Investigating Bayesian Optimization for rail network optimization

Abstract

Optimizing the operation of rail networks using simulations is an ongoing task where heuristic methods such as Genetic Algorithms have been applied. However, these simulations are often expensive to compute and consequently, because the optimization methods require many (typically >10 4 ) repeat simulations, the computational cost of optimization is dominated by them. This paper examines Bayesian Optimization and benchmarks it against the Genetic Algorithm method. By applying both methods to test-tasks seeking to maximize passenger satisfaction by optimum resource allocation, it is experimentally determined that a Bayesian Optimization implementation finds 'good' solutions in an order of magnitude fewer simulations than a Genetic Algorithm. Similar improvement for realworld problems will allow the predictive power of detailed simulation models to be used for a wider range of network optimization tasks. To the best of the authors' knowledge, this paper documents the first application of Bayesian Optimization within the field of rail network optimization.

Machine learning for CO2 conversion driven by dielectric barrier discharge plasma and Cs2TeCl6 photocatalysts

“…Therefore, for the basic BPANN model, two hyperparameters (hidden layer nodes and regularization term parameters) are optimized via three different methods: Genetic Algorithm (GA), 41 Particle Swarm Optimization (PSO) 42 and Bayesian Optimization. 43 The meanings of the two hyperparameters, their corresponding impacts on the BPANN model, and the principles and application descriptions of each optim-ization algorithm in the BPANN can be found in the ESI † in detail.…”

NOAH as an Innovative Tool for Modeling the Use of Suburban Railways

“…Canca et al, 2017 [1] Line planning O Yan et al, 2019 [2] MILP Hickish et al, 2020 [3] Network optimization GA Schlechte et al, 2011 [4] NH Weik et al, 2020 [5] Nodes (location, classification) NH Ahmed et al, 2020 [6] Station location GA Dong et al, 2020 [7] O Qiannan et al, 2022 [8] NH Binder et al, 2021 [9] Timetable creation and rescheduling SA Jamili and Pourseyed Aghaee, 2015 [10] SA Shang et al, 2018 [11] NH Wang et al, 2018 [12] MILP Jensen et al, 2016 [13] NH Yin et al, 2021 [14] O Zhang et al, 2018 [15] O Zhang et al, 2019a [16] MILP Zhang et al, 2019b [17] NH Zhang et al, 2022 [18] O Zhao et al, 2021 [19] MILP Whitbrook et al, 2018 [20] MS Zheng et al, 2014 [21] BBO Satoshi et al, 2011 [22] Operations (delay management, train routing) TS Sama et al, 2017 [23] SI Oneto et al, 2017 [24] O Xiao et al, 2018 [25] Demand side of the transport system NH Shen et al, 2016 [26] O Wu et al, 2013 [27] O Liu et al, 2019 [28] ANN Hansson et al, 2021 [30] Bus systems and networks NH Wei et al, 2021 [31] NH Ngo et al, 2021 [32] Ride-hailing services NH Behrends, 2017 [33] Nodes, coordination NH Stead et al, 2019 [34] NH Yang et al, 2020 [35] GA Lianhua and Xingfang, 2022 [36] SA Chen et al, 2020 [37] Autonomous Vehicles O Lopez and Farooq, 2020 [39] Smart mobility data-markets BC Xavier and Xavier, 2016 [44] Not related to transport systems O Martinelli and Teng, 1996…”

Modified Ant Colony Optimization as a Means for Evaluating the Variants of the City Railway Underground Section

“…Additionally, in [ 15 , 16 ], a genetic algorithm is used to optimize public transportation routes. There are also attempts to optimize entire transit networks [ 17 , 18 , 19 ]. In [ 20 ], the authors use a genetic algorithm to solve a network design problem in an urban area in a multi-criteria manner.…”