Long-Term and Short-Term Traffic Forecasting Using Holt-Winters Method: A Comparability Approach With Comparable Data in Multiple Seasons

Long-Term and Short-Term Traffic Forecasting Using Holt-Winters Method: A Comparability Approach With Comparable Data in Multiple Seasons

Aditya R. Raikwar (Kashibai Navale College of Engineering, India), Rahul R. Sadawarte (Kashibai Navale College of Engineering, India), Rishikesh G. More (Kashibai Navale College of Engineering, India), Rutuja S. Gunjal (Kashibai Navale College of Engineering, India), Parikshit N. Mahalle (Kashibai Navale College of Engineering, India) and Poonam N. Railkar (Kashibai Navale College of Engineering, India)
Copyright: © 2018 |Pages: 15
DOI: 10.4018/978-1-5225-5643-5.ch077

Abstract

The need of faster life has caused the exponential growth in No. of vehicles on streets. The adverse effects include frequent traffic congestion, less time efficiency, unnecessary fuel consumption, pollution, accidents, etc. One of most important solution for resolving these problems is efficient transportation management system. Data science introduces different techniques and tools for overcoming these problems and to improve the data quality and forecasting inferences. The proposed long-term forecasting model can predict numerical values of effective attributes for a particular day on half-hourly basis, at least 24 hours prior to the time of prediction. The proposed forecasting model for short-term analysis will be having access to data as close as 30-minute difference from the time of prediction. Our proposed solution has integrated use of Holt-Winters (HW) method along with comparability schemes for seasonal approach.
Chapter Preview
Top

Traffic is a critical problem that inspired several researchers (Meier, & Lee, 2009; Tchepnda et al., 2009; Biswas et al., 2016; Erden et al., 2016; Mukherjee et al., 2016; Roy, 2017). Various techniques have been utilized in forecasting traffic Out of which primary focus is on trend, seasonality or combination of both in time series data. Seasonality is considered for forecasting in techniques such as Repeatability and Similarity of graphs (Hou et al., 2016), Block regression model (Pan et al., 2015), Time-Aware Multivariate Nearest Neighbor Regression (TaM-NNR) (Dell'Acqua et al., 2015). In (Hou et al., 2016), similar and repeatable nature of traffic flow has been utilized with help of statistic average values of basic series and deviation series whereas (Pan et al., 2015) includes block regression model which consist of seasonal differentiation. (Dell'Acqua et al., 2015) takes into account TaM-NNR which is an extension of multivariate method. It utilizes time and day (i.e. seasonality) for fetching neighbors. However, these methods focus principally on periodic behaviors, thereby affecting the predicted results, as they may also depend upon local mean values and trend in recent data. Trend and seasonality are considered in (Ahn et al., 2016; Tan et al, 2009). SVR with Bayesian classifier is used for estimating cliques of dependent roads. This increases the accuracy of forecasting results since mapping is considered, however selection of parameters is difficult and wrong selection affect the predicted values adversely. In (Tan et al, 2009), trend is forecasted by fitting curve of general trend on traffic level where as seasonality is predicated by simulated annealing-support vector regression machine (SASVR).

Purchase this chapter to continue reading all 15 pages >

Complete Chapter List

Search this Book:
Reset