Authors :
Dwijesh Shah; Paramsmit Sanghani; Mrugendrasinh Rahevar
Volume/Issue :
Volume 6 - 2021, Issue 12 - December
Google Scholar :
http://bitly.ws/gu88
Scribd :
https://bit.ly/3GUTyDs
Abstract :
Due to a significant increase in the number of
automobiles, traffic congestion has become a serious issue
in recent years. This paper discusses various techniques
for forecasting traffic flow to resolve the issue of traffic
congestion. To begin, we will demonstrate how time series
can be applied in this field. Second, we will attempt to describe which time-series models will be most beneficial in
resolving the most pressing issue. Following that, we'll
compare the results obtained using various methods using
accuracy parameters. Additionally, we observe that the
ARIMA time series forecasting method is incapable of producing appropriate results due to the seasonality observed
in the data. We discovered in this research paper that the
SARIMA time series forecasting method produces more
accurate results when forecasting traffic flow at 15-minute
and 30-minute intervals. Additionally, we discovered that
for short time intervals, i.e., one minute, FBProphet outperforms SARIMA.
Keywords :
Traffic time-series, ARIMA, SARIMA, Facebook Prophet, Traffic Prediction
Due to a significant increase in the number of
automobiles, traffic congestion has become a serious issue
in recent years. This paper discusses various techniques
for forecasting traffic flow to resolve the issue of traffic
congestion. To begin, we will demonstrate how time series
can be applied in this field. Second, we will attempt to describe which time-series models will be most beneficial in
resolving the most pressing issue. Following that, we'll
compare the results obtained using various methods using
accuracy parameters. Additionally, we observe that the
ARIMA time series forecasting method is incapable of producing appropriate results due to the seasonality observed
in the data. We discovered in this research paper that the
SARIMA time series forecasting method produces more
accurate results when forecasting traffic flow at 15-minute
and 30-minute intervals. Additionally, we discovered that
for short time intervals, i.e., one minute, FBProphet outperforms SARIMA.
Keywords :
Traffic time-series, ARIMA, SARIMA, Facebook Prophet, Traffic Prediction