Comparison between multiresponse‐robust process design and numerical optimization: A case study on baking of fermented chickpea flour‐based wheat bread

Abstract

AbstractThe baking conditions (relative humidity, baking time, and temperature) for a fermented chickpea flour‐based wheat bread were optimized considering multiple responses: crumb hardness, specific loaf volume, color change in crust and crumb. Rotatable central composite design was applied within the baking domain of 175–225 °C/13–22 min/45–95% RH. An intelligent and robust technique based on Taguchi factor‐effect approach, principal component analysis, gray relational analysis, artificial neural networks, and genetic algorithms was employed for this purpose. Response surface methodology (RSM) with numerical optimization was also applied and the results were compared with the intelligent system‐multiresponse‐robust process design (IS‐MR‐RPD). The optimized conditions suggested by RSM and IS‐MR‐RPD were 195 °C/15 min/85% RH and 201.5 °C/15 min/85% RH, respectively. The ability to predict by RSM toward inferring the relative importance of factors is commendable; however, IS‐MR‐RPD predicted a more accurate data compared to the largely deviated data obtained from RSM.Practical applicationsIn the case of a staple food like bread, optimizing the baking condition to produce a cost‐effective, attractive, and nutritive product is the real need for the bakery industry. To fulfill both the consumer and manufacturer demands, different intercorrelated responses like texture, color appearance, and loaf volume are considered prior to process optimization. This study showed the efficacy of optimization between two different techniques for the baking of fermented chickpea flour‐based wheat bread. The two techniques are response surface modeling and multiresponse‐robust process design. The outcome of this study will help the industry to apply the robust process design for any food process optimization.