Time Series Analysis and Forecasting of Vehicular Network Communication Severity: A Comparative Study of ARIMA and SARIMA Models

The research evaluates vehicular network communication severity using two well-known statistical models namely ARIMA (AutoRegressive Integrated Moving Average) and SARIMA (Seasonal ARIMA) for time series analysis and forecasting. The Bologna and Erlangen simulation datasets provide vehicular communication data to analyze especially packet transmission stops and communication interruption scales. The objective of this study investigates time-based dependencies and future severity prediction for transport networks to enhance reliability and performance of Vehicular Ad-Hoc Networks (VANETs). The data undergoes a complete exploratory data analysis (EDA) to reveal hidden patterns together with data trends and unusual data points. The results indicate that ARIMA and SARIMA models deliver accurate predictions yet SARIMA establishes a better ability to understand seasonal dynamics present in severity data sets. This investigation offers essential knowledge about time series models in VANET applications while presenting recommendations for vehicle communication system enhancement.