A Cloud-Based Temporal Pattern Mining Framework for Enhanced Compliance Monitoring in Municipal Solid Waste Systems

Abstract

 Managing municipal solid waste effectively requires knowing how waste generation changes over time and whether people are following segregation rules. This paper presents a computer-based system that automatically finds patterns in waste data and checks compliance using real-world collection records.

We analyzed 9,839 waste collection records—totaling 1,704,275 kilograms—from Deolali Pravara, India, covering the years 2022 to 2024. Our method uses a technique called STL decomposition to detect seasonal patterns, statistical methods to identify unusual data points, and new measures to evaluate segregation quality.