Related Experiment Video
Updated: Dec 30, 2025
06:24
Generic Protocol for Optimization of Heterologous Protein Production Using Automated Microbioreactor Technology
Published on: December 15, 2017
10.6K
Improved Pullulan Production and Process Optimization Using Novel GA-ANN and GA-ANFIS Hybrid Statistical Tools
Parul Badhwar1, Ashwani Kumar2, Ankush Yadav3
1Microbial Process Development Laboratory, University Institute of Engineering and Technology, Maharishi Dayanand University, Rohtak-124001, Haryana, India.
Biomolecules
|January 16, 2020
Summary
Mathematical tools optimized pullulan production from Aureobasidium pullulans, increasing yield. Genetic Algorithm-Adaptive Neuro-Fuzzy Inference System (GA-ANFIS) accurately predicted optimal conditions for maximum pullulan yield.
Area of Science:
- Biotechnology
- Bioprocess Engineering
- Mathematical Modeling
Background:
- Pullulan, a polysaccharide produced by Aureobasidium pullulans, has diverse industrial applications.
- Optimizing pullulan production is crucial for enhancing yield and economic viability.
- Traditional optimization methods like one variable at a time (OVAT) can be time-consuming and may not capture complex interactions.
Purpose of the Study:
- To explore advanced non-linear hybrid mathematical tools for optimizing pullulan production process variables.
- To compare the predictive accuracy of genetic algorithm coupled with artificial neural network (GA-ANN) and genetic algorithm coupled with adaptive network based fuzzy inference system (GA-ANFIS).
- To experimentally validate the optimized parameters and predicted pullulan yield.
Main Methods:
- Utilized one variable at a time (OVAT) approach for initial optimization.
- Employed GA-ANN and GA-ANFIS as non-linear hybrid mathematical tools for process optimization.
- Performed experimental validation of the parameters predicted by the mathematical models.
- Analyzed regression values (Levenberg-Marquardt algorithm for ANN) and epoch error for GA-ANFIS to assess model performance.
Main Results:
- OVAT approach yielded a maximum pullulan concentration of 35.16 ± 0.29 g/L.
- GA-ANN predicted a maximum pullulan yield of 39.4918 g/L.
- GA-ANFIS predicted a maximum pullulan yield of 36.0788 g/L with high precision (epoch error: 6.1055 × 10⁻⁵).
- Experimental revalidation of GA-ANFIS parameters achieved 98.82% accuracy.
- Optimal conditions predicted by GA-ANFIS include substrate concentration (49.94 g/L), incubation period (182.39 h), temperature (27.41 °C), pH (6.99), and agitation speed (190.08 rpm).
Conclusions:
- Non-linear hybrid mathematical tools, particularly GA-ANFIS, offer superior prediction accuracy for pullulan yield optimization compared to OVAT.
- GA-ANFIS successfully identified optimal process parameters leading to a predicted pullulan yield of 36.0788 g/L.
- The study demonstrates the potential of data-driven modeling for efficient bioprocess optimization in polysaccharide production.
