Skripsi
PENERAPAN SMART TRANSPORTATION PADA SMART CITY UNTUK MENENTUKAN RUTE TERBAIK MENGGUNAKAN METODE RECURRENT NEURAL NETWORK YANG DIOPTIMASI DENGAN BAYESIAN OPTIMIZATION (RNN-BO)
Every year, traffic jams get worse as more and more vehicles fill the roads, causing delays for drivers. The solution to this problem is the implementation of Smart Transportation in Smart City which can determine the best route for drivers. To create this system, the You Only Look Once version 8 (YOLOv8) algorithm is used to count the number of vehicles in CCTV footage, while a Recurrent Neural Network optimized with Bayesian Optimization (RNN-BO) is used to predict road conditions based on reference tables. The Best First Search algorithm is then used to determine the best route for the driver. The dataset used consists of 4224 vehicles and a reference table with 5 columns and 320 rows of road conditions in .csv form. YOLOv8 produced a model with a Mean Average Precision (mAP) of 85.4% and a test accuracy of 69.52% for motorcycles and 87.71% for cars. Recurrent Neural Network (RNN) produces a model accuracy of 49.84% and prediction accuracy of 95.75%, which is then increased to a model accuracy of 57.46% through Bayesian Optimization. Finally, the Best First Search algorithm determines the best route based on road conditions and distance traveled, with the result that route 4 has the lowest weight for all conditions, including morning at 08:00 and 09:00, afternoon at 13:00 and 14:00, and in the afternoon at 16:00 and 17:00.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2307004707 | T126783 | T1267832023 | Central Library (REFERENS) | Available but not for loan - Not for Loan |
No other version available