The application of deep dual q network in urban rail transit network planning

Urban rail transit network planning is a complex issue that involves multiple considerations, including traffic flow, passenger demand, geographical environment, economic costs, environmental protection. To address the pressing issues of inadequate station coverage, protracted planning timelines, and suboptimal passenger satisfaction in conventional urban rail transit network planning methodologies, the paper delves into the application of a Deep Dual Q Network in urban rail transit network planning. By initially analyzing the distribution density of Points of Interest (POI), the study employs the K-means clustering algorithm to meticulously select optimal locations for urban rail transit stations. Subsequently, a tailored urban rail transit network planning model is constructed, incorporating the site selection outcomes and delineating its pertinent constraints. Leveraging the prowess of a Deep dual Q Network, the model is efficiently solved, yielding an optimized urban rail transit network planning scheme. Experimental validation reveals remarkable outcomes, with a maximum station coverage rate of 91.9%, a minimized planning duration of 23.6 minutes, and a pinnacle passenger satisfaction rating of 98.1%, underscoring the practical efficacy and significance of the proposed approach.