Comparing manual and automated time and tree volume data for estimating harvester productivity in a Pinus taeda stand

Abstract

Abstract

Cut-to-length harvesters are widely used in South Africa as they can be efficient, enhance operator safety, deliver consistent quality, and minimize environmental impacts. These machines are increasingly becoming automated, incorporating digital systems that produce real-time operational data to support informed decision-making. This research compared manually and machine-collected time and log measurement data to assess agreement between the two methods. A Ponsse Ergo harvester equipped with an H7 head was observed while working in a loblolly pine (P. taeda) stand. The cut-to-length harvester productivity was assessed by comparing manual time study and log measurement data with the stem file and timestamp data recorded by the onboard computer system. Statistical analyses, including regression modeling, correlation analysis, and non-parametric tests, were used to evaluate time distribution and the level of agreement between manual and machine measurements. A high degree of agreement was found, indicated by small median differences and strong correspondence between manual and machine datasets. Within machine data, variation increased with log size, with less consistency observed in sawlogs. Although significant differences between the two datasets were observed in log volume and log length, no significant difference was found for the thin-end (small end) diameter. Machine data proved to be reliable for measuring productivity.