Browsing by Author "Palma, Cristian D"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
Item Impact of timber volume and grade estimation error on the british columbia coastal supply chain(2015) Vergara, Francisco P; Palma, Cristian D; Nelson, John DTimber supplies are particularly affected by volume and timber grade uncertainty in the current forest inventory. The performance of the British Columbia coastal timber industry could therefore be improved by developing policies to manage this uncertainty in timber supplies. However, most research has focussed only on the effect of timber volume uncertainty on forest scheduling decisions. Hence, the objective of the research presented here was to determine the impact of volume and grade errors in the forest inventory when solving the tactical lumber planning problem, and to develop a policy to address customer service reduction due to uncertainty. A decision support system based on a linear program was formulated as a case study to implement production planning for the British Columbia coastal forest. Four error scenarios were analyzed based on the sample size used to estimate timber grading error, the size of the cut block in use, and volume/grade error assumptions. Solutions were evaluated based on these four error scenarios and 12 lumber demand targets. A probability curve for fullling lumber demand was determined, as well as prot variations for the error scenarios. Prot variation was consistent with the error magnitude. In the normal-size cut-block error scenario, the rate of change in average prot was 6.05% lower than the perfect prot estimate with perfect information. A 25% reduction in timber grade estimation, which reduced the prot variation to 1.43%, was also tested. However, in the large-size cut block (cut block >50 ha) error scenario, the rate of change of average prot was 2.24% lower than the prot determined by perfect estimation. On the other hand, applying a 25% reduction in timber grade estimation reduced the prot variation to 0.48%. The ability to full lumber demand was compromised by timber volume and grading errors. Naturally, the reductions were greater for higher-error scenarios; however, if timber grading errors can be reduced, lumber demand targets can be reduced as well, leading to a less risky operation. Although this performance drawback is not particularly relevant to the U.S. housing market, under changing market conditions, it could compromise the industry’s ability to participate in other cut-to-order low-value lumber markets with more demanding due dates and lumber demand fullment requirements (e.g., the Japanese thin-board market).