Farhad Salour Doctoral Thesis

 
Figure 31.
 
Accumulated permanent deformation from RLT tests and calibrated models 
for Luleå subgrade at moisture content: (a): 6.0%; (b): 8.0%; (c): 9.3%; (d): 10.1%.
(c) 
(d) 
(a) 
(b) 

---

49 
 
 
 
Figure 32.
 
Accumulated permanent deformation from RLT tests and calibrated models 
for Torpsbruk subgrade at moisture content: (a): 4.0%; (b): 5.5%; (c): 6.8%; (d): 7.4%.
(c) 
(d) 
(a) 
(b) 

---

50 
Main findings in Paper VI 

Generally, it was observed that the modified models that are based on the 
shakedown approach performed reasonably well in capturing the permanent 
deformation behaviour of the selected silty sand subgrade soils using data 
obtained from the RLT tests. 

Minor discrepancies between the predictive models were observed confirming 
the potential of using RLT test data and the presented approach for 
characterizing the permanent deformation of subgrade soils. 

Among the models used, the Tseng and Lytton (1989) model exhibited the best 
performance in most of the cases (different moisture content), considering the 
goodness of fit of the model and the accuracy in predicting the shakedown 
ranges at different stress paths. 

In most of the cases in the studied models, it was observed that usually several 
combinations of parameters would provide acceptable fit to the test data. The 
inconsistency in the acceptable range for the parameters was also observed in 
the literature, implying the apparent need for conducting more permanent 
deformation tests in order to set up a widely accepted range limits for the 
parameters of different models and/or possibly develop simpler models with a 
minimum number of parameters and validating these models through full-scale 
experiments on pavement structures. 

---

51