A Static Traffic Assignment Model Combined with an Artificial Neural Network Delay Model

Development of Delay Models with a Deep Learning Approach For Adaptive Signalized Intersections

“…Yapay sinir ağları (YSA), insan beynindeki biyolojik nöronların çalıştığı mekanizmalardan ilham alan ve insan beyninin öğrenme fonksiyonunu örnekler yardımı ile gerçekleştiren bilgisayar sistemleridir (Ding, 2007;Öztemel, 2006). YSA, verilerden elde ettikleri bilgiler ile insan beyninin fonksiyonel özelliklerine benzer bir şekilde öğrenme, ilişkilendirme, sınıflandırma, genelleme, özellik belirleme ve optimizasyon yapabilmektedir (Öztemel, 2006).…”

Improving Analytical Travel Time Estimation for Transportation Planning Models

“…Zhao and Ding's work shed some light on the way to overcome the above constraints (Zhao and Ding 2006;Ding 2007;Ding et al 2009). However, their models did not address link delay estimation separately from intersection delays.…”

“…The link travel time, t link (x), is estimated using the modified BPR equation with consideration to cycle length and link width, as follows: (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) where L = link distance (meters); The intersection delay is given below: In the model, street width is used instead of number of lanes; this is due to driving behavior in Tehran, where the number of lanes does not necessarily dictate the number of cars being accommodated across the width of a street (Aashtiani 1999). The significance of this model is that it provides a simple method to approximate cycle length and red time as follows: Link delay and intersection delay are separately expressed as follows: Ding (2007) and Ding et al (2009) proposed an artificial neural network (ANN) model to predict intersection delay based on traffic volumes from all movements of intersection. In their study, the authors assumed that signal timing plans are optimized in a future year and created simulated traffic data from TRANSYT-7F for which signal timing plans were optimized.…”