Segregation of Municipal Solid Waste Based on Their Biodegradability Using Spectroscopy Sensor

The world today is facing a colossal increase in environmental pollution and degradation, resulting in ecological imbalance due to urbanization and development of smart cities. This necessitates the requirement for a robust waste management system, which can properly segregate waste based on its types and biodegradability. Segregating small nonbiodegradable waste from a heterogeneous mix of waste is a very big challenge. This letter proposes a rapid and efficient system to tackle this challenge where the system uses a spectroscopy sensor to collect information from the waste object and an algorithm to classify the nonbiodegradable waste materials from this heterogeneous mix of waste. Samples belonging to three object classes, namely, plastics, paper/ cardboard, and organic waste, were considered for testing the system, which uses regression model, confusion matrix, and an optimization technique to reduce error and obtain the best output closest to the ideal solution. This system gives the best output at 705 nm of wavelength with threshold values of 40.65 and 25.04. If the sensor output for a particular sample lies between 40.65 and 25.04, then it is a nonbiodegradable waste, whereas if the value lies above 40.65 or below 25.04, then it is a biodegradable waste. The experimental results demonstrate that the proposed system with the algorithm satisfactorily classifies the waste into biodegradable and nonbiodegradable waste.