Open Collections

Arora, Rohit

University of British Columbia

Forest harvesting consists of multiple sequential activities to convert trees into logs ready for delivery to mills: felling, processing, yarding and loading. The forest harvesting cost contributes significantly to the delivered cost of logs. Therefore, it is important to optimize the schedule of forest harvesting activities to minimize costs. Previous studies have developed mathematical programming models to optimize the forest harvesting scheduling at the operational level. However, these studies did not consider the precedence relationship between forest harvesting activities, multiple machine assignment decisions, and the use of multi-task machines. The goal of this dissertation is to optimize the scheduling of harvesting activities at the operational level considering the mentioned research gaps. To achieve this goal, three mathematical programming models are developed in this work. In the first model, the precedence relationship between harvesting activities and the movement of individual machines is considered. In the second model, the multiple machine assignment decisions are incorporated in addition to those conditions considered in the first model. Also, the precedence relationship based on the slope of cut blocks is included in the second model. In the last model, the use of multi-task machines is incorporated in addition to other considerations in the second model. In this model, the scheduling of activities related to road construction within a cut block is also included. All models determine the start time and end time of each harvesting activity at each cut block, and where the machine should move after completing its operation in one cut block. In addition, the second and third models also determine the number of machines to be assigned at each cut block for each activity. All the models are applied to the harvesting operations of a real case study in the coast of British Columbia. The results indicate that the harvesting cost from the models is (at most) 4.5% higher than that of the defined ideal cost benchmark. All the developed models can be easily applied to other cases and regions by modifying the sets of succeeding and preceding activities in the input data according to the requirement of the harvesting system.

Text

2023-08-17

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/85509

Forestry

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/