Population-based simulation optimization for urban mass rapid transit networks

Abstract

In this paper, we present a simulation-based headway optimization for urban mass rapid transit networks. The underlying discrete event simulation model contains several stochastic elements, including time-dependent demand and turning maneuver times as well as direction-dependent vehicle travel and passenger transfer times. Passenger creation is a Poisson process that uses hourly origin-destination-matrices based on anonymous mobile phone and infrared count data. The numbers of passengers on platforms and within vehicles are subject to capacity restrictions. As a microscopic element, passenger distribution along platforms and within vehicles is considered. The bi-objective problem, involving cost reduction and service level improvement, is transformed into a single-objective optimization problem by normalization and scalarization. Population-based evolutionary algorithms and different solution encoding variants are applied. Computational experience is gained from test instances based on real-world data (i.e., the Viennese subway network). A covariance matrix adaptation evolution strategy performs best in most cases, and a newly developed encoding helps accelerate the optimization process by producing better short-term results.

Multi-objective simulation optimization for complex urban mass rapid transit systems

“…3). This approach has been used previously as well (Schmaranzer et al 2018(Schmaranzer et al , 2019; for this study its use is limited to the first phase of our optimization scheme and the determination of the required minimum/maximum objective values (Sect. 4).…”

“…This section contains a brief description of the simulation model of the Viennese network as of 2016. Detailed model descriptions can be found in Schmaranzer et al (2016Schmaranzer et al ( , 2018Schmaranzer et al ( , 2019. The second paper contains an earlier version using data from 2014.…”

“…Our industrial partner provided count data from 2016, which we used to create new, updated versions of the origin-destination-matrices. For the precise procedure, see Schmaranzer et al (2019). Origin-destination-pairs represent the number of passengers who wish to travel, within a certain time frame (e.g., from 08:00 to 08:59), from one geographical location to another.…”

“…Visual goodness-of-fit tests (Q-Q-plots and density-histogram plots) conducted in R provided strong support for this choice. A more detailed analysis can be found in Schmaranzer et al (2019) and Schmaranzer et al (2018). Once a vehicle has reached one end of a line, it remains in the station for 0.41 min (± 0.02), accounting for about 25 s of dwelling time.…”

“…The base solution is the very first individual in the initial population, and the remaining ones are generated by sampling from a normal distribution with the currently employed headways as mean and 20% standard deviation. Schmaranzer et al (2019) study four different solution encoding variants (including continuous and discrete values) and find that the SO-CMA-ES achieved the best results, in less time, when using continuous factor encoding. This encoding uses factors instead of values, applied to the currently employed headways.…”