Moisture accumulation within road pavements, particularly in unbound granular materials with or without thin sprayed seals, presents significant challenges in high-rainfall regions such as Queensland. This infiltration often leads to various forms of pavement distress, eventually causing irreversible damage to the pavement structure. The moisture content within pavements exhibits considerable dynamism and directly influenced by environmental factors such as precipitation, air temperature, and relative humidity. This variability underscores the importance of monitoring moisture changes using real-time climatic data to assess pavement conditions for operational management or incorporating these effects during pavement design based on historical climate data. Consequently, there is an increasing demand for advanced, technology-driven methodologies to predict moisture variations based on climatic inputs. Addressing this gap, the present study employs five traditional machine learning (ML) algorithms, K-nearest neighbors (KNN), regression trees, random forest, support vector machines (SVMs), and gaussian process regression (GPR), to forecast moisture levels within pavement layers over time, with varying algorithm complexities. Using data collected from an instrumented road in Brisbane, Australia, which includes pavement moisture and climatic factors, the study develops predictive models to forecast moisture content at future time steps. The approach incorporates current moisture content, rather than averaged values, along with seasonality (both daily and annual), and key climatic factors to predict next step moisture. Model performance is evaluated using R2, MSE, RMSE, and MAPE metrics. Results show that ML algorithms can reliably predict long-term moisture variations in pavements, provided optimal hyperparameters are selected for each algorithm. The best-performing algorithms include KNN (the number of neighbours equals to 15), medium regression tree, medium random forest, coarse SVM, and simple GPR, with medium random forest outperforming the others. The study also identifies the optimal hyperparameter combinations for each algorithm, offering significant advancements in moisture prediction tools for pavement technology.

The majority of the Australian road network consists of unbound granular pavements with a thin bituminous surfacing. In Queensland, regional and remote areas, economic and environmental considerations encourage the use of locally available materials for the provision of granular pavements resulting in lower use of finite resources and a reduction in material transportation costs and associated emissions. These materials, known as non-standard or marginal materials, typically do not meet all the Queensland standard specification requirements but provide satisfactory performance when properly managed. At present, a universally accepted testing procedure for assessing the performance of non-standard materials is lacking. This paper reports on the first completed stages of a study aiming to investigate the physical and mechanical properties of non-standard materials using a range of laboratory testing: wheel tracking, modified Texas triaxial, California bearing ratio tests, and physical characterisation testing such as particle size distribution, Atterberg limits, compaction test, and apparent particle density measurement. This paper assessed 10 different non-standard materials together with one standard material using the selected laboratory tests. Later stages of the ongoing project, conducted for the National Asset Centre of Excellence (NACoE), will expand this testing and aims to develop a performance-based non-standard material screening tool to assist in material selection and assessments for road pavement construction and maintenance in low-traffic and low rainfall areas.

The behaviour of unbound granular materials (UGMs) used in road construction is crucial in determining the longevity and performance of road pavement. Geotechnical analysis can assist engineers in selecting suitable materials and designing road pavements that meet industry standards. This paper presents the results of laboratory geotechnical tests conducted on unbound granular materials (UGMs) collected from three sites (Roses Gap, Rules East, and Polkemmet Road) in Horsham, Victoria, Australia. UGMs were investigated for their mechanical behaviour and suitability as subgrade materials for road pavements. The study utilised laboratory geotechnical tests, including particle size distribution (PSD), Atterberg limits, compaction (Proctor) test, unconfined compressive strength (UCS), California Bearing Ratio (CBR), and repeated load triaxial (RLT) tests, to evaluate the physical and mechanical properties of the UGM samples. The study indicates that UGM samples collected from different locations displayed variations in their geotechnical properties, such as particle size distribution, water absorption, and CBR strength. Roses Gap samples showed weak cohesion properties, and significant vertical displacements after repeated triaxial tests. However, among the samples in this site, samples with higher clay content (RG21) demonstrated the most promise in triaxial tests. Similarly, the Rules East samples were found to be suitable for low-traffic subgrades due to their satisfactory CBR and RLT testing results, albeit with little cohesion from clay content. Out of three locations, Polkemmet samples were identified as potential subgrade applications, with PR12 being the top recommendation overall. It satisfied PSD, CBR, and RLT test conditions due to acceptable particle size in the largest range, highest CBR strength value, and lowest permanent displacement. The study's findings provide useful information for the design of road pavements using these materials and the characterisation of rural materials around the Horsham region for future use in various other contexts.

Unbound granular materials (UGMs) are extensively used in pavements mostly as subgrade and subbase materials. Excessive permanent settlement or rutting is the main damage mechanism encountered in UGMs. Rutting is a result of accumulated gradual plastic strain in the subbase and subgrade layers subjected to repetitive traffic loadings. Axisymmetric triaxial apparatus or repeated lateral triaxial (RLT) devices are commonly used to explore the rutting of UGMs. However, these devices are not able to capture the actual stress state generated in traffic. A soil element in pavement layers is in a three-dimensional (3D) stress state and includes all three components of cyclic principal stresses. A typical pavement also can be considered geometrically as a plane strain structure. Accordingly, aim of this study is to carry out experiments to determine the long-term deformation of a silty sand in plane strain and in a 3D stress state using a multistage true triaxial apparatus (TTA). It is found that the permanent deformation of UGMs under plane strain and 3D anisotropic stress state differs significantly from that under axisymmetric stress. An increase in the intermediate principal stress was observed to decrease the total and permanent deformation. An increase in cyclic stress level was also found to increase the rutting in UGMs. The deformation of soil under the plane strain state was found to be less than that in the axisymmetric stress state but falls into an intermediate range when compared to tests involving 3D cyclic loading.

In road engineering, the applications of piezoelectric ceramics on road energy harvesting or electronic sensing have drawn many attentions. However, the application potential of piezoelectric sensor on UGMs (unbound granular materials) in road is unclear. The main problem is the applicability of piezoelectric equation considering that the soil structure of UGMs and the working environment that the transducer may suffer may influence the transferred stress on the piezoelectric transducer and then the electricity generation. In this study, a twodimensional piezoelectric transducer made by PZT-5H piezoelectric ceramics was installed in the sample of UGMs, and a series of cyclic tests were performed on the UGMs samples by a large-scale triaxial apparatus. This test method is specially designed for the case that the piezoelectric transducer is installed in UGMs of pavement base/subbase layers and then undergo traffic loading under complex working environment. Test results show that the electricity generation is primarily dependent on the magnitude and frequency of cyclic loading, while the influences of other factors are limited. The piezoelectric equation to calculate the open-circuit voltage is deduced and verified under the complex working environment. This indicates that the piezoelectric equation is applicable to the case that the applied stress on the piezoelectric transducer is transferred from the soil medium, and the two-dimensional piezoelectric transducer has a sensor potential to monitor the dynamic vertical and horizontal soil stresses under traffic loading.