A Stackelberg approach to transportation infrastructure planning: Integrating User Decision-making

• Published in: Computers & operations research Vol. 184; p. 107211
• Main Authors: Cepeda-Valero, Oscar-Mauricio, Torres-Delgado, Jose-Fidel, González, Andres D.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.12.2025
• Subjects: Centrality measures; Metro networks; Stackelberg model; Urban transport network design; Metro networks; Urban transport network design; Stackelberg model; Centrality measures
• ISSN: 0305-0548
• DOI: 10.1016/j.cor.2025.107211

The design of transportation networks is an intriguing and complex subject, primarily driven by the high costs and existing deficiencies in transit systems. Effective network design necessitates the consideration of multiple factors, including the rapid growth of urban centers, the demand for new routes, and the evolving preferences of citizens. A poorly designed network can result in significant drawbacks, such as delays, congestion, and escalated air pollution levels. Consequently, addressing this challenge requires the implementation of targeted modifications in the transport infrastructure, such as constructing new streets, expanding existing roads, enhancing capacity, and improving coverage. The paper introduces a Stackelberg model that addresses the planning of transport infrastructure, focusing on incorporate the user decisions. The model functions on two levels: the upper level captures the decision-making process of users, influenced by factors like time savings and transport system accessibility. Conversely, the lower level relates to the planner’s responsibility in designing and configuring the transport network to align with user demands and preferences while minimizing costs. Four types of transport networks were evaluated based on criteria such as used capacity, network growth, system cost, and unmet user needs. The evaluation revealed that mesh networks perform better due to their ability to distribute user flow across multiple lines, reducing dependency on a few edges. Additionally, the versatility of the procedure is demonstrated through its implementation in the Lausanne metro network. [Display omitted] •The Stackelberg model represents the trade-off between users and service providers.•Optimizes transport networks using real-world data from OpenStreetMap with Gurobi.•Uses centrality measures to assess the performance of transport infrastructure.