A Decomposition-Based Framework for Large-Scale Multi-Period Log-Truck Routing and Scheduling: A Case Study in Canadian Forestry

This paper addresses the complex multi-period log-truck routing and scheduling problem (LTRSP) in the forest industry, proposing an enhanced mathematical programming formulation and a decomposition heuristic to solve large-scale instances. The Canadian forest industry faces significant logistical challenges due to vast distances, seasonal variability, market fluctuations, and environmental concerns. Efficient transportation is essential for maintaining both economic viability and environmental sustainability. This research presents a comprehensive framework for routing and scheduling, starting with an analysis of industry rules to design a routing network. A mixed-integer linear programming (MILP) model is then formulated to capture these rules, integrating spatial and temporal constraints. Subsequently, a solving approach, Relax-and-Fix, is applied to historical data provided by a Canadian forest company. The results demonstrate that the framework can generate optimal solutions for daily problems and near-optimal solutions for weekly problems within reasonable computation times. This work ofers an end-to-end framework for tackling LTRSP, developed in collaboration with forest companies and incorporating all their critical business constraints, distinguishing it from existing approaches in the literature.