OPTIMIZATION OF STATION LOCATIONS AND TRACK ALIGNMENTS FOR RAIL TRANSIT LINES

Loading...
Thumbnail Image

Files

Lai_umd_0117E_13053.pdf (3.8 MB)
No. of downloads: 3410

Publication or External Link

Date

2012

Citation

DRUM DOI

Abstract

Designing urban rail transit systems is a complex problem, which involves the determination of station locations, track geometry, right-of-way type, and various other system characteristics. The existing studies overlook the complex interactions between railway alignments and station locations in a practical design process. This study proposes a comprehensive methodology that helps transit planners to concurrently optimize station locations and track alignments for an urban rail transit line. The modeling framework resolves the essential trade-off between an economically efficient system with low initial and operation cost and an effective system that provides convenient service for the public. The proposed method accounts for various geometric requirements and real-world design constraints for track alignment and stations plans. This method integrates a genetic algorithm (GA) for optimization with comprehensive evaluation of various important measures of effectiveness based on processing Geographical Information System (GIS) data.

The base model designs the track alignment through a sequence of preset stations. Detailed assumptions and formulations are presented for geometric requirements, design constraints, and evaluation criteria. Three extensions of the base model are proposed. The first extension explicitly incorporates vehicle dynamics in the design of track alignments, with the objective of better balancing the initial construction cost with the operation and user costs recurring throughout the system's life cycle. In the second extension, an integrated optimization model of rail transit station locations and track alignment is formulated for situations in which the locations of major stations are not preset. The concurrent optimization model searches through additional decision variables for station locations and station types, estimate rail transit demand, and incorporates demand and station cost in the evaluation framework. The third extension considers the existing road network when selecting sections of the alignment. Special algorithms are developed to allow the optimized alignment to take advantage of links in an existing network for construction cost reduction, and to account for disturbances of roadway traffic at highway/rail crossings. Numerical results show that these extensions have significantly enhanced the applicability of the proposed optimization methodology in concurrently selecting rail transit station locations and generating track alignment.

Notes

Rights