Department of Mechanical Engineering, Sharda University, SET, Greater Noida, Uttar Pradesh, India
Abstract
The global pursuit of lightweight, high-performance, and sustainable materials has paved the way for the widespread use of plant fiber composites in various industries, ranging from automotive to aerospace, construction, and consumer goods. Despite their potential, predicting the mechanical properties of plant fiber composites remains a complex challenge due to the inherent variability of natural fibers, diverse processing parameters, and the multitude of performance requirements. This paper proposes a conceptual framework for developing a multi-objective predictive model that elucidates the relationships among material constituents, fabrication parameters, and resulting mechanical properties in plant fiber composites. The proposed model integrates methodologies from statistical, computational, and artificial intelligence (AI) domains—namely response surface methodology (RSM), finite element analysis (FEA), artificial neural networks (ANNs), adaptive neuro-fuzzy inference systems (ANFIS), and multi-objective genetic algorithms (MOGAs)—to address simultaneous optimization objectives. An extensive, PhD-level literature review underscores the theoretical foundations and best practices for studying plant fiber composites, while also highlighting critical research gaps such as moisture sensitivity, thermal stability, interfacial adhesion, and limited predictive frameworks. By incorporating state-of-the-art approaches, the paper proposes a robust, multi-objective decision support tool capable of guiding the selection and design of plant fiber composites with targeted mechanical properties. This conceptual framework offers a blueprint for future experimental validations and industrial applications, ultimately advancing the sustainable adoption of plant fiber composites in high-performance engineering solutions.
Keywords: Adaptive neuro-fuzzy inference system, Artificial neural network, Finite element analysis, Genetic algorithm, Mechanical properties, Multi-objective modelling, Plant fiber composites
View PDF