Calculation of Effective Structural Number Using Simple Neural Network for Some Road Links in Indonesia
In the pavement maintenance system, the parameter of effective structural number (SNeff) would be a considered factor in deciding whether a road link would be repaired or not. To calculate this parameter, it is required the testing of Falling Weight Deflectometer (FWD) and information of layer composition and thicknesses. The combination of these information and using the method of AASHTOâ€™93, it can be calculated the SNeff. These two information generally would be gained through the testings of core drill and test pit which would take time and cost. To overcome these problems, the neural network method or precisely the artificial neural network is developed for analysis of pavement structure. From the analysis, it can be said that the neural network of single perceptron can be used for predicting the SNeff with an acceptable error. In general the value of SNeff obtained from neural network calculation is lower than that of AASHTOâ€™93. In this paper it is also recommended to develop the neural network using multi layer perceptron for the use on pavement system analysis that might be decreasing the error.
 B. Xu, S. R. Ranjithan, and Y. R. Kim, Development Of Relationships Between FWD Deflections And Asphalt Pavement Layer Condition Indicators, Inter-national Journal of Compu-tational Intelligence Volu-me 3 Number 1, the TRB, Washington, D.C., 2001.
 Puspitaningrum , 2006. Pengantar Jaringan Syaraf Tiruan, Andi Offset, Yogyakarta
 Rojas R, 1996. Neural Networks A Systematic Introduction, Springer, Berlin
 Siegfried, 2009. Prediksi Tebal Lapisan Beraspal Meng-gunakan Data Lendutan FWD dan Metoda Neural Network Untuk Single Layer Perceptron. Jurnal Jalan dan Jembatan, Bandung.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.