Akademik Veri Yönetim Sistemi
Arabian Journal for Science and Engineering, 2025 (SCI-Expanded, Scopus)
This paper addresses a real-world urban transportation network design problem considering cost efficiency, environmental impact, and citizen satisfaction. To realistically design this network, we introduce two novel variants of the rich vehicle routing problem that consider both electric and fossil fuel vehicles: (i) the capacitated, green, heterogeneous fleet, and time-dependent vehicle routing problem with simultaneous pickup and delivery (CGHTD-VRPSPD) for direct/indirect carbon emissions and fuel/energy consumption, and (ii) an extended version of CGHTD-VRPSPD that incorporates speed-dependent and gear selection for (CGHTD-VRPSPD-sdgs). Both problems are formulated as mixed-integer linear programming and mixed-integer nonlinear programming models, respectively. For the large-sized problems, we propose protected simulated annealing (P-SA), which enhances the standard SA algorithm with a mechanism that protects the best solution. The proposed models and algorithms are applied to real-world scenarios involving a shift-worker transportation system of a furniture factory in Kayseri, Türkiye. In the large-sized scenario of the CGHTD-VRPSPD-sdgs, P-SA achieves lower costs with 74% and 75% shorter convergence times than SA and variable neighborhood search, and shows a competitive performance against the genetic algorithm with a 12% improvement in convergence time. The results show that the proposed methodology provides practical managerial insights to support decision-makers in sustainable urban transportation systems.