New Routing Algorithm Weighs Traffic and Air Pollution for Nairobi Drivers

Researchers at Vellore Institute of Technology in India have developed a route-planning framework for Nairobi, Kenya, that considers not only traffic congestion but also air pollution when calculating the best path between two points. The study, published as an original research article in Frontiers in Environmental Science, addresses a gap the authors identify in most existing traffic-management approaches, which typically ignore air quality when determining routes.

The framework begins by extracting the city’s road network from Google Maps and converting it into a weighted graph using image-processing techniques. The authors chose Google Maps over OpenStreetMap because it offers integrated environmental data support, real-time traffic updates and improved spatial accuracy for dynamic route optimization. However, they note that the framework remains fully compatible with open-source basemaps such as OpenStreetMap, which could be used in future deployments without changing the underlying method.

Air-quality inputs come from two sources. Monitoring stations run by the National Environment Management Authority and the Kenya Meteorological Department provide direct measurements, while locations without sensor coverage are supplemented by a validated Urban Emission Zone model. Traffic conditions are represented through calibrated traffic simulations informed by GPS-derived traffic characteristics and available Nairobi traffic benchmarks, rather than assumed uniform conditions across the network.

To categorize congestion levels across the road network, the method applies Fuzzy C-Means clustering, a technique that assigns road segments to congestion categories based on their traffic characteristics. Both air pollution levels and traffic density then feed into the calculation of the graph’s edge weights, so that each possible road segment carries a cost reflecting how crowded and how polluted it is. The routing problem is subsequently formulated as a linear programming optimization problem, which the authors use to identify the most effective and least polluting path through the network.

The framework was evaluated on a case study of a major transportation corridor connecting Jomo Kenyatta International Airport with Nairobi’s Central Business District. Nairobi is described by the authors as one of Africa’s most heavily congested cities, making it a representative test case for the rapidly growing urban centers of the continent, where finding the most direct and unencumbered path is described as essential. The authors present the results as demonstrating the framework’s potential applicability for sustainable urban mobility and environmentally informed route planning.

The work sits at the intersection of several established techniques: remote sensing and image processing for road extraction, fuzzy clustering for congestion classification, emissions modeling for pollution estimation where sensors are sparse, and mathematical optimization for path selection. By combining these into a single pipeline, the study offers a template that other African cities with similar data constraints could adapt, since the method does not depend exclusively on proprietary map data.

As with any modeling-based study, the reliability of the routes produced depends on the quality of the underlying inputs, including the accuracy of the emission model in areas without direct monitoring and the fidelity of the calibrated traffic simulations. The authors report the framework’s evaluation on a single corridor, and broader deployment across a full urban network would be the natural next step for testing the approach at scale.

Subject of Research: A traffic- and pollution-aware route optimization framework for Nairobi, Kenya

Article Title: An optimal path-finding algorithm for African cities: a case study of Nairobi, Kenya, considering traffic congestion and air pollution

Article References: S, P., & A, P. (2026). An optimal path-finding algorithm for African cities: a case study of Nairobi, Kenya, considering traffic congestion and air pollution. Frontiers in Environmental Science, 14, Article 1793670. https://doi.org/10.3389/fenvs.2026.1793670

Image Credits: AI Generated

DOI: 10.3389/fenvs.2026.1793670

Keywords: Nairobi, traffic congestion, air pollution, route optimization, fuzzy c-means clustering, linear programming, urban mobility, smart city, Kenya, weighted graph, optimal, path-finding