Simsir, MisraYildirim, U. MahirSen, Doruk2026-04-042026-04-0420241300-70092147-5881https://doi.org/10.5505/pajes.2024.76299https://search.trdizin.gov.tr/tr/yayin/detay/1297694https://hdl.handle.net/11411/10735In a world with limited resources, it is crucial for individuals to utilise shared systems and develop strategies to optimise their usage. To cope with this, 'servicizing' has emerged as a rapidly growing promising solution, especially in car-sharing systems. These systems can be split into two: station-based and free-floating. The latter introduces more flexibility to the customers as free-floating systems allow users to pick up and drop off vehicles anywhere within predetermined operational zones. This flexibility may come with an additional cost by bringing a potential imbalance between demand and supply. This imbalance can harm the company's profitability and customer satisfaction. In this study, the imbalance problem of the system of free-floating car sharing is considered. A mixed integer linear programming model is developed and tested with real data for free floating car sharing systems to solve this problem. The proposed system consists of four modules: clustering, forecasting, optimization model, and relocation strategy. According to the results, it is observed that the system is more balanced with satisfying 9% more demand and more profitable with earning 6% more. The study was conducted on a car-sharing company that is based in Istanbul, but the results can be applied to any free-floating car-sharing system. This ensures customer satisfaction by meeting demand and balancing the system.eninfo:eu-repo/semantics/openAccessCar-SharingFree-FloatingUser-Based RelocationMachine LearningArticle10.5505/pajes.2024.7629910.5505/pajes.2024.762999437934129769430Q3WOS:001381269300010