|
September 16, 2025 - GTU Office of Press and Public Relations Academics from Gebze Technical University and Sabancı University develop innovative algorithms to address network problems in a wide range of areas, from logistics and traffic management to internet infrastructure. These algorithms aim to improve connection times within networks, reducing costs while increasing efficiency.
Asst. Prof. Burak Paç from the Department of Industrial Engineering at Gebze Technical University and Esra Koca (PhD) from Sabancı University have focused on an important topic known as the “edge improvement problem.” In its simplest terms, this problem seeks to optimize the nodes in a network for maximum efficiency. The new models they developed provide exact solutions for small systems while delivering fast and reliable results even in large and complex networks.
Benders Decomposition Algorithm Makes a Difference In particular, the developed Benders decomposition algorithm overcomes challenges encountered in solving medium- and large-scale problems. Thanks to this algorithm, processes that were previously very time-consuming or even impossible to solve can now be handled much more quickly and reliably.
The new approaches are expected to have significant impacts in several fields: Logistics: Helps companies plan routes more quickly, reducing costs and increasing delivery speed. Smart Cities: Optimizes traffic management by alleviating congestion in urban areas. Telecommunications: Accelerates data flow in internet infrastructure, minimizing data loss. Energy Management: Helps prevent potential disruptions in energy networks.
Strong Support for Future Infrastructure The work of Koca and Prof. Paç provides a powerful tool that can be directly applied in strategic sectors such as smart city planning, e-commerce logistics, energy management, and digital infrastructure.
This significant study has been published in one of the field’s most respected journals, the European Journal of Operational Research, under the title “Exploring the discrete and continuous edge improvement problems: Models and algorithms.” The article can be accessed via DOI: 10.1016/j.ejor.2024.12.051. |
