PEMODELAN ESTIMASI KERUSAKAN PERKERASAN JALAN (Studi Kasus Ruas Jalan Nasional di Wilayah Kerja PPK 3, Satker PJN Wilayah Propinsi DIY)

• Main Authors: , Emma Kartika C, , Arief Setiawan Budi Nugroho, S.T, M.Eng, Ph.D
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2014
• Subjects: ETD
• Online Access: https://repository.ugm.ac.id/128400/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=68743
• Institution: Universitas Gadjah Mada

Pavement Management System (PMS) gives information which is required in road management decision making, for example maintenance and rehabilitation (M&R) planning. Planning is required in order to choose the right scheme between repairing poor pavement and maintaining good pavement to prevent further degradation. PMS requires accurate pavement condition monitoring and appropriate pavement performance prediction so that M&R plan can be effective. This research was conducted by developing pavement distress estimation models. The developed models were used as base on road maintenance planning. This research used the national road network of PPK 3, Satker PJN, Yogyakarta Province as a case study. Pavement distress estimation models were developed based on Markov and Artificial Neural Network (ANN) theory, that used crack area, crack width, potholes, rutting, IRI and SDI as parameter. Markov model development used homogen Markov Chain which was generated based on condition transition probability for low and high traffic load category. Meanwhile, ANN model used data from traffic and road condition survey as input variable. Future road condition estimation values in ANN model were simulated by Matlab software . Both models were evaluated using regression from degradation curve approach. Analysis result shown that in general, based on regression of degradation curve approach, the R2 value of ANN model is higher compared to Markov model. ANN model is better than Markov model in case of its generality in modeling structure so that it can be used flexibly. Because of its flexibility, using related input variables, ANN model is able to develop road deterioration model based on IRI and SDI parameter which is not able to be developed by Markov model for high traffic load category. On the other hand, Markov model is superior in term of its easiness for maintenance plan arrangement. Maintenance plan which was made based on the developed model shown that annual routine maintenance gives optimum result, considered from cost term.