An integrated mathematical programming approach for drone-based drug delivery under uncertainy

This thesis examines the potential use of unmanned aerial vehicles in healthcare logistics and proposes a mathematical programming model that integrates warehouse selection, drone type determination, and distribution planning in the pharmacy pharmaceutical distribution process. Delays and operational costs experienced in traditional road distribution have increased the search for alternative methods for time-sensitive deliveries in healthcare logistics. Drone-based distribution systems stand out as a fast and flexible solution for short-distance and urgent deliveries. The study considered 103 pharmacies in a specific district, candidate warehouse locations, and five drone types with varying technical specifications. Furthermore, five weather scenarios were defined to examine the impact of meteorological uncertainty on drone performance. Changes in range and energy consumption were integrated into the model with scenario-based performance multipliers. The model develops a stochastic structure by integrating operational factors such as drone battery life, range constraints, drone capacity, and uncertain pharmacy demands. The study utilizes a Goal Programming approach to simultaneously address objectives such as cost, time, and energy consumption. Deviation variables were defined for each objective, and a multi-objective structure was created that minimizes the weighted sum of these deviations. The model developed in this context offers a applicable and holistic distribution planning approach for healthcare logistics, taking into account the impact of different weather conditions.