DESIGN OF AN INTELLIGENT WASTE COLLECTION ROUTE OPTIMIZATION FRAMEWORK USING DECISION TREE ALGORITHMS FOR A CLOUD-BASED BIG DATA REPOSITORY

Abstract

Poor waste management is one of the factors that had significant implications for public health. Currently, the process of solid waste collection in most urban cities of developing countries is curbside collection and is marred with poor disposal infrastructure, time consuming and high transportation cost. The study proposes a cloud-based big data repository for solid waste management in Owerri metropolis, Eastern Nigeria, by using a decision tree algorithm for optimizing collection routes. This system will consider factors such as: data storage, real-time processing, cloud resources, data sources, GPS data integration, user accessibility data analytics and reporting. The proposed system will make use of machine learning with decision tree algorithms for waste classification and route optimization, Azure Power BI for data visualization and reporting, and service-oriented architecture with Microsoft Azure platform as a service (PaaS) and Python programming. Azure Data Factory will provide secure and scalable data storage, and IoT Hub will automate the process of ingesting data from bin sensors, IoT devices, or manual data input. In order to ensure optimum use of disposal infrastructures, minimize transportation costs related to trash collection, and increase labor productivity, the system will assist in optimizing waste collection routes for an efficient and effective waste collection process