Intelligent decision support and agent-based techniques applied to wood manufacturing

A rough mill is a manufacturing facility where loads of lumber of approximate dimensions are cut into components of specific sizes, priorities and qualities. These components are used in making furniture, doors and windows. Lumber is a very valuable natural resource and is a significant expense to the rough mill. By improving the processes in the rough mill, the cost is reduced and the waste of natural material is decreased. In this research, the operations in a Canadian rough mill are described, and the decisions that operators take are identified. The rough mill scheduling of components on machines is a challenging problem that cannot be solved by traditional methods because the defects in the wood are not known in advance. In addition, the wood sizes are approximate and often inaccurate. Scheduling algorithms are implemented using constraint satisfaction and heuristic methods. An agent-based system for decision support and simulation is designed and implemented.