Enhancing operational efficiency in a voluntary recycling project through data-driven waste collection optimization

Area of Science:

• Operations Research
• Environmental Science
• Computer Science

Background:

• Recycling in developing nations relies on voluntary, often non-profit, private sector initiatives.
• Operational efficiency is a major challenge for these initiatives, particularly in logistics.
• Traditional optimization methods like the Traveling Salesman Problem are too time-consuming for practical, daily use.

Purpose of the Study:

• To develop a data-driven, three-step approach to optimize waste collection logistics for non-profit recycling projects.
• To enhance the efficiency and practicality of waste collection operations in resource-constrained environments.
• To address the challenges of maximizing collected materials and minimizing travel distances.

Main Methods:

• Geographical grouping of collection points using K-means clustering.
• Optimization models to maximize recyclable plastic per trip and determine efficient routes.
• Integration of real-time waste volume data and live traffic information via an Intelligent Traffic Information Center.

Main Results:

• Demonstrated significant daily increases in the average amount of plastic collected.
• Achieved a notable reduction in the average distance traveled for waste collection.
• Validated the approach through a case study involving 152 collection points for the Won Project.

Conclusions:

• The proposed data-driven method provides prompt, reliable solutions for daily waste collection operations.
• Open-source software enables practical implementation, overcoming logistical hurdles in voluntary recycling projects.
• The approach enhances the sustainability and effectiveness of recycling initiatives in developing countries.