A decision-support system to assist forest industry planning and investment when data are scarce: the case of hardwood veneering in subtropical eastern Australia. Issue 2 (3rd April 2021)

• Record Type: Journal Article
• Authors: Venn, T. J.; McGavin, R. L.
• Is Part Of: Australian forestry. Volume 84:Issue 2(2021)
• Page Start: 59
• Page End: 72
• Publication Date: 2021-04-03
• Subjects: operations research -- non-linear programming -- processing scale -- facility location -- log procurement -- rotary veneer -- financial performance; Forests and forestry -- Australia -- Periodicals; Forests and forestry -- Australasia -- Periodicals; 634.90994
• DOI: 10.1080/00049158.2020.1836453
• Languages: English
• ISSNs: 0004-9158

Abstract

ABSTRACT: An aspatial mathematical model has been developed to simultaneously support log processing investment decisions with respect to processing scale, facility location and log procurement when data are scarce. A key design principle was to make the model suitable for use by industry. The objective function maximises gross margins per hour of log processing time, and the model accounts for potential processing efficiencies with larger-diameter logs. To support log procurement decisions, the model estimates mill-delivered log costs at which a log procurement officer should be indifferent between purchasing alternative log types. The utility of the model is demonstrated with an application to rotary veneer processing of hardwood logs in subtropical eastern Australia. Complex interactions between processing scale, facility location and log procurement strategies were revealed by substantial differences in gross margins between modelled scenarios. Log procurement decisions were found to have the greatest potential impact on gross margins, followed by facility location and processing scale. The model highlighted that substantially higher returns can be earned from optimal log procurement strategies relative to approaches that either minimise log costs, maximise product recovery or do not differentiate between log types and simply utilise all available log volume.