Multi-Response Optimization of Banana Pseudostem Biopolymer Films Using Python for Sustainable Packaging
Contributors
Chetan Kapadnis
Dr. Shrikaant Kulkarni
Keywords
Proceeding
Track
General Track
License
Copyright (c) 2026 Sustainable Global Societies Initiative

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Abstract
Banana pseudostems are underused agricultural waste which has the potential to be used as a sustainable resource in the production of biodegradable food packaging. However, the performance of films made from these materials will depend upon how different formulation variables interact with each other. A formulation that enhances one characteristic could also result in decreased performance of another. Therefore, this study utilizes a workflow based in Python and derived from peer-reviewed literature to identify through use of Multi-Response Desirability Analysis and Pareto Analysis, published banana-pseudostem film formulations. Six studies were evaluated in terms of material form, matrix, amount of reinforcement and measured mechanical/barrier characteristics. Due to differences in matrices and testing methods, the data from those experiments cannot be pooled together to create a single predictive model. For the principal analysis, the data from the five Starch/Banana-Pseudostem formulations reported by Othman et al., (2020) was analyzed separately from the Potato-Starch/Banana-Cellose data reported by Gudeta, (2026). The results from the balanced desirability analysis indicated that 20 wt % pseudostem powder from Othman's database would be the best option. In addition, the results from both analyses indicated that there existed response trade-offs among all tested formulations. The outcome of this analysis provides a literature-derived screening candidate; however experimental verification under uniform test conditions is necessary.