A Scheduling Algorithm Tailored for a Dredging Operation

This thesis presents the development of a scheduling algorithm specifically tailored for a dredging operation. Addressing the unique challenges posed by underwater excavation in consolidated silica sand deposits using cutter suction dredgers (CSDs). Conventional open-pit scheduling approaches fail to capture the continuous, constrained, and direction-dependent nature of dredging systems. To bridge this gap, an algorithm in Python was developed that integrates 3D geological and quality models into a Directed Acyclic Graph (DAG) framework and employs topological sorting to generate operationally valid mining sequences. This ensures that all spatial, geotechnical, and operational dependencies are respected throughout the scheduling process. The approach transforms detailed block models into Smallest Mining Units (SMUs) aligned with actual dredger production volumes and applies stochastic, constraint-based sequencing to generate feasible mining scenarios. Validation using real operational data demonstrated that the algorithm follows the constraints while stabilizing product quality that goes into the mixing pond, reducing fluctuations by approximately 2–5% compared to historical data. The results suggest that the proposed method can generate feasible and consistent schedules, serving as a basis for future data-driven approaches to mine planning.