Adaptive control of bucking on harvesters to meet order book constraints

Modern mechanized harvesters are often fitted with sensors that measure stem dimensions and with computers that optimally buck each stem to maximize the value gained based on stem dimensions, qualities, log prices, and desired specifications. Optimally bucking individual stems, based on market prices, is unlikely to provide yields that meet order book constraints at the harvest unit or forest level. An adaptive control heuristic was developed by imbedding an individual stem optimal bucking dynamic programming procedure in a threshold accepting algorithm which adjusts relative prices and minimum small-end diameter specifications to meet order book constraints. The adaptive control heuristic was tested on four stands where the location and detailed stem description of every tree was known. Three of the stands were virtual stands designed specifically to test the adaptive control heuristic. The fourth stand was a real-world stand. This paper describes the adaptive control heuristic, presents results from a series of tests on its performance, and compares the use of stem information collected during pre-harvest inventory and by the harvester as it works its way through a harvest unit to calculate the relative prices that will best satisfy the market constraints. Improvements in meeting order book, target proportions were found for all four stands when the heuristic was used: 17 to 22 percent improvement with pre-harvest inventory data and 19 to 26 percent improvement with stem information gathered as the harvester works its way through the unit.

You must be logged in to download any documents. Please login (login accounts are free) or learn how to Become a Member