Yoosin Kim, Journal of Internet Computing and Services, Vol. 26, No. 1, pp. 55-63, Feb. 2025
10.7472/jksii.2025.26.1.55, Full Text:  HTML
Keywords: Smoking Detection, Rent-a-car, IoT, Air-Quality Sensor, Machine Learning

Abstract

As users shift from owning vehicles to using them, the rental car market, including shared vehicles, is experiencing continuous growth. However, complaints are on the rise as more users report discomfort from indoor smoking in shared cars. To address this issue, this study developed a system capable of detecting smoking in real time by IoT based air quality sensors within vehicles. The system was tested on 30 vehicles from domestic car-sharing company A over three months, detecting a total of 1,369 smoking incidents, with 50% involving e-cigarettes. Smoking rates were higher on weekdays than on weekends, and multiple smoking occurrences were observed in vehicles operating during early morning hours 2:00 am. As a result, this IoT sensor-based smoking detection system allows car-sharing and rental companies to detect smoking in real-time. Consequently, the system can encourage non-smoking behavior, help maintain cleanliness inside shared vehicles, and prevent potential complaints from non-smoking customers about unclean conditions due to smoking.

Statistics
Cite this article