Design and evaluation of an intelligent waste monitoring system based on RGIS integration for smart cities

Abstract

The object of this study is the municipal solid waste management system within a modern urban environment, where rapid urbanization and population growth pose significant challenges to ecological sustainability. The key problem addressed is the inefficiency of waste collection due to overflowing containers, poor route planning, and suboptimal resource allocation. To tackle these issues, an intelligent waste monitoring system has been developed that integrates Internet of Things (IoT) technologies, computer vision, data analytics, and a Regional Geographic Information System (RGIS). The system includes a computer vision model that analyzes images of waste containers to determine their fill level. Fine-tuning the model on locally collected image data, reflecting regional characteristics such as lighting, container types, and weather conditions, significantly improved detection accuracy and adaptability. Route optimization for waste collection is implemented using a mathematical formulation of the Traveling Salesman Problem (TSP), solved via Mixed Integer Linear Programming (MILP), which helped reduce fuel consumption, travel time, and staff workload. Integration with RGIS and GPS enables dynamic routing and real-time geospatial visualization of operational data. The proposed system forms a closed-loop control cycle that links automated detection, spatial analysis, and decision-making. Experimental results demonstrate high efficiency, adaptability to regional conditions, and scalability, confirming the system’s practical applicability to other urban areas. In the future, the system may be expanded to include environmental monitoring modules such as air quality, noise, and soil conditions and predictive modeling of waste generation, thereby supporting the sustainable development of smart city infrastructure

More by Nurzhamal Kashkimbayeva

• Evidence-Linked Measurement of Indirect Prompt Injection in Web Security Agents: Paired Trials and Application-Specific Evidence AuditsTechnologies · 2026
• Kazakhstan Agri Recommender — Data Collection PipelineZenodo · 2026
• Replication Data and Code for “Added Predictive Value and Temporal Transferability of MODIS and ERA5-Land Data for District-Level Wheat Yield Forecasting”Harvard Dataverse · 2026
• Multimodal Data Fusion for District-Scale Crop Yield Prediction in Northern KazakhstanIEEE International Conference on Smart Information Systems and Technologies · 2026