Quantifying natural variation of formaldehyde emissions for wood composite panels
Young, T. M., J. Otjen, and N. O. Andre.  2014.  PTF BPI 2014. 3rd Intern. Conf. Processing Tech. for the Forest and Bio-based Products Industries, Salzburg/Kuchl, Austria, 24-26 September 2014. p. 461-465.

Quantifying the natural variation of formaldehyde (HCHO) emissions during the manufacture of wood composite panels (e.g., particleboard, medium density fiberboard or MDF, etc.) and estimating the costs incurred by controlling the natural variation of HCHO emissions are the goals of this study. Natural variation of HCHO emissions from wood composite panels in the presence of regulatory limits directly influences the operating target that manufacturers’ must maintain, i.e., larger natural variation of HCHO emissions requires lower operating targets for HCHO emissions which increases the costs of manufacture. Taguchi’s “Loss Function” is used to estimate the costs incurred from controlling variability in HCHO emissions. Taguchi’s key premise is that costs in a manufacturing process increase at a non-linear rate for deviations from the operating target (two-sided for lower and upper specifications) or specification limit (one-sided for only one specification limit). In this study, the one-sided loss function was used to estimate the costs incurred due to variation in HCHO emissions. Statistical models are developed of HCHO emissions as a function of multiple independent process variables. Statistical models with R2 > 0.70 are used to estimate the total variance of HCHO emissions from particleboard manufacturing processes. Examples of Taguchi loss analyses for producers with a range of variability in HCHO emissions will be reviewed.