Premium
Review article: Basic steps in adapting response surface methodology as mathematical modelling for bioprocess optimisation in the food systems
Author(s) -
Nwabueze Titus U.
Publication year - 2010
Publication title -
international journal of food science and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.831
H-Index - 96
eISSN - 1365-2621
pISSN - 0950-5423
DOI - 10.1111/j.1365-2621.2010.02256.x
Subject(s) - response surface methodology , bioprocess , computer science , process (computing) , design of experiments , mathematical model , multivariate statistics , polynomial regression , software , mathematical optimization , biochemical engineering , machine learning , regression analysis , mathematics , engineering , statistics , chemical engineering , programming language , operating system
Summary Techniques involving choosing process combinations for optimisation without due consideration for relevant experimental designs is scientifically unreliable and irreproducible. Mathematical modelling, of which response surface methodology (RSM) is one, provides a precise map leading to successful optimisation. This paper identified key process variables, building the model and searching the solution through multivariate regression analysis, interpretation of resulting polynomial equations and response surface/contour plots as basic steps in adapting the central composite design to achieve process optimisation. It also gave information on appropriate RSM software packages and choice of order in RSM model and data economy in reducing the factorial experiments from large number parameter combinations to a far less number without losing any information including quadratic and interaction (if present) effects. It is expected that this paper will afford many food scientists and researchers the opportunity for adapting RSM as a mathematical model for achieving bioprocess optimisation in food systems.