Open Collections

Lahrsen, Steffen Thomas

University of British Columbia

The increasing complexity of modern forest operations demands advanced methodologies for productivity assessment. While onboard computers (OBC) and sensor technologies are well-integrated into cut-to-length (CTL) harvesting systems, their adaption in whole-tree (WT) harvesting is limited. The lack of standardized automated data collection in WT operations makes productivity analysis more resource-intensive and reliant on manual data gathering. This research seeks to bridge this gap by leveraging long-term large-scale production data recorded by FPDat II OBC to develop productivity models for felling machines in WT harvesting. A multi-stage methodology was employed, beginning with a systematic synthesis of key productivity-influencing factors in feller buncher and feller director operations. Subsequently, a validation study assessed the accuracy of FPDat II OBCs in estimating machine time metrics. Building on these findings, a novel approach integrating OBC-generated GNSS data with high-resolution LiDAR-based forest inventory data was developed to automate production analysis. The final phase involved long-term productivity modeling using a heteroscedastic mixed-effects framework to identify the influence of the variables stem size, stems per hectare, volume per hectare and ground slope on productivity. Results indicate that OBC data loggers provide reliable machine time estimates, with errors in productive time remaining below 1% when appropriate preprocessing thresholds are applied. Integrating GNSS-derived machine tracks with forest inventory data enabled accurate estimations of harvested volume at a machine-level resolution, reducing dependency on manual field measurements. The developed productivity models offer actionable insights on WT felling at the machine- and cutblock-level. This research contributes to the advancement of digital forest machine connectivity by providing a scalable framework for automated productivity monitoring in WT harvesting. The findings support industry stakeholders in optimizing machine deployment, improving operational planning, and enhancing supply chain efficiency. Future research should explore AI-driven predictive modeling and real-time machine connectivity to further refine productivity assessment methodologies in industrial forestry.

Text

2025-04-23

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/90725

Forestry

University of British Columbia

UBCV

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