Adar, Uğur GüvenÇayli, OsmanYilmaz, Atınç2025-03-092025-03-092025978-303181454-91865-0929https://doi.org/10.1007/978-3-031-81455-6_2https://hdl.handle.net/20.500.12662/49881st International Conference on Optimization and Data Science in Industrial Engineering, ODSIE 2023 -- 16 November 2023 through 17 November 2023 -- Istanbul -- 326219Traffic congestion in large cities negatively affects daily life and is a significant concern. In crowded cities like Istanbul, not only waiting in traffic but also the availability of parking spaces at the destination is a critical issue. Predicting the occupancy of parking lots in advance can supply a significant advantage for drivers. In this study, traffic congestion in Istanbul is attempted to be estimated using artificial intelligence techniques such as K-NN, Decision Trees, and Artificial Neural Networks, with 2022 vehicle traffic data from the Istanbul Metropolitan Municipality (IBB) Open Data Portal. As a result of the tests, the decision tree method provided the best results in the dataset, estimating traffic congestion with an R2 value of 0.8469. The occupancy status of parking lots in areas close to the traffic congestion estimation points was determined using the İspark Parking Detailed Information dataset. Based on the obtained information, it is aimed to predict the occupancy status of parking lots according to traffic congestion without the need for live data using the Dijkstra algorithm in the future. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.eninfo:eu-repo/semantics/closedAccessartificial neural networksdecision treesk-nnparking lottraffic congestion estimationConference Object10.1007/978-3-031-81455-6_22-s2.0-8521850628731Q3192204