Using Particle Swarm and Brain Storm Optimization for Predicting Bus Arrival Time
Imam B. Mores (a), Muhammad Fauzan(a), Yul Y. Nazaruddin(a,b*), and Parsaulian I. Siregar(a)
a)Instrumentation and Control Research Group,
Department of Engineering Physics, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, Indonesia
b)National Center for Sustainable Transportation Technology, CRCS Building, 2nd floor, Jl. Ganesha 10, Bandung 40132, Indonesia
*yul[at]tf.itb.ac.id
Abstract
Particle Swarm Optimization (PSO) and Brain Storm Optimization (BSO) are alternative methods to find out the optimized solution of a non-linear equation. This paper will discuss the application of both methods to find out the weight of neurons from Adaptive Neuro Fuzzy Inference Systems (ANFIS) technique, which is used in predicting the bus arrival time at the bus stop. Comparison of the performance from both methods will also be made. After the modeling, training and testing of the proposed algorithm, the RMSE value produced from ANFIS which was trained by the PSO testing was 0.8145, and if it was trained by BSO was 0.8352. These results also conclude that the ANFIS with PSO algorithm yields better predicting bus arrival time better rather than ANFIS BSO in this case
Keywords: Adaptive Neuro-fuzzy Inference System, bus arrival time, Brain Storm Optimization, Particle Swarm Optimization
Topic: Control System