United States Department of Transportation
i
Genetic algorithm-based stochastic optimization for preempted signals at highway railroad grade crossing.
-
2013-09-01
[PDF-870.12 KB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Corporate Publisher:
-
Abstract:Decreasing the number of accidents at highway-railway grade crossings (HRGCs) is an important goal in the transportation field. The preemption of traffic signal operations at HRGCs is widely used to prevent accidents by clearing vehicles off the tracks before a train arrives. However, by interrupting normal traffic operations, preemption operations can contribute to congestion in highway traffic networks. This report presents a genetic algorithm (GA)-based stochastic optimization approach for preempted signals that is designed to minimize highway delays while improving safety. The first step of proposed method determines the preemption phase sequences that prevent the queue from backing on to the HRGC. The second step is to implement a GA-based algorithm to find the optimized signal phase lengths for reducing highway traffic delay. The GA-based Stochastic Optimization of Preempted Signals (GASOPS) model optimizes signal timing plans for both normal and preemption operations simultaneously, while current signal optimization models can optimize for only normal operations. Results show that the proposed approach is more efficient in signal optimization than traditional methods. This optimization approach reduces the delay by a maximum of 17% compared to optimal timing plans found using state-of-the-art methods. This model also improves safety because all queue lengths in GASOPS scenarios are 0, even when demand is doubled. This approach will be useful for designing and improving the preemption operations for signalized intersections near HRGCs.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
