Journal
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
Volume 13, Issue 4, Pages 1727-1737Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TITS.2012.2203122
Keywords
Autoregressive integrated moving average (ARIMA); combined forecasting; exponential smoothing method; Holt's method; irregularly spaced time series data; neural network; short-term traffic speed forecasting
Categories
Funding
- Research Grants Council of HKSAR, China [HKU7176/07E]
- Hui Oi Chow Trust Fund [200902172003]
- University Research Committee of the University of Hong Kong [201011159026, 201111159056]
- Research Postgraduate Studentship from the University of Hong Kong
- Outstanding Researcher Award from the University of Hong Kong
Ask authors/readers for more resources
Recent growth in demand for proactive real-time transportation management systems has led to major advances in short-time traffic forecasting methods. Recent studies have introduced time series theory, neural networks, and genetic algorithms to short-term traffic forecasting to make forecasts more reliable, efficient, and accurate. However, most of these methods can only deal with data recorded at regular time intervals, which restricts the range of data collection tools to presence-type detectors or other equipment that generates regular data. The study reported here is an attempt to extend several existing time series forecasting methods to accommodate data recorded at irregular time intervals, which would allow transportation management systems to obtain predicted traffic speeds from intermittent data sources such as Global Positioning System (GPS). To improve forecasting performance, acceleration information was introduced, and information from segments adjacent to the current forecasting segment was adopted. The study tested several methods using GPS data from 480 Hong Kong taxis. The results show that the best performance in terms of mean absolute relative error is obtained by using a neural network model that aggregates speed information and acceleration information from the current forecasting segment and adjacent segments.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available