Log cut optimization

Abstract

When building log houses, you must first cut full-length logs from stock to smaller logs in the factory. These logs are packed into their own pallets in the order of installation and the pallets are delivered to the construction site. While cutting the full-length logs to smaller logs, there will be some trim loss, also known as waste. The purpose of this thesis is to develop an algorithm to minimize the trim loss. On top of the trim loss minimization the algorithm must consider the space limitations in the production lines in the factory. There is only room for a certain number of log pallets to simultaneously exist next to the production lines, and the logs must be cut so that this limit is never exceeded. The algorithm developed in this thesis used dynamic programming with a heuristic search. The algorithm is not guaranteed to provide an optimal solution, however, it allowed us to reduce the trim loss from dozens of percentages to near one percentage. At the same time, the number of allowed open log pallets was never exceeded, and each log pallet contained only logs that were installed after each other.