Non Normal Distribution Transformations in Continuous Reactor Scale Up
Applying Box-Cox and Johnson transformations to skewed residence time data prevents major capability miscalculations during continuous reactor scale up.
Johnson SB Distribution
Meaning
Probability distribution belonging to the Johnson system that models variables restricted to a specific finite interval. The johnson sb distribution is defined by four parameters that allow it to cover a wide range of skewness and kurtosis combinations within bounded limits. It is frequently employed in quality control to characterize measurements that cannot exceed physical or engineering tolerances.
This flexibility makes it more versatile than a simple beta distribution for fitting empirical data from manufacturing processes.
Parameter Estimation
Fitting the model to a data set involves calculating the translation and shape parameters. The johnson sb distribution requires a transformation of a standard normal variable to map onto the desired range. Moments or percentiles from the sample provide the basis for these calculations.
Transformation Logic
Values are mapped from a normal curve to a bounded range using a specific logarithmic function. In a johnson sb distribution, the resulting curve can be unimodal or bimodal depending on the shape parameters. This allows for the representation of complex process behaviors that exhibit asymmetry.
The transformation remains mathematically stable for computational simulations.
Capability Analysis
Process engineers use this model to predict the probability of producing parts outside of specification limits. Because a johnson sb distribution accounts for non normal data, it yields more reliable Cpk values for skewed distributions. Applying a standard normal assumption to bounded data often leads to an overestimation of defect rates.