Multi-Response Optimization of Quality Parameters in Pumpkin–Oat Composite Noodles using Response Surface Methodology

Abstract

Pumpkin (Cucurbita maxima) and oat (Avena sativa L.) flours are valuable functional ingredients that can enrich cereal-based staples such as noodles with β-carotene, β-glucan, and dietary fibre while displacing part of the refined wheat flour normally used [1]. However, partial replacement of wheat flour dilutes gluten and alters water-binding behaviour, which can compromise cooking stability, texture, and consumer acceptability if substitution levels and hydration are not carefully controlled [1,3,4]. This study applied a three-factor, three-level Box–Behnken design (BBD) to evaluate and optimize the effect of pumpkin flour (5–15%), oat flour (5–15%), and water addition (30–38%) on six quality responses of composite noodles: cooking time, cooking loss, water absorption index (WAI), texture hardness, overall sensory acceptability, and β-carotene content. Second-order polynomial models were fitted to each response (R² = 0.974–0.997), and the Derringer–Suich desirability function was used to identify the formulation that simultaneously satisfied all quality targets [6]. The numerically optimized formulation (12% pumpkin flour, 9% oat flour, 35% water addition) gave an overall desirability of 0.79, with predicted cooking time of 6.76 min, cooking loss of 5.07%, WAI of 2.13 g/g, hardness of 952 N, sensory score of 7.55 (9-point hedonic scale), and β-carotene content of 225.2 µg/100 g. Validation trials conducted at the optimum point showed relative errors below 2% between predicted and experimental values for all responses, confirming the adequacy of the fitted models. The results demonstrate that RSM-based multi-response optimization is an effective strategy for developing nutritionally enhanced pumpkin–oat composite noodles without compromising technological and sensory quality.