Optimization Models for Smart Rail Transport in Sustainable Logistics Systems: A Case Study of Semi-Arid Smart Cities in Southeastern Chihuahua (Parral Region)

Abstract

This paper proposes a comprehensive and integrated smart rail transport optimization framework specifically designed for semi-arid logistics environments, with a focused application in the Parral region of southeastern Chihuahua, Mexico. The proposed approach addresses the structural inefficiencies and environmental constraints characteristic of resource-extraction and low-density territories by combining advanced optimization theory with emerging smart city technologies. At the core of the framework lies a multi-objective mathematical model that simultaneously minimizes transit time (T), carbon emissions (E), and operational costs (C). These objectives are balanced through adaptive, region-specific weighting parameters that reflect the economic priorities, environmental regulations, and infrastructural limitations of semi-arid regions. The model incorporates network constraints, stochastic demand behavior, rail capacity limitations, and intermodal transfer conditions, ensuring a realistic and scalable representation of freight transport dynamics. To enhance real-time decision-making and system responsiveness, the framework is augmented with an Internet of Things (IoT) architecture composed of distributed sensors deployed across rail infrastructure, cargo units, and transfer nodes. These sensors enable continuous monitoring of variables such as temperature, load conditions, vibration, and transit status. In parallel, machine learning modules are integrated to perform predictive analytics, including demand forecasting, anomaly detection, and dynamic route optimization under uncertain conditions. The system is further embedded within a smart city control layer, allowing centralized coordination, data fusion, and adaptive policy enforcement. A simulation-based evaluation was conducted using regionally calibrated datasets reflecting the logistical patterns of southeastern Chihuahua. The results indicate that the proposed framework achieves a reduction of approximately 30–35% in CO₂ emissions, driven by modal shift and improved energy efficiency of rail systems. Additionally, logistic flow efficiency improves by nearly 25%, as measured by reduced congestion, enhanced scheduling, and optimized routing. Operational costs exhibit a decrease ranging from 20% to 28% when compared to traditional highway-based freight transport, highlighting the economic viability of the approach. These findings underscore the strategic potential of intelligent rail transport systems as a sustainable and scalable alternative for freight mobility in semi-arid regions. Beyond immediate performance improvements, the proposed framework contributes to the broader transition toward smart, resilient, and low-carbon logistics ecosystems, particularly in regions where economic activity depends heavily on mining and agriculture.