Advances in Genetic Algorithm Optimization of Traffic Signals

Kesur, Khewal Bhupendra

doi:10.1061/(asce)0733-947x(2009)135:4(160)

Cited by 28 publications

(29 citation statements)

References 86 publications

(197 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For all critical movements during a cycle, the summation of their effective green time plus lost time (Lost) is equal to the cycle length. In this study, a generic algorithm (Kesur, 2009) is used to solve this optimization problem with complicated function structures.…”

Section: Intersection Levelmentioning

confidence: 99%

Investigating Emission Reduction Benefit From Intersection Signal Optimization

Zhang

Zietsman

2012

Journal of Intelligent Transportation Systems

View full text Add to dashboard Cite

Most existing intersection signals are timed based on delay minimization. However, minimizing delay does not necessarily lead to the minimization of emissions at an intersection. No study has focused on the difference or the trade-off between delay based and emissions based signal optimization. Delay-based optimization typically uses macroscopic flow conditions such as traffic demands, saturation flow rates, and average delay. However, the latest emission model, MOVES (Motor Vehicle Emission Simulation), requires second-by-second individual vehicle speed profiles, which makes the model difficult to formulate directly in an emission-based signal optimization problem. This study first develops a methodology to derive vehicle profiles given macroscopic inputs so that MOVES can be applied to estimating emissions. Then an optimization methodology of signal timing is developed and solved with a genetic algorithm. The objective function of the optimization problem considers both delay and emissions, with the signal timing elements being the decision variables. Through a case study, the air quality benefit by reducing vehicle emissions through intersection signal control is demonstrated, and thetrade-off between operational and emission performance measures is investigated. Furthermore, the air quality benefit from intersection signal control is discussed under different scenarios of cycle lengths, percentages of turning vehicles, and traffic demands on major/minor roads.

show abstract

Section: Intersection Levelmentioning

confidence: 99%

Investigating Emission Reduction Benefit From Intersection Signal Optimization

Zhang

Zietsman

2012

Journal of Intelligent Transportation Systems

View full text Add to dashboard Cite

show abstract

“…The encoding scheme developed by the author for optimization of signal timings under a common cycle length is extended to accommodate mixed cycle lengths. The encoding scheme is a combination of real and binary encodings.…”

Section: Methodsmentioning

confidence: 99%

“…Modeling logic and parameters are based on findings from the literature on driver behavior and vehicle characteristics. MSTRANS has been found to produce results comparable with CORSIM , a widely used commercial stochastic microscopic traffic simulator . A thorough account of the functional details of the model, a review of the literature justifying the model logic, and a complete code listing are given by Kesur .…”

Section: Methodsmentioning

confidence: 99%

Optimization of mixed cycle length traffic signals

Kesur

2012

J of Advced Transportation

Self Cite

View full text Add to dashboard Cite

SUMMARY Mixed cycle length operation has been recommended for networks where individual intersections process considerably different traffic volumes. The signals to operate at lower or higher cycle lengths are determined heuristically. This paper demonstrates that the use of mixed cycle lengths as given by the heuristic is inferior to operation under a common cycle length. This contradicts findings in earlier studies, and the difference in conclusion is due to the use of updated optimization methodology. A procedure for incorporating the allocation of mixed cycle lengths into the global optimization of all signal timing variables by a genetic algorithm is proposed. The mixed cycle length timing plans obtained from this procedure are an improvement over those determined heuristically. Mixed cycle length operation is found to be of a more limited application than indicated in previous studies. Copyright © 2012 John Wiley & Sons, Ltd.

show abstract

“…BLX-a crossover operator has been successfully used in benchmark studies (Cervone et al 2000;Zhao et al 2007) and in real-life applications such as traffic signals (Kesur 2009), image processing (Cordón et al 2006;Santamaría et al 2009) or financial predictions (Delgado et al 2006). In the following, we briefly describe the BLX-a crossover operator.…”

Section: Blx-a Crossover Operatormentioning

confidence: 99%

“…In particular, a specific version for real-coded optimization, real CHC, was proposed by its authors using the crossover operator BLX-a , and it has been successfully used in real-life applications (e.g. Delgado et al 2006;Kesur 2009;Santamaría et al 2009). However, there are no theoretical studies to describe CHC behavior or to predict its performance when applied to optimize real-coded functions.…”

Section: Introductionmentioning

confidence: 99%

Modeling dynamics of a real-coded CHC algorithm in terms of dynamical probability distributions

2011

View full text Add to dashboard Cite

Some theoretical models have been proposed in the literature to predict dynamics of real-coded evolutionary algorithms. These models are often applied to study very simplified algorithms, simple real-coded functions or sometimes these make difficult to obtain quantitative measures related to algorithm performance. This paper, trying to reduce these simplifications to obtain a more useful model, proposes a model that describes the behavior of a slightly simplified version of the popular real-coded CHC in multi-peaked landscape functions. Our approach is based on predicting the shape of the search pattern by modeling the dynamics of clusters, which are formed by individuals of the population. This is performed in terms of dynamical probability distributions as a basis to estimate its averaged behavior. Within reasonable time, numerical experiments show that is possible to achieve accurate quantitative predictions in functions of up to 5D about performance measures such as average fitness, the best fitness reached or number of fitness function evaluations.

show abstract

Advances in Genetic Algorithm Optimization of Traffic Signals

Cited by 28 publications

References 86 publications

Investigating Emission Reduction Benefit From Intersection Signal Optimization

Investigating Emission Reduction Benefit From Intersection Signal Optimization

Optimization of mixed cycle length traffic signals

Modeling dynamics of a real-coded CHC algorithm in terms of dynamical probability distributions

Contact Info

Product

Resources

About