The evolution of knowledge in the forest products manufacturing
Young, T. M., A. Petutschnigg, and M. Barbu.  2013.  9th edition of the International Conference “Wood Science and Engineering in the Third Millennium”, p. 22-27.  (invited)

The manuscript is conceptual; it assesses the use of real-time process data by forest products manufacturers as related to data discovery and the co-evolutionary existence of data mining inductive algorithms. Forest products companies have intellectual latency concerning process variation and such latency leads to higher than necessary costs of manufacturing, inferior product quality, and inaccurate prediction. Strength properties from the laboratory are unreliable for real-time prediction given their historic temporal orientation. Induction of real-time process data when aligned with destructive test data from the laboratory improves manufacturing latency and results in discovery of unknown sources of process variation. This discovery of unknown sources of process variation allows for quantification of variation and defines the relationship with product attributes which improves accuracy of predictions. The manuscript outlines the importance of the real-time relational database and data quality verification as essential steps in the data discovery journey. These essential steps are often overlooked and lead to antidotal analyses that limit process improvement and predictability. Statistical methods for inductive data mining are discussed, e.g., multiple linear regression analysis (MLR), regression trees (RT), principal component analysis (PCA), and quantile regression. The strengths and weaknesses of each method with appropriate citations are noted.