced from 108 days to 44 days, which meant a 40% improvement in speed for administering vaccines. In this case study, we presented a vaccination programme in which the average antibody rate of individuals does not fall below the critical-time threshold.