TY - JOUR
T1 - Uncertainty and sensitivity analysis: Mathematical model of coupled heat and mass transfer for a contact baking process
AU - Feyissa, Aberham Hailu
AU - Gernaey, Krist
AU - Adler-Nissen, Jens
PY - 2012
Y1 - 2012
N2 - Similar to other processes, the modelling of heat and mass transfer during food processing involves
uncertainty in the values of input parameters (heat and mass transfer coefficients, evaporation rate
parameters, thermo-physical properties, initial and boundary conditions) which leads to uncertainty in
the model predictions. The aim of the current paper is to address this uncertainty challenge in the modelling
of food production processes using a combination of uncertainty and sensitivity analysis, where the
uncertainty analysis and global sensitivity analysis were applied to a heat and mass transfer model of a
contact baking process. The Monte Carlo procedure was applied for propagating uncertainty in the input
parameters to uncertainty in the model predictions. Monte Carlo simulations and the least squares
method were used in the sensitivity analysis: for each model output, a linear regression model was constructed
and the standardized regression coefficients (SRCs) and R2 were computed. The effect of input
parameters on model predictions was calculated, and the relative impact of the parameters on each of
the outputs was ranked. Results of the uncertainty and sensitivity analysis can be used to prioritize future
experimental efforts, as discussed for the contact baking process.
AB - Similar to other processes, the modelling of heat and mass transfer during food processing involves
uncertainty in the values of input parameters (heat and mass transfer coefficients, evaporation rate
parameters, thermo-physical properties, initial and boundary conditions) which leads to uncertainty in
the model predictions. The aim of the current paper is to address this uncertainty challenge in the modelling
of food production processes using a combination of uncertainty and sensitivity analysis, where the
uncertainty analysis and global sensitivity analysis were applied to a heat and mass transfer model of a
contact baking process. The Monte Carlo procedure was applied for propagating uncertainty in the input
parameters to uncertainty in the model predictions. Monte Carlo simulations and the least squares
method were used in the sensitivity analysis: for each model output, a linear regression model was constructed
and the standardized regression coefficients (SRCs) and R2 were computed. The effect of input
parameters on model predictions was calculated, and the relative impact of the parameters on each of
the outputs was ranked. Results of the uncertainty and sensitivity analysis can be used to prioritize future
experimental efforts, as discussed for the contact baking process.
U2 - 10.1016/j.jfoodeng.2011.09.012
DO - 10.1016/j.jfoodeng.2011.09.012
M3 - Journal article
SN - 0260-8774
VL - 109
SP - 281
EP - 290
JO - Journal of Food Engineering
JF - Journal of Food Engineering
IS - 2
ER -