A soft computing approach to prediction of wheel induced rut depth: Appraisal of Artificial Neural Network

Que Wang, Feihu Leng


An effort to predict wheel induced rut depth under the effect of wheel load, velocity, number of passage and slippage using artificial neural network (ANN) technique trained by three network functions of feed-forward network (FFN), cascade-forward network (CFN) and Elman back-propagation network (EBN) is presented. Data were used from the series of experiments where a total of 81 samples were available for training, validating and testing the neural networks. Comparing a one-hidden layer architecture with a two layered one, the best topology to yield better performance on the criteria of lower root mean squared error (MSE), and coefficient of determination (R2) was the latter provided that the developed model is not trapped in the over-learning drawback. It was divulged that the ANN with 4-20-14-1 topology provided the best performance in terms of MSE and R2 equal with 5.733×10-15 and 0.991, respectively. It was inferred that ANNs could reliably be applied as a promising tool for the prediction of rut depth created by wheeled vehicle trafficking. 

Full Text:



Bayat H, Neyshabouri MR, Hajabbasi MA, Mahboubi AA, Mosaddeghi MR (2008) Comparing neural networks, linear and nonlinear regression techniques to model penetration re-sistance. Turkish Journal of Agriculture and Forestry 32: 425-433.

Bilgili M (2011). the use of artificial neural networks for forcasting the monthly mean soil temperatures in Adana, Turkey. Turkish Journal of Agriculture and Forestry35: 83-93.

Botta, G. F., Becerra, A. T., & Tourn, F. B. (2009). Effect of the number of tractor passes on soil rut depth and com-paction in two tillage regimes. Soil and Tillage Re-search, 103(2), 381-386.

Çarman K (2002) Compaction characteristics of towed wheels on clay load soil in a soil bin. Journal of Soil and Tillage Research 65(1):37-43.

Edlund, J., Bergsten, U., &Löfgren, B. (2012). Effects of two different forwarder steering and transmission drive sys-tems on rut dimensions. Journal of Terramechanics, 49(5), 291-297.

Jaiswal S, Benson ER, Bernard JC, Van Wicklen GL, (2005) Neural Network Modelling and Sensitivity Analysis of a Mechanical Poultry Catching System. Biosystems En-gineering 92(1): 59-68.

Kurjenluoma, J., Alakukku, L., &Ahokas, J. (2009). Rolling resistance and rut formation by implement tyres on tilled clay soil. Journal of Terramechanics,46(6), 267-275.

Liu, K., Ayers, P., Howard, H., & Anderson, A. (2009). Influence of turning radius on wheeled military vehicle in-duced rut formation. Journal of Terramechanics, 46(2), 49-55.

Liu, K., Ayers, P., Howard, H., & Anderson, A. (2010). Influence of soil and vehicle parameters on soil rut for-mation. Journal of Terramechanics, 47(3), 143-150.

Mohsenimanesh A, Ward SM (2010) Estimation of a three-dimensional tyre footprint using dynamic soil-tyre contact pressures. Journal of Terramechanics 47(6): 415-421.

Raper RL, Bailey AC, Burt EC, Way TR, Liberati P. Infla-tion pressure and dynamic load effects on soil deformation and soil-tire interface stresses. Trans ASAE 1995; 38(3): 685-9.

Roul AK, Raheman H, Pansare MS, Machavaram R(2009) Predicting the draught requirement of tillage implements in sandy clay loam soil using an artificial neural network, Bio-systems Engineering 104(4): 476-485.


  • There are currently no refbacks.