Optimizing tensile strength of PLA-Lignin Bio-composites using machine learning approaches

• Main Authors: Manshor, Mohd Romainor, Kamarulzaman, Amjad Fakhri, Anuar, Hazleen, Toha @ Tohara, Siti Fauziah, Ali, Fathilah, Sukindar, Nor Aiman, Suhr, Jonghwan, Haris, Nursyam Dzuha
• Format: Conference or Workshop Item
• Language: English; English
• Published: Springer 2023
• Subjects: TS195 Packaging
• Online Access: http://irep.iium.edu.my/104852/2/104852_Optimizing%20tensile%20strength.pdf; http://irep.iium.edu.my/104852/7/104852_Optimizing%20tensile%20strength_SCOPUS.pdf; http://irep.iium.edu.my/104852/; https://link.springer.com/chapter/10.1007/978-981-19-9509-5_58
• Institution: Universiti Islam Antarabangsa Malaysia

It is imperative to accurately estimate the final performance of composite parts during the initial design phase of the manufacturing process. In generating sustainable bio composites with superior mechanical properties such as tensile strength, the combination of fillers and plasticizers, as well as their concentration in the mixture, are always deemed crucial. In order to reduce the number of experimental runs and their associated costs and timescales, statistical optimization of the core design elements has become increasingly important. The filler and plasticizer concentrations of extruded bio composites were adjusted in this study utilizing both statistical (analysis of variance (ANOVA) and response surface methodology (RSM)) and machine learning (Multilayer Perceptron (MLP)) approaches. Initial ANOVA results indicated that lignin, epoxidized palm oil (EPO), and their respective combinations were the most influential factors in enhancing the durability of lignin/polylactic acid (PLA) bio composites. In this work, RSM and MLP were used to model and predict the data in order to maximize the various solutions and establish the nonlinear relationship between the concentration of lignin and EPO.