Parameter Identification for a Mechanistic Model of Poly-β-hydroxybutyrate Production

The use of detailed models for bioprocess design and control has been limited as accurate estimation of model parameters is often difficult. In this paper, the parameter estimation problem for a mechanistic model of the production of a biopolymer, poly-β-hydroxybutyrate (PHB), is examined in detail. Parameter estimation was undertaken using previously published data. Three parameter sets were obtained with large uncertainty in some parameters. Parametric sensitivity, identifiability and estimability analyses indicated that only certain parameters were uniquely identifiable. Experiment design studies indicated that carefully designed experiments could significantly reduce the uncertainty in the parameter estimation problem.