Prompt prediction of spillovers in urban street networks by using rolling time occupancy data

ABSTRACT

Prompt and accurate prediction of spillovers is essential in traffic control systems. Many studies on spillover focus on the identification methods, but little attention is paid to spillover prediction. In this paper, a new method to identify the imminent spillover is proposed. Firstly, we introduce the computation algorithm of the new index of rolling time occupancy (RTO). As a single RTO is strongly influenced by buses and other factors, the criteria for spillover prediction are divided into two parts: the threshold for abnormal occupancy and the minimum number of RTOs for spillover prediction. The occupancy threshold is determined by analyzing the relationships among the holding time within a horizon, effective vehicle lengths, and vehicle speeds. The second part is determined by quality control charts. Finally, misjudgment ratio and postponement time are used to evaluate the proposed method, and the results show that the new method provides the best compromise criterion.

KEYWORDS:

• Urban transportation system
• quality control
• spillover
• RTO

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Sheng Jin http://orcid.org/0000-0001-6110-0783

Additional information

Funding

This research is supported by the Zhejiang Provincial Natural Science Foundation [grant number LY17F030009]; the National Natural Science Foundation of China [grant numbers 61773337, 61773338 and 61304191]; the Key Research and Development Program of Zhejiang [grant number 2018C01007]; the MOE (Ministry of Education in China) Project of Humanities and Social Sciences [grant number 17YJCZH124] and the Hangzhou Science and Technology Development Plan Program [grant number 20160533B97].