<p>This study presents an innovative approach to enhance the efficiency of composite material by utilizing industrial waste by-products like leather trimming waste and high-density polyethylene waste. This study utilizes advanced machine learning approaches such as artificial neural networks (ANN) and linear regression models, to predict the mechanical properties of composite materials over a range of compositions. The produced composite materials are evaluated using a thorough series of systematic experimental testing. Mechanical test assess important parameters such as tensile strength, tear strength, flexural strength, and hardness. These tests demonstrate the adaptability and promising potential of the materials. The ANN and logistic regression models have impressive prediction accuracies of 93.38% and 90.25% respectively. This suggests that they are highly effective in forecasting mechanical properties by analyzing composite composition. Moreover, the models' predictive powers are validated by performance metrics such as precision, recall, and F1 score. By conducting a thorough investigation, we have identified a composition that possesses superior mechanical properties compared to traditional materials. This suggests that it can be efficiently used in a wide range of industries for packaging, construction, and automotive. This study highlights the possibility of using industrial waste for improving the performance of composite materials, which in turn promotes sustainability and resource efficiency. Furthermore, it lays the groundwork for future research on enhancing material properties through the application of machine learning methods Thereby, promoting cutting-edge engineering methodologies and facilitating the progress of sustainable material advancement.</p>