Exploring the Feasibility of Using GANs for Traffic Mitigation.
Reimagining Urban Planning for Tirana's Emerging City Center.
Author
Andia VLLAMASI
Affiliation
PhD IDAUP/ Polis University, Tirana, Albania
Abstract
Tirana and other rapid growing cities are experiencing fast urbanization, which has increased traffic congestion and caused major delays and disruptions in transportation systems. In order to overcome these obstacles, network modeling and transportation indices have emerged as crucial instruments for comprehending and reducing urban traffic problems. Predicting these indices, however, becomes essential for sustainable urban planning and efficient traffic management as cities become increasingly complex.
Along with other recent advances in deep learning, the introduction of Generative Adversarial Networks (GANs) and their adaptations for spatial data analysis have provided urban planners with powerful tools to construct hyper- realistic urban layouts. Presenting a methodology for using GANs to produce new suitable city layouts with an emphasis on traffic mitigation is the aim of this study, which also aims to explore and showcase the potential of AI, specifically GANs, in urban planning. This approach surpasses some of the traditional limitations in urban planning, particularly the ability to facilitate iterative upgrades and provide prompt performance feedback at the first stages of design. This study investigates how the generative capabilities of GANs could speed up the design process and enable urban planners to dynamically alter layouts in response to shifting constraints and objectives. In order to create sustainable, ideal urban landscapes, this approach seeks to assess how well GANs support data-driven decision- making. Urban planners will be able to precisely assess urban plans prior to implementation through analyzing the potential for providing traffic estimates in sequential time slots based on varying travel demands. The combination of GAN and traffic predictions will enable the generation of rapid scenarios to explore multiple design alternatives and their traffic impact. These developments offer a revolutionary perspective to contemporary urban planning by facilitating the investigation of efficient city plans that not only reduce traffic jams but also encourage sustainable growth.
