Framework for Waste Collection Problem in a Low-Tech Urban Setting in India

This paper presents an operations-research-based improvement framework for a waste collection system in the capital city of India, Delhi. The existing manual system divides the urban region into ‘wards’, each with its depot (or garage) of trucks, and a ‘greedy’ method to assign trucks to collect waste from collection points. Here, we propose a framework that encourages the use of OR-based modelling in a low-tech setting. It is based on two types of interventions: system change and system optimization. In system change, a new routing process is developed using a mixed-integer linear programming (MILP) model to address the waste collection problem, which is formulated as a combinatorial optimization known as the vehicle routing problem (VRP). In system optimization, an appropriate intervention link is identified, and minimal changes are introduced to improve the system. A clustering method is utilized for this intervention to achieve efficiency. We found that the system change and the system optimization models led to a decrease of 0.26% and 0.66%, respectively, over the baseline cost.