Multi-objective multi-path COVID-19 medical waste collection problem with type-2 fuzzy logic based risk using partial opposition-based weighted genetic algorithm

The transportation and handling of the most infectious coronavirus disease 2019 (COVID-19) waste are vital from the point of view of risk and safety. Here, a multi-objective, multi-path medical waste collection routing problem with occupational and transportation risks is considered. A specially equipped medical collection van with appropriately dressed collectors starts from a dumping campus and returns to it for disposal after collecting the COVID-19 hazardous wastes from rural medical centers/hospitals. Transportation, fixed charge, and loading costs are considered in this system. The occupational risk in each medical center depends on the number of workers involved, the collected amount of hazardous waste, and the storage period. The transportation risk depends on the probability of an accident which is determined by Type-1 and Type-2 fuzzy logic, the environmental condition of the accident, the length of the route, the weight of carried waste, and the population around the accident site which are the key applications of artificial intelligence. Here, the objective is to minimize the total cost, occupational, and transportation risks. This multi-objective problem is converted to a single-objective one through the weighted sum method. To solve it, we develop and implement a genetic algorithm with partial opposition-based learning-dependent initialization and mutation, probabilistic selection, and weightage-based comparison crossover. The proposed algorithm is tested against some standard benchmark instances, its effectiveness is shown through statistical tests and performance metrics. From the numerical experiments, it is observed that total cost is inversely related to both risks, but their behavior is non-linear. Some managerial insights are presented.