การวิเคราะห์ระดับการให้บริการบริเวณทางแยกที่มีสัญญาณไฟจราจร โดยใช้แบบจาลองฮิดเดนมาร์คอฟ

Abstract

An signalized intersection may have significant impact to traffic condition. Level of Service (LOS) measurement is an essential component to evaluate and improve the vehicle management at the intersection. However, LOS could cost a highly budget especially when planning a long-term project. This study present analyzing and prediction LOS at an intersection where it is operated with signalized intersection and real-time traffic condition data on Google Map. The study took place in Rin Kham intersection, Muang district, Chiang Mai province, Thailand. Hidden Markov Model (HMM) was applied in this study, both discrete state and continuous state, to find correlation between LOS and traffic color shading of traftic conditions on Google Map. Those parameters will be determined in Transition Probability Matrix, Observation Probability Matrix, and Initial Probability Matrix. In conclusion, the HMM in both discrete state and continuous state gave similar results in which accurately forecasted LOS.