Univariate time series prediction of air quality inside a public transportation bus using available software

Accurate predictions of the in-vehicle air contaminants is an integral task in the process of evaluating indoor air quality (IAQ) of a transportation micro-environment and developing control measures to keep the contaminants inside the vehicle cabin below recommended IAQ guidelines. Time series analysis has been adopted by researchers worldwide in forecasting the atmospheric pollutants on a short-term basis with acceptable results. One can choose to model the time series from a wide range of available software. Currently, there are several software that have in-built features to directly import a dataset and perform the time series modeling. The developed univariate carbon dioxide and carbon monoxide time series models were validated using a comprehensive set of IAQ operational performance measures and graphical representations with different software. The developed univariate carbon dioxide and CO time series models met all the required IAQ model acceptance criteria, with emphasis on the established extreme-end contaminant concentration evaluation procedure.