Estimating harvester productivity in Pinus radiata plantations using StanForD stem files

Abstract

Productivity models produced using time differences between consecutive StanForD stem files collected by harvester onboard computers were compared with models produced using traditional time and motion techniques for the same initial trees. Three sites were studied in Pinus radiata plantation clearfell operations across southern Australia. Delays and trees with multiple leaders or broken tops were removed from the data. This was done for the stem file data using filters. The same filters were applied to data from all sites. No significant differences were found between the models at each site, though the stem file productivity models generally had a poorer fit than the time and motion models. The advantages of using stem files for modelling are the ready availability of stem file data, which enables rapid creation of generalised harvester productivity models and avoids short-term changes in productivity caused by the presence of an observer (the “Hawthorne effect”). Disadvantages are the inability to account for unforeseen changes in conditions during data collection, and the inability to isolate work-cycle time-element data.