Efficient and safe maritime navigation in complex and congested coastal regions requires advanced route optimization methods that surpass the limitations of traditional shortest-path algorithms. This study applies Deep Q-Network (DQN) and Proximal Policy Optimization (PPO) reinforcement learning (RL) algorithms to generate and refine optimal ship routes in East Asian waters, focusing on passages from Shanghai to Busan and Ulsan to Daesan. Operating within a grid-based representation of the marine environment and considering constraints such as restricted areas and Traffic Separation Schemes (TSS), both DQN and PPO learn policies prioritizing safety and operational efficiency. Comparative analyses with actual vessel routes demonstrate that RL-based methods yield shorter and safer paths. Among these methods, PPO outperforms DQN, providing more stable and coherent routes. Post-processing with the Douglas-Peucker (DP) algorithm further simplifies the paths for practical navigational use. The findings underscore the potential of RL in enhancing navigational safety, reducing travel distance, and advancing autonomous ship navigation technologies.

This paper proposes a methodology for gantry route optimization in order to maximize the productivity of a odd-type surface mount device (SMD). A odd-type SMD is a machine that uses a gantry to mount electronic components on the placement point of a printed circuit board (PCB). The gantry needs a nozzle to move its electronic components. There is a suitability between the nozzle and the electronic component, and the mounting speed varies depending on the suitability. When it is difficult for the nozzle to adsorb electronic components, nozzle exchange is performed, and nozzle exchange takes a certain amount of time. The gantry route optimization problem is divided into the mounting order on PCB and the allocation of nozzles and electronic components to the gantry. Nozzle and electronic component allocation minimized the time incurred by nozzle exchange and nozzle-to-electronic component compatibility by using an mixed integer programming method. Sequence of mounting points on PCB minimizes travel time by using the branch-and-price method. Experimental data was made by randomly picking the location of the mounting point on a PCB of 800mm in width and 800mm in length. The number of mounting points is divided into 25, 50, 75, and 100, and experiments are conducted according to the number of types of electronic components, number of nozzle types, and suitability between nozzles and electronic components, respectively. Because the experimental data are random, the calculation time is not constant, but it is confirmed that the gantry route is found within a reasonable time.

A sweep-based heuristic using common area is developed for multi-vehicle VRPs under time various and unsymmetric forward and backward vehicle moving speed. One depot and 2 delivery vehicle are assumed in this research to make the problem solving strategy simple. A common area is held to make adjustment of possible unbalance of between two vehicle delivery completion times. The 4 time zone heuristic is used to solve for efficient delivery route for each vehicle. The current size of common area needs to be studied for better results, but the suggested problem solving procedures can be expanded for any number of vehicles.