Multi-Type traffic sensor system design for multi-period network-wide speed-flow estimation with day-to-day uncertainties

Abstract

Owing to budget limitations, traffic sensors cannot be installed on each link of a road network in practice. Thus, optimizing the locations of a limited number of sensors is crucial for development of various intelligent transportation systems. This paper devises a multi-type traffic sensor location model with consideration of day-to-day flow variations by hourly period. Based on the devised model, network-wide link speeds and flows in multiple periods, even on those links without installed sensors, can be estimated. The link speed and flow estimator is formulated as a bi-level model, in which the upper-level model estimates multi-period origin–destination (OD) demands, and the lower-level model assigns the resultant OD demands onto the study network to obtain the hourly link flows and speeds. The case studies on both synthetic and real road networks demonstrate the cost-effectiveness of the multi-period multi-type sensor location scheme and provide valuable insights.

KEYWORDS:

• Sensor location
• speed-flow estimation
• multi-type sensors
• multiple periods

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Talent Fund of Beijing Jiaotong University [grant number 2023XKRC026]; National Natural Science Foundation of China [grant numbers 72301023 and 72288101];   Smart Traffic Fund from the Transport Department of the Hong Kong Special Administrative Region, China [grant number PSRI/06/2108/PR]; and Research Grants Council of the Hong Kong Special Administrative Region, China [grant numbers PolyU/25209221 and PolyU/15206322].