Softwood sawlog secondary transport travel speed prediction for the South African forestry industry

Abstract

The objective of this study was to develop a predictive model for travel speed of softwood sawlog timber transport (STT) vehicles over a range of forest and provincial roads of varying condition for the South African forestry industry. Data was accumulated from both the Eastern Cape/KwaZulu-Natal and Mpumalanga forest regions of South Africa. Vehicle location and payload data were collected remotely using a combination of GPS tracking and remotely sensed data. Road condition, including road width, was assessed for each identified road segment in-field according to a visual condition indicator (VCI) index. Two STT contractors were selected from each forest region transporting pine sawlogs directly from compartment roadside landings to either a processing plant or log storage area. Five STT vehicle types, representative of transport operations in the two forest regions, were assessed. Principal component analysis was conducted to determine the degree of communality between the respective predictor variables combining road width and VCI as one factor, and truck maximum power and the percentage of (legal) maximum payload as a second factor. Comparisons of the correlations between average speed and the respective predictor variables showed that road width and percentage of maximum (legal) payload had the highest correlations. Multiple linear regression of these two factor variables were used in the model showing both variables as significant (p < 0.05) with an adjusted r2 value of 0.52.