Optimization of plantation sawmill performance in Uganda.

Abstract

Sawmills operating in Uganda’s plantations have limited decision support systems for sawing, which partly accounts for their poor performance in terms of both volume recovery and value recovery. The objective of this study was to contribute to improvement of plantation sawmill performance by developing a tool to aid in decision making when sawmilling. Specific objectives were to determine plantation sawmill performance, determine potential conflict in sawing strategies for optimizing performance, and determining potential improvement in performance when a decision support tool is used. The study was conducted on four sawmills operating in three plantation clusters. Quasi experiments were used to collect data on log diameter, length and timber sizes produced for determining performance and the potential improvement. Mathematical analysis was used to develop sawing patterns for determining potential conflicts in sawing strategies for optimal performance. Timber volume recovery was 32% while value recovery was 123,800 UGX/m3. A two-way ANOVA indicated significant difference (p =0.00) in volume recovery between sawmills and a significant difference (p = 0.00) in value recovery between sawmills. The medium band sawmill had the highest volume and value recovery while the mobile circular sawmill had the lowest volume and value recovery rates. A paired samples t-test indicated a significant difference (p = 0.00) in both value and volume recovery between sawing strategies. Volume recovery of plantation sawmills was 39% and would reduce by 2% if value maximizing strategy was used. On the other hand, the value recovery was 195,200 UGX/m3 and would reduce by 11% if volume maximizing sawing strategy was used. An independent t-test indicated a significant difference (p = 0.00) empirical volume recovery (41%) and simulated volume recovery (45%) obtained by using a decision support tool. Empirical value recovery (217,400 UGX/m3) was also significantly different (p = 0.00) from Simulated timber value recovery (277,400 UGX/m3). Therefore, plantation sawmill performance is poor and is being affected by sawmill technology and log diameter sizes. There also exist conflicts in sawing strategies for optimal performance and a potential improvement in performance of sawmills when a decision support tool is used. Further studies on the effect of demand, sawyer’s experience and product mix on performance and testing, validation of the decision tool and assessing feasibility of process modification for tool adoption were recommended.