Optimization of Vacuum Frying Process for Sweet Potato Chip Manufacturing Using Response Surface Methodology and Artificial Neural Network Model

The purpose of this study was to optimize product yield and quality of the sweet potato chip manufacturing process in a pilot-scale industrial fryer using vacuum frying (VF) technology, response surface methodology (RSM), and artificial neural network (ANN) model. The variables, osmotic dehydration...

Full description

Saved in:
Bibliographic Details
Published inBiotechnology and bioprocess engineering Vol. 28; no. 4; pp. 554 - 567
Main Authors Kim, Da-Song, Lee, Jung Heon, Shin, Hyun-Jae
Format Journal Article
LanguageEnglish
Published Seoul The Korean Society for Biotechnology and Bioengineering 01.08.2023
Springer Nature B.V
한국생물공학회
Subjects
Agricultural production
Artificial neural networks
Biotechnology
Chemistry
Chemistry and Materials Science
Computer science
Crop yield
Dehydration
Design of experiments
Engineering
Food
Food quality
Frying
Industrial and Production Engineering
Manufacturing
Manufacturing industry
Neural networks
Optimization
Performance evaluation
Performance prediction
Potatoes
Research Paper
Response surface methodology
Statistics
Sweet potatoes
Vacuum
Vegetables
생물공학
artificial neural network model
vacuum frying
response surface methodology
optimization
sweet potato chip
Online AccessGet full text

Cover

More Information
Summary:The purpose of this study was to optimize product yield and quality of the sweet potato chip manufacturing process in a pilot-scale industrial fryer using vacuum frying (VF) technology, response surface methodology (RSM), and artificial neural network (ANN) model. The variables, osmotic dehydration (OD) concentration, OD temperature, and VF temperature were designed to optimize the yield, oil content, and browning index (BI) of vacuum-fried sweet potato chips. Yield, oil content, and BI achieved optimal conditions for 52.46%, 10.65%, and 61.14 in RSM, and 53.52%, 11.58%, and 60.40 in ANN, respectively. Based on the statistical evaluation performance, the ANN model had a higher predictive performance than the RSM model. These findings highlight the high-quality pilot-scale manufacturing process along with a better statistical approach. Moreover, the optimized process can be used for the commercial production of vacuum-fried sweet potato chips.
ISSN:1226-8372
1976-3816
DOI:10.1007/s12257-023-0061-0