Although there is a plethora of methods to acquire traffic data, the ability to accomplish it at large-scale, low-cost, low-maintenance, high accuracy and on continuous basis, is far from achieved. Despite the difficulties, the conventional wisdom in traffic management has been that one should acquire as much traffic data as possible for effective automated control. In this paper, we investigate this assumption to understand how much data collection can be avoided without adversely affecting traffic monitoring objectives. We empirically find that a significant proportion of traffic segments can be predictable and furthermore, they can be identified with substantially less data than may be blindly collected today. This lays the basis for collecting traffic data using people as sensors. We explore what kind of information may be sought from citizens and present a method to use it to measure the scale of traffic. This is particularly advantageous in situations of emerging countries where traffic complexity and data quality needs make traditional data collection methods insufficient and costly, while people are already sharing traffic information.
[1]
Demetrios Zeinalipour-Yazti,et al.
Ieee Icdm 2010 Contest Tomtom Traffic Prediction for Intelligent Gps Navigation
,
2022
.
[2]
Biplav Srivastava.
A New Look at the Traffic Management Problem and Where to Start
,
2011
.
[3]
Douglas W. Hubbard,et al.
How to Measure Anything: Finding the Value of "Intangibles" in Business
,
2007
.
[4]
Mashrur Chowdhury,et al.
Fundamentals of Intelligent Transportation Systems Planning
,
2003
.
[5]
E. Huang,et al.
Real-time multi-sensor multi-source network data fusion using dynamic traffic assignment models
,
2009,
2009 12th International IEEE Conference on Intelligent Transportation Systems.
[6]
F. W. Kellaway,et al.
Advanced Engineering Mathematics
,
1969,
The Mathematical Gazette.
[7]
Yilin Zhao,et al.
Mobile phone location determination and its impact on intelligent transportation systems
,
2000,
IEEE Trans. Intell. Transp. Syst..