Design and Development of GIOT based Intelligent Smart Waste Management and Predictive Modelling

Waste management is one of the important factors which determine the health of an urban/rural area. Maintaining the surroundings clean and tidy in waste management becomes a tedious task for the government. To resolve this issue intelligent monitoring takes immediate effect on solid waste dust bins for making a green and safe environment. In this paper, a design of an intelligent smart-based waste management system (ISWM) with a predicted forecast is proposed. The proposed smart bin model uses ultrasonic, flame, gas, and accelerometer sensors these data is processed by the low power microcontroller called ESP-32. Initially, all the sensor data are processed by using ESP32 and statistical data is stored in the cloud. The proposed system consists of two parts the first step is a notification of the smart bin status to the user by the Global Internet of Things (GIoT) and secondly, smart bin sensor data is captured to form a dataset and the level of bin status in the dataset monitored and interpreted by using multivariate Seasonal AutoRegressive Integrated Moving Average with eXogenous (SARIMAX) model. The proposed SARIMAX model forecasts the abnormality of the sensor data immediately sends the notification to the concerned user. In this work, the bin management framework is designed to monitor solid waste at all times. This work promotes cleanliness and efficient management of resources. Our proposed model performance metrics such as RMSE and MAPE are 0.6145 and 3.20 respectively, which shows the forecasting values have fewer deviations when compared to the actual values.