Application of Machine Learning in Waste Management: An Introductory Approach.
Publication Date: 19/07/2024
Author(s): Uchenna Chinyere Onyemauche, Jonathan Ugwu Okwor , Charity Elochukwu Mbanusi, Blessing Ulunma Osondu.
Volume/Issue: Volume 4 , Issue 2 (2024)
Abstract:
Waste Management is a daily task in urban areas, which needs a huge amount of labour resources and this affects natural, budgetary, efficiency and social aspect of our cities. Manual sorting of garbage is a difficult process that is expensive and that is why scientists create and study automated sorting methods that increase the efficiency of the recycling process. Most recently, there has been a drift in combining waste that is prime scheme with low cost IoT architectural design on a test board. However, the results from all these past approaches and techniques are still not clear and cannot be applied in real systems, such as in cities and campuses. This work introduces the design of a micro controller that is single, low cost, straight forward with an ultrasound sensor which can measure the filling height of a garbage Trash Bin and send information using LORA Technology. A novel IoT based Machine Learning method in combination with Genetic Algorithm to predict the probability of collecting waste in real environment based on historical data were used in this study. This is combined with a microcontroller system designed with a sensor module for measuring the height that is the fillings levels of each trash bin. The system can optimize the collection of waste with the shortest path by using genetic algorithm. Python was used for analyzing the data.Using the above Machine Learning techniques like Logical Regression cum Genetic Algorithm to compute the paths of wastes collection with different time schedules, it is cumbersome to get efficient route optimization; hence the aim of this paper was to present an IoT cloud solution combining device connection, data processing, control and ensuring route optimization. Genetic Algorithm ensures perfect route optimization.
Keywords:
IoT, Machine Learning, Deep Learning, Waste Management, Genetic Algorithm, Logical Regression.
