Optimal Route Selection in Fleet Management Using Predictive Air Quality Data via a Smart Modular IoT-Enabled Sensing Platform

In the era of connected vehicles, fleet management is evolving beyond traditional logistics optimization, addressing challenges like insufficient real-time tracking and the perishability of goods due to poor air quality conditions. This paper introduces a pioneering smart connected-vehicle-based Internet of Things (IoT) system selecting future optimum routes essential for fleet safety and operational efficiency based on predictive air quality data. The system features an advanced modular sensing package with integrated battery management system, continuously monitoring volatile organic compounds (VOCs), carbon monoxide (CO), carbon dioxide (CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>), relative humidity, temperature, pressure along with vehicle motion and real-time GPS location data. Data are then transmitted to a centralized server, where different predictive algorithms are deployed to predict future air pollutant concentrations. These predictions enable proactive route optimization, allowing fleet managers to navigate based on lowest predicted air quality indices (AQI), ensuring cleaner and safer routes by leveraging real-time environmental data and predictive algorithms. The system has been tested across different environments, validating its sensing, IoTenabled and predictive abilities. Results demonstrate its efficacy in accurate predictions, achieving RMSE of 1.4, 8.1 and 2.5 ppm and mean absolute errors (MAE) of 1.0,5.2 and 1.6 ppm for CO, CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> and VOCs respectively. Additionally, the artificial neural network (ANN) model achieved an impressive R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.92 in predicting CO.