Abstract

Increasing environmental awareness and promoting sustainable soil management are crucial for addressing ecological challenges posed by plastic pollution and ensuring the long-term health of our soils. This study introduces a broad, algorithm-guided experimental approach that integrates environmental monitoring, data science and microbiological analysis to assess the biodegrading dynamics of various biodegradable materials like polylactic acid (PLA), polyhydroxyalkanoates (PHAs)and the starch-based composites with advanced machine learning models in a data-handled platform that evaluates the degradation behavior of biodegradable polymers and their impact on the soil's physicochemical and biological properties. Utilizing the Random Forest Regression (RFR) and Support Vector Machines (SVM) examines nutritional cycling efficiency, microbial social structure, carbon dynamics, and parameters such as soil enzyme activities under both laboratory control and field-relevant conditions. Real-time soil monitoring is activated through IoT-based sensors that measure moisture, temperature, pH, and CO₂ flux, which combined with laboratory analyses, feeds in models of predictions that guide repetitive adjustments in material soil interaction. Among the tested materials, starch-based composites performed the fastest biological degradation (72%), followed by PHA (55%) and PLA (28%), correlated with increased microbial activity and enzyme function. The machine learning model performed high predictions (R2> 0.89) able to make real-time decisions. This adaptive model supports deep insight into soil-biopolymer interactions and promotes environmental skills through a visual dashboard that allows users to explain the dynamic soil condition. The interdisciplinary structure not only provides technological progress for soil monitoring but also as an educational tool, which encourages the practices that are informed in permanent agriculture and public environmental stewardship. Overall, the integration of biodegradable materials with intelligent monitoring systems provides a double advantage: An available platform for tangible improvement in soil health and participation science.