The VTI National Transport Library Catalogue

Normal view MARC view

Regional land use and transportation planning with a genetic algorithm Balling, Richard ; Lowry, Michael ; Saito, Mitsuru

By: Contributor(s): Publication details: Transportation Research Record, 2003Description: nr 1831, s. 210-8Subject(s): Bibl.nr: VTI P8169:2003 Ref ; VTI P8167Location: Abstract: A new approach to regional land use and transportation planning, which uses a genetic algorithm as an integrated optimization tool, is presented. The approach is illustrated by applying it to the Wasatch Front Metropolitan Region, which consists of four counties in the state of Utah. This genetic algorithm-based approach was applied earlier to the twin cities of Provo and Orem in Utah, but here it is adapted to regional planning. Three issues make regional planning particularly difficult: (a) individual cities have significant planning autonomy, (b) the search space of possible plans is immense, and (c) preferences between competing objectives vary among stakeholders. The approach used here addresses the first issue by the way the problem is formulated. The second issue is addressed with a genetic algorithm. Such algorithms are particularly well suited to problems with large search spaces. The third issue is addressed by using a multiobjective fitness function in the genetic algorithm. It was found that a genetic algorithm could produce a set of nondominated future land use scenarios and street plans for a region, from which regional planners can make a selection. Execution of the algorithm to produce 100 plans per generation for 100 generations took about 4 days with a high-end personal computer. Interesting trends for reducing change and traffic congestion were discovered.
Item type: Reports, conferences, monographs
Holdings
Current library Call number Status Date due Barcode
Statens väg- och transportforskningsinstitut Available

A new approach to regional land use and transportation planning, which uses a genetic algorithm as an integrated optimization tool, is presented. The approach is illustrated by applying it to the Wasatch Front Metropolitan Region, which consists of four counties in the state of Utah. This genetic algorithm-based approach was applied earlier to the twin cities of Provo and Orem in Utah, but here it is adapted to regional planning. Three issues make regional planning particularly difficult: (a) individual cities have significant planning autonomy, (b) the search space of possible plans is immense, and (c) preferences between competing objectives vary among stakeholders. The approach used here addresses the first issue by the way the problem is formulated. The second issue is addressed with a genetic algorithm. Such algorithms are particularly well suited to problems with large search spaces. The third issue is addressed by using a multiobjective fitness function in the genetic algorithm. It was found that a genetic algorithm could produce a set of nondominated future land use scenarios and street plans for a region, from which regional planners can make a selection. Execution of the algorithm to produce 100 plans per generation for 100 generations took about 4 days with a high-end personal computer. Interesting trends for reducing change and traffic congestion were discovered.

Powered by Koha