Sang Hyun Kim

This study addresses a drone delivery routing optimization problem with battery constraints, a variant of a vehicle routing problem (VRP), where drones deliver goods to multiple customer nodes while adhering to operational limitations such as battery and payload capacities. Our study proposes Adaptive-Policy Optimization with Multiple Optima (A-POMO), an enhanced framework based on POMO, which is a reinforcement-learning-based method for solving combinatorial optimization problems like VRP without relying on labeled data. A-POMO for battery-constrained drone delivery routing improves solution quality and computational efficiency through a constraint-guided probability distribution and an adaptive elite reward mechanism. Especially, A-POMO adjusts reward weights adaptively and employs a refined baseline for elite solutions to strike a balance between exploration and exploitation. Numerical experiments with 20, 50, and 100 customer nodes demonstrate that A-POMO achieves near-optimal solutions compared to MILP, with significant time efficiency for small-scale instances. It also outperforms the original POMO in large-scale instances for constraint-guided route optimization.

By generalizing the unconstrained p-hub approach for the park-and-ride (P&R) facility location problem under the multinomial logit demand function, the capacitated p-hub approach for the problem under the nested logit demand function captures a broader range of real-world cases. To solve this problem optimally, we introduce a mixed-integer linear program and accelerate its solution by enhancing the branch-and-cut procedure. To address the problem at a large scale, we introduce two other polyhedral approaches: variable neighborhood search (VNS) and adaptive randomized rounding (ARR). Downtown areas in Seoul have a high modal share of public transportation and congested road traffic, yet P&R has not been widely implemented. Therefore, we apply the ARR procedure to solve a real-world problem using traffic and geographic data from the Seoul metropolitan area. ARR performs better than VNS and addresses real-world cases. The solutions obtained by ARR present a phased expansion plan that encourages policymakers to start installing a small number of P&Rs immediately.

Airport runway operations are managed hierarchically. Initially, a runway configuration is determined several hours before the commencement of actual runway operations. Thereafter, each aircraft is...