Activity Number:
|
664
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 4, 2016 : 8:30 AM to 10:30 AM
|
Sponsor:
|
IMS
|
Abstract #319412
|
|
Title:
|
Dynamic Weighted Average Approach for Predicting Street Parking Availability
|
Author(s):
|
Yi Hua* and Ouri E. Wolfson and Xudong Lin and Jie Yang
|
Companies:
|
University of Illinois at Chicago and University of Illinois at Chicago and South China Agricultural University and University of Illinois at Chicago
|
Keywords:
|
Transportation ;
Parking Availability Prediction ;
Cross-Validation ;
Root Mean Squared Error ;
Historical Mean ;
Kalman Filter
|
Abstract:
|
Real-time estimation and prediction of street parking availability have become a challenging problem for both transportation authorities and drivers, especially in big cities. It has been a concern of the government because of its prospect in terms of reducing traffic congestion and pollution. It also provides real-time parking availability information to save drivers' time significantly. One application-based solution called PhonePark utilizes the GPS and accelerometer sensor signals in smart phones to detect parking/deparking activities. Here we propose a dynamic weighted average method for estimating the current parking availability and predicting the status in the near future. It incorporates the application-based signal data and the historical profile data. The weights of historical profile data and the signal data chosen by cross validation vary across time and parking blocks. Using the real parking availability data collected in San Francisco, we show that our method is superior to the existing methods in the literature including Weighted Average and Kalman Filter methods in terms of accuracy and feasibility.
|
Authors who are presenting talks have a * after their name.