Feedback Control for Metro Lines With a Junction

Schanzenbacher, Florian; Farhi, Nadir; Leurent, Fabien; Gabriel, Gérard

doi:10.1109/tits.2020.2974342

Cited by 8 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, the value of m is defined such that 1 ≤ m ≤ m C = 68 . The description of the phases and the characteristic points of the fundamental diagram can be found in Schanzenbächer et al ( 24 ) and Schanzenbächer ( 25 ).…”

Section: Review Of Analytical Results On a Metro Line With A Junctionmentioning

confidence: 99%

“…Its value is included in an interval I ( m ) , such that ∀ m * < m ≤ m C and ∀ Δ m ∈ I ( m ) = [ Δ m min * false( m false) , Δ m max * false( m false) ] , f ( m , Δ m ) = f max . The equations giving the interval I for the capacity phase can also be found in Schanzenbächer et al ( 24 ) and Schanzenbächer ( 25 ).…”

Section: Review Of Analytical Results On a Metro Line With A Junctionmentioning

confidence: 99%

“…The equations giving the interval I for the capacity phase can also be found in Schanzenba¨cher et al (24) and Schanzenba¨cher (25).…”

Section: N Umentioning

confidence: 99%

See 2 more Smart Citations

Impact of a First-In–First-Out Rule on the Merge of a Metro Line with a Junction

Farrando

Farhi

Christoforou

et al. 2023

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

On a metro line with a junction, the merge of the line is a key point for the operation. This paper proposes a discrete event traffic model for a metro line with a junction. It is based on the line’s existing signaling system and determines trains’ departure times at all line nodes. Furthermore, the model estimates trains’ average headway and frequency. We apply our model to Paris metro line 13, which is currently not operated with a FIFO (first-in–first-out) rule on the merge. Thus, we seek to evaluate the effect of a FIFO rule on the nominal frequency. Compared to the current line operation, we also test the FIFO rule as a control strategy for daily disturbances. At the steady state, we show that the train frequency is maximized regardless of the distribution of the trains on the three parts of the line (the central part and two branches). Then, we study the train frequency in two disturbed situations that often occur on a metro line. We show that the FIFO rule reduces the impact of these disturbances in time and intensity. This research is motivated by the Paris metro operator, as the considered line is expected to be fully automated in the coming years. The operator wants to better understand the potential benefits of new operating rules and how operating protocols can improve the throughput and reliability of the system. This work is the first step in seeing frequency evolution on a line with a junction.

show abstract

Section: Review Of Analytical Results On a Metro Line With A Junctionmentioning

confidence: 99%

Section: Review Of Analytical Results On a Metro Line With A Junctionmentioning

confidence: 99%

See 1 more Smart Citation

Impact of a First-In–First-Out Rule on the Merge of a Metro Line with a Junction

Farrando

Farhi

Christoforou

et al. 2023

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

show abstract

“…Specifically, the train dynamics are represented by a discrete event system that captures various traffic constraints such as dwell and safe-separation time constraints [16][17][18]. Additional research has extended this approach to incorporate passenger demand [19][20][21][22][23][24], model traffic on mass transit lines with junctions [25][26][27], and simulate mass transit lines with skip-stop policies [28,29]. In our work, we extend the existing traffic model by introducing the control of passenger inflows to platforms.…”

Section: Overview Of the Proposed Approachmentioning

confidence: 99%

Deep Reinforcement Q-Learning for Intelligent Traffic Control in Mass Transit

Khozam,

Farhi

2023

Sustainability

View full text Add to dashboard Cite

Traffic control in mass transit consists of the regulation of both vehicle dynamics and passenger flows. While most of the existing approaches focus on the optimization of vehicle dwell time, vehicle time headway, and passenger stocks, we propose in this article an approach which also includes the optimization of the passenger inflows to the platforms. We developed in this work a deep reinforcement Q-learning model for the traffic control in a mass transit line. We first propose a new mathematical traffic model for the train and passengers dynamics. The model combines a discrete-event description of the vehicle dynamics, with a macroscopic model for the passenger flows. We use this new model as the environment of the traffic in mass transit for the reinforcement learning optimization. For this aim, we defined, under the new traffic model, the state variables as well as the control ones, including in particular the number of running vehicles, the vehicle dwell times at stations, and the passenger inflow to platforms. Second, we present our new deep Q-network (DQN) model for the reinforcement learning (RL) with the state representation, action space, and reward function definitions. We also provide the neural network architecture as well as the main hyper-parameters. Finally, we give an evaluation of the model under multiple scenarios. We show in particular the efficiency of the control of the passenger inflows into the platforms.

show abstract

“…For the case of a metro line with a junction, the fundamental diagram gives the relation between the macroscopic variables train number, the difference between the number of trains on the branches, asymptotic average train time-headway and the parameters of the line (train run, dwell and safe separation times). It is new and has first been presented by Schanzenbächer et al [33,39].…”

Section: Contributionsmentioning

confidence: 99%