Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 82
Edited by: B.H.V. Topping
Paper 52

Stability Prediction of Asphaltic Concrete Mixes using Neural Networks

M.H. Alawi, M.A. Saif and M.S. El-Bisy

Department of Civil Engineering, Umm Al-Qura University, Makkah, Saudi Arabia

Full Bibliographic Reference for this paper
M.H. Alawi, M.A. Saif, M.S. El-Bisy, "Stability Prediction of Asphaltic Concrete Mixes using Neural Networks", in B.H.V. Topping, (Editor), "Proceedings of the Eighth International Conference on the Application of Artificial Intelligence to Civil, Structural and Environmental Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 52, 2005. doi:10.4203/ccp.82.52
Keywords: neural networks, asphaltic concrete mixture, Marshall stability, Marshall flow.

The design of asphalt paving mixtures is a matter of selecting and proportioning materials to obtain a mix that will produce a durable pavement with the ability to carry the traffic loads without undergoing excessive distortion or displacement. The mix design should also provide for sufficient air voids and workability to allow for additional compaction under traffic loads and efficient placement of the mix [1].

Neural Networks technology provides several reliant analysis in many science and technology applications. In particular, Neural Networks (NN) are often applied to the development of statistical models for intrinsically non-linear systems, since NN behave better in complex conditions [2].

The present work employs the technique of the neural networks in order to estimate Marshall stability of asphaltic concrete mixtures. To predict the Marshall stability of asphaltic concrete mixtures with Neural Networks, samples were collected from different regions in Makkah area during the construction of new roads and tested (at the laboratories of the College of Engineering and Islamic Architecture of the Umm Al-Qura University) for bitumen content and gradation of aggregate determination. In addition Marshall stability and Marshall flow were measured.

A conventional model of Marshall stability of asphaltic concrete mixtures is developed on the basis of data collected. The data set is divided into three data files: a training file, a validation file, and a test file. Validation data are used to monitor the neural model's performance during training to prevent problems such as over-training. Test data are used to measure the performance of the trained neural model. The suitability of using artificial neural networks in developing a model of the Marshall stability of asphaltic concrete mixtures is explored; a model is developed and validated.

The results indicate that the model could predict the Marshall stability of asphaltic concrete mixtures with sufficient accuracy for practical design purposes. Transportation and highway engineers may use these models to predict the Marshall stability of asphaltic concrete mixtures and avoid conducting costly experimental tests that require specialized equipments and expertise.

Asphalt Institute, "Mix Design Methods for Asphalt Concrete and other Hot-Mix Types", The Manual Series NO.2 (MS-2), 6th Edition, 1995.
Charalambous, C., "Conjugate gradient algorithm for efficient training of artificial neural networks", IEEE Proceedings, 139(3), 301-310, 1992.
Hecht-Nielsen, R., "Neurocomputing", Addison-Wesley, Inc., 1990.
Maren, A., C. Harston, and R. Pap, "Handbook of Neural Computing Applications. Academic", San Diego, California, 488 pp., 1990.
Ministry of Transportation (M.O.T), "Bituminous Mix Design Form MRDWS 410D.c", Circular No. 2401/2403, Riyadh, Kingdom of Saudi Arabia, 1998.
Salim, Mahmood, "Highway Engineering (1)", Dar Alratib Aljameiah, Lebanon, 1984.
Smith, M., "Neural Networks for Statistical Modeling", Van Nostrand Reinhold, New York, 1993.
Willmott, C.J., "On the validation of models", Phys. Geogr., 2, 184-194, 1981.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description