An IoMT platform for an integrated sustainable energy-efficient disaster relief supply chain to prevent severity-driven disruptions during pandemics

Abstract

Pandemics, such as the Influenza virus and the contemporary COVID-19, can lead to widespread disruptions in key segments, including the supply chain, beyond the capacity of communities or governments. The establishment of a robust relief supply chain network can alleviate the destructive effects of pandemics and strengthen the distribution of relief supplies across medical centers and demand zones. Furthermore, the Internet of Medical Things (IoMT), which connects medical devices and applications over the internet, provides healthcare providers with real-time data collection, transmission, and sharing. Considering this motivation, in this work, a multi-objective sustainable network is firstly modeled mathematically to not only curb the direct flow of relief supplies among specific components, but also to operate the reverse flow of waste within the network. Furthermore, in accordance with the Sustainable Development Goals (SDGs), the model emphasizes the energy consumption of critical activities like production and transportation. Additionally, an IoMT configuration is propounded to strengthen the mathematical model with real-time data. metaheuristic optimizers are effective toolkits owing to the NP-hardness of the model. To ensure that the model is compatible and applicable under varying conditions, a suite of tuned metaheuristic optimizers is utilized as well as five scales of test problems. Additionally, the performance of optimizers is examined using a number of recognized performance indicators. The normality of the results is evaluated through statistical tests, namely the Kolmogorov-Smirnov and Shapiro-Wilk tests. Following this assessment, a comprehensive analysis is carried out using the Wilcoxon test and Paired-Samples T Test to compare the results in a pairwise manner, while maintaining a significance level of 0.05. The outcomes derived from these tests reveal the presence of significant disparities among the performance indicators. To ascertain the algorithm with superior performance, an evaluation is conducted using interval plots and the Friedman test, considering each individual performance indicator. The empirical evidence derived from this analysis indicates that the Multi-objective Seagull Optimization Algorithm (MOSOA) exhibited the highest overall mean rank score of 1.93, surpassing other metaheuristic algorithms in terms of performance.