Next Article in Journal
Heart Failure Prediction Based on Bootstrap Sampling and Weighted Fusion LightGBM Model
Previous Article in Journal
The Effect of Green Areas on Urban Microclimate: A University Campus Model Case
Previous Article in Special Issue
Effect of Waste Palm Oil Reclaiming Agent on Viscoelastic and Mechanical Properties of Hot-in-Place Recycled Mixture
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 15
Issue 8
10.3390/app15084359
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Experimental Investigation into Permeable Asphalt Pavement Based on Small-Scale Accelerated Testing

by
Bing Yang
1,2,
Hui Li
2,*,
Yingtao Li
1,
Murong Cheng
1,
Yang Sun
2 and
Yuzhao Han
2
1
Jiangsu Expressway Engineering Maintenance Technology Co., Ltd., Nanjing 210000, China
2
The Key Laboratory of Road and Traffic Engineering, Ministry of Education, Tongji University, Shanghai 201804, China
*
Author to whom correspondence should be addressed.
Appl. Sci. 2025, 15(8), 4359; https://doi.org/10.3390/app15084359
Submission received: 26 January 2025 / Revised: 21 March 2025 / Accepted: 28 March 2025 / Published: 15 April 2025
(This article belongs to the Special Issue Sustainable Asphalt Pavement Technologies)
Browse Figures
Versions Notes

Abstract

:
The durability of permeable pavement needs to be further studied by accelerated pavement testing (APT). Full-scale APT facilities are commonly associated with a very high initial investment and operational costs. A piece of small-scale accelerated testing equipment, the model mobile load simulator (MMLS), was used to investigate and evaluate the mechanical properties of three types of permeable asphalt pavements, including a 4 cm porous asphalt layer with cement-treated permeable base (4PA-CTPB), 7 cm porous asphalt layer with cement-treated permeable base (7PA-CTPB), and 7 cm porous asphalt layer with cement-treated base (7PA-CTB). Under different conditions of subgrade soil, transverse and longitudinal strains at the bottom of the porous asphalt layer and average rut depth and temperature data were collected. The results indicated that 4PA-CTPB produced the maximum average rut depth but minimum resilient tensile strain. The transverse resilient tensile strain of 7PA-CTPB was significantly higher than the other two structures under both wet and dry conditions. The transverse resilient tensile strain significantly increased with increasing loading cycles with a decreasing rate, which could be affected by both load and temperature. MMLS could be used to explore and evaluate the mechanical properties of permeable asphalt pavement. From the data under dry and wet conditions, it may be better to increase the strength of the subgrade, where a suitable hydraulic conductivity coefficient should be considered.

1. Introduction

The progression of urbanization has led to an extensive coverage of impermeable surfaces across metropolitan regions [1]. This phenomenon causes diminished infiltration and evaporation capacities, subsequently elevating surface runoff volumes while reducing groundwater replenishment rates [2]. As an eco-friendly stormwater management strategy, permeable pavement mitigates environmental impacts while enhancing vehicular safety and ride quality for motorists and adjacent communities [3,4]. This pavement system facilitates hydraulic connectivity through surface infiltration and potential subgrade percolation. The base layer functions as a temporary water retention reservoir to attenuate runoff volumes. Since 1970s, the fully permeable pavement had been implemented in numerous U.S. states for low-volume roadways and light-duty vehicle applications. The design prioritizes high void ratios in the surface and base layers to achieve optimal permeability and stormwater storage capacity. However, conventional designs employing minimally compacted subgrade soils frequently exhibit moisture susceptibility and a reduced service life. Alternative configurations utilizing high-strength permeable bases with a reduced porosity may improve structural durability when hydraulic storage functions are deprioritized, though the associated mechanical performance requires comprehensive investigation.
Previous studies have identified clogging of the surface layer, raveling, and cracking as primary types of distress of permeable pavement [5]. While laboratory testing remains fundamental for material characterization, such empirical data demonstrate a limited correlation with in-service pavement performance. Accelerated pavement testing (APT) has emerged as a robust methodology for evaluating structural response and distress mechanisms, including rutting, fatigue cracking, and deformation under controlled loading conditions [6,7,8,9,10]. APT enables the precise simulation of axle load magnitudes and repetition frequencies on constructed pavement sections, facilitating the short-term accumulation of damage equivalent to long-term field performance. Nevertheless, full-scale APT implementations incur substantial capital and operational expenditures. To address this limitation, a down-scale laboratory wheel tracking device, the model mobile load simulator (MMLS), was selected to test the mechanical response and performance of pavement. The model mobile load simulator (MMLS) has been adopted for mechanistic–empirical analysis. The MMLS3 apparatus, a one-third scale accelerated loading system, accommodates both in situ pavements and laboratory-constructed test sections. Kim [11] demonstrated the efficacy of a scaled APT for performance prediction through a comparative full-scale validation, emphasizing critical scaling factors including loading frequency, pavement layer thickness proportionality, and dimensional similitude requirements. Huang [12] conducted multi-scale APT comparisons, concluding that MMLS3 provides a cost-effective rutting and fatigue characterization when structural equivalency criteria are maintained. Subsequent investigations have employed scaled APT to analyze tire configuration effects [13], the scale of model pavements [14], and temperature on the performance of pavement [15].
Many studies have pointed out that MMLS3 has the potential to obtain the mechanical responses and performance of pavements in term of fatigue performance, permanent deformation [16,17,18,19], and other properties [20,21,22]. While full-scale APT initiatives continue to advance pavement engineering [23], their implementation remains cost-prohibitive for routine testing. Current MMLS3 applications predominantly focus on porous asphalt material characterization rather than holistic structural evaluation [24]. Numerical modeling approaches employing finite element analysis (FEA) and discrete element methods (DEMs) have supplemented experimental permeable pavement research [25,26,27,28]. The University of California Pavement Research Center (UCPRC) developed a mechanistic–empirical design framework for permeable interlocking concrete pavement (PICP), though the validation relied on FEA simulations rather than APT-derived mechanical responses [29].
The objective of this study was to investigate the mechanical properties of permeable pavement using a one-third scale model mobile load simulator (MMLS3). Three types of structures were designed to study the strain at the bottom of the surface layer under different thicknesses of the surface layer and different types of base materials. A dense cement-treated base and cement-treated permeable base were used in this study. Permeable pavements with permeable and impermeable bases were compared in both dry and wet conditions. In addition, the average rut depth of different structures was measured. Also, the variation in temperature of different structures and soil conditions are discussed.

2. Materials and Methods

2.1. Model Mobile Load Simulator MMLS3

The experimental investigation employed a one-third scaled accelerated loading apparatus, designated MMLS3, to evaluate model permeable pavement performance as illustrated in Figure 1. The MMLS3 system incorporates four reciprocating wheels generating 2.7 kN axle loads with a 0.7 MPa tire contact pressure. The technical specifications of the MMLS3 configuration are detailed in Table 1. Scaled and prototype systems demonstrate analogous mechanistic responses when maintaining dimensional similitude. For equivalent pavement materials with proportional layer thicknesses and linear elastic material behavior, prototype-to-model correlations require the adherence to scaling criteria: length, 1:n; load, 1:n2; time, 1:n; materials properties, 1:1; pressure on the surface, 1:1 (where n is scale factor, and n = 3 in this study). The scaled surface and base layer thicknesses were maintained at one-third of the prototype dimensions. To obtain full-scale displacements, the load duration in the model pavement should be equal to that of the prototype multiplied by the same scale factor used for the geometric dimensions. In other words, if the full scale and small scale use the same speed, the displacement of the small scale should multiply the scale factor, which will be equal to the displacement of the full scale [11]. In fact, the higher speed can result in the lower critical strain response in a full-scale APT [30,31]. The small-scale MMLS3 can produce the opposite results [28]. However, speed was outside the scope of this study. The maximum speed of MMLS3 is 9 km/h. The higher speed can reduce the duration, and 7.5 km/h was chosen for safety in this study. In addition, due to the small size of the scaled specimens, the rigid constraints of the mold walls could restrict the lateral deformation of the material, leading to stress concentration at the edges, which could lead to an overestimation of the rutting resistance or fatigue life. At the same time, the rigid boundaries could also suppress interlayer slippage and lateral plastic flow, causing the rutting deformation pattern to deviate from that of the actual pavement. In this study, to avoid the significant influence of the boundaries on strain and deformation, the strain and deformation under the wheel track were mainly investigated.

2.2. Structures and Materials

In this study, three types of permeable asphalt pavement, including two fully permeable structures and one drainage structure, were prepared in a steel track, which was 270 cm long, 90 cm wide, and 32 cm deep. Since there were two partitions at the bottom of the steel track, the length of the three structures were respectively divided into 85 cm (4PA-CTPB), 105 cm (7PA-CTPB), and 85 cm (7PA-CTB), as shown in Figure 2 and Table 2. An aquiclude using waterproof geotextile was installed between every two structures to prevent water from moving laterally. In addition, a filtration geotextile was installed between the subgrade layer and permeable base. The materials of the three structures were tested in the laboratory [32,33], and their properties are presented in Table 3. In this paper, a porous asphalt mixture (PA) was prepared with 4.6% high-viscosity modified asphalt and an aggregate with a maximum nominal size of 13 mm. The gradations of the porous asphalt mixture and base materials are presented in Table 4. The aggregate size should be small to prevent stripping. The permeability of the surface and base layer was tested in the laboratory. The field permeability was tested and compared with the laboratory test results. The results indicated that the field permeability was a little lower than that in the laboratory but still met the requirements of the infiltration function. In addition, the laboratory dynamic modulus master curve was obtained by fitting the modulus data using a sigmoid function (Equation (1)) and a temperature shift factor αT (Equation (2)).
log ⁡ E = δ + α 1 + e β + γ log ⁡ ( f α T )
l o g α T = − C 1 ( T − T r ) C 2 + T − T r
where E is the modulus (MPa); f is the frequency (Hz); δ, α, β, γ, C1, and C2 are the fitting parameters; T is the test temperature; and Tr is the reference temperature. The fitting parameters are presented in Table 5, and the master curve is plotted in Figure 3. The reference temperature was 20 °C in this study. It could be seen that as the loading frequency increased, the dynamic modulus of the mixture gradually increased. As the temperature rose, the dynamic modulus gradually decreased. Therefore, it could be foreseen that in the accelerated loading test, under the condition that the loading frequency and pressure were fixed, as the temperature rose, the modulus of the mixture might decrease and the resulting strain might increase.

2.3. Preparation of Structures

In the first phase of this study, an experimental small-scale model pavement was built for MMLS3 testing. Figure 4 presents an overall view of building these structures. The building procedures are as follows:
(1)
Subgrade: The waterproof geotextile was spread all over the bottom and all around the steel tank, soil evenly sprinkled on the inside of the steel tank, and a small roller used to compact it to the required thickness and compaction;
(2)
Base: The filtration geotextile was laid on top of the subgrade of 4PA-CTPB and 7PA-CTPB, and the waterproof geotextile was used to separate 4PA-CTPB, 7PA-CTPB, and 7PA-CTB. Then the base materials were evenly paved and rolled to the required thickness;
(3)
Interface bonding layer: After 7 days, a layer of tack coat was evenly spread on top of the base, and the permeability of base was examined;
(4)
Surface layer: After 2 days, the porous asphalt mixture was paved and rolled to the required thickness with a small roller; the paving temperature was 175 °C, and the rolling temperature was 150 °C.
These structures were instrumented with horizontal fiber Bragg grating strain gauges and temperature gauges as shown in Figure 5. These strain gauges were used to monitor transverse and longitudinal strain at the bottom of the surface layer. They were embedded 4 cm, 7 cm, and 7 cm below the surface of 4PA-CTPB, 7PA-CTPB, and 7PA-CTB, respectively, in the interlayers of the asphalt concrete courses. They were positioned following a line coincident with the axis of the wheel track. Each strain gauge was alternatively set perpendicular or longitudinal to the wheel track.

2.4. Test Procedure

In this study, the effects of dry and wet subgrade on the mechanical properties were explored. Wheel track 1 was firstly loaded under dry condition as shown in Figure 5. After 0.3 million cycles of loading, the MMLS3 was moved to wheel track 2. At the same time, water was poured into the inner pavement through a pre-set pipe. In addition, during the entire loading, water was poured into the pavement to maintain the wet condition of the subgrade by checking the water level in the pipe. From Figure 4b, it can be seen that a PVC pipe was installed in the pavement as an observation well. In this study, these 3 types of structures were loaded for 0.3 million cycles at wheel track 1 and 0.85 million cycles at wheel track 2. In fact, the structures of 4PA-CTPB-D and 7PA-CTPB-D were loaded for 0.35 million cycles, where the deformation and strain barely increased with increasing cycles after 0.35 million cycles. The structure of 7PA-CTB-D was loaded for 0.3 million cycles. Moreover, since the length of the wheel track was only 1260 mm, it was impossible to load three structures at the same time. Therefore, 4PA-CTPB and 7PA-CTPB were firstly loaded for 0.35 million cycles under dry condition, then 7PA-CTB was loaded for 0.3 million cycles. After tests under dry conditions, water was poured into the inner structure of 4PA-CTPB and 7PA-CTPB to maintain the wet state of the top of the subgrade. 7PA-CTPB and 7PA-CTB were loaded for 0.85 million cycles in this study. More loading cycles were chosen to investigate the effects of water on the structural mechanical properties.
During the period of testing, MMLS3 was used to load the experimental pavement, and no temperature control was used during the entire testing. The strain responses, rut depth, and temperature data were measured during the entire MMLS testing. The model mobile load simulator was running at daytime and stopped at nighttime. The strain data were stored for 30 s every hour. Due to the stopping of the MMLS3 device (MIS, Randburg, South African) at the nighttime, the entire testing lasted several days. The data collection frequency was 150 Hz.
In this study, the permeability coefficient of pavement before and after loading was tested according to ASTM C1701 [34]. Before loading, the permeability coefficients of 4PA-CTPB, 7PA-CTPB, and 7PA-CTB were 0.18 cm/s, 0.17 cm/s, and 0.14 cm/s, respectively. After loading, the permeability coefficients of the three pavement structures were 0.17 cm/s, 0.15 cm/s, and 0.12 cm/s, respectively. The load resulted in a decrease in the air void of the asphalt mixture in the wheel track and a decrease in the permeability coefficient of the pavement structure. Because the accelerated loading test was an indoor test, no excess soil or contaminants blocked the air void, and the effect of clogging on the long-term performance of permeable pavement is not discussed.

3. Results and Discussion

3.1. Dry Condition

The strains at the bottom of the surface layer of the three structures were collected. Figure 6 presents the history of the longitudinal and transverse strain versus time at the initial loading stage. In general, the total strain response of the asphalt pavement includes two parts: permanent strain and recoverable resilient strain. The resilient strain is defined as the magnitude of the strain recovering after a complete loading and unloading cycle. It was calculated as the maximum recoverable strain in 30 s of data acquisition. As shown in Figure 6, the peak of the strain curve of each structure corresponded to a wheel load. It was obvious that the longitudinal strain had alternations of tension and compression, while the transverse strain only demonstrated a tension state [35,36]. Each tension peak corresponded to a wheel load located right above the strain gauge. When the strain reached a peak, there was next a slow recovery process, which was the stress relaxation process duo to the viscoelastic property of the asphalt mixture. In addition, the effect of the material’s viscoelasticity results in the asymmetric shape of the strain curves under transient tire loading, especially due to delayed recovery after wheel load has passed the strain gauge position. Due to damage to the longitudinal strain gauge at the bottom of 7PA-CTB-D, it was impossible to obtain accurate longitudinal strain data under dry conditions. Therefore, the longitudinal strain of 7PA-CTB-D is not discussed in this paper. From Figure 6a, the resilient tensile strain of 4PA-CTPB-D is close to the compressive strain at a value of 28 με. However, the resilient tensile strain of 7PA-CTPB-D was 12 με, which was obviously less than the 36 με of the resilient compressive strain. From Figure 6b, the transverse resilient tensile strain of 4PA-CTPB-D is less than the other two structures. The resilient tensile strain of 7PA-CTPB-D was 94 με, which was significantly larger than that of 7PA-CTB-D at a value of 42 με.
Figure 7 presents the variation in the resilient strain of the whole loading period and the first day of loading under dry conditions. From Figure 7a, the transverse resilient strain of 7PA-CTPB-D is significantly larger than the other strains. If the thickness of the surface layer was thin enough, this could lead to more compression or even induce compressive strain. These strains were partially from the viscoelastic property and partially from the material’s movement under the wheel track due to the permanent deformation [16]. And the effect of permanent deformation was much higher for the transverse strain gauges compared with the longitudinal strain gauges [37,38,39]. In addition, the edges of the steel tank were close to the wheel track, resulting in higher transverse resilient tensile strain. However, the longitudinal strain could be higher than the transverse strain, as 4PA-CTPB-D shows. Some studies showed that when multiple axles and multiple wheels are used with a wander, the longitudinal strain can be higher than the transverse strain. However, only a single wheel and single axle were used in this study. Therefore, that was probably because a thinner surface layer was used in 4PA-CTPB-D.
In particular, stopping the machine resulted in typical relaxation curves between the stop and subsequent start time, indicating similar relaxation behavior for any loading period. In addition, during a day loading, the transverse resilient strain of 7PA-CTPB-D increased with increasing cycles, and the range in the variation of strain was 80~100 με. However, the longitudinal resilient strain of 7PA-CTPB-D did not change obviously. Moreover, with increasing loading cycles, the ranges in both the longitudinal and transverse resilient strains of 4PA-CTPB-D were relatively small, which indicated that the thinner surface layer resulted in less tension than the thicker surface layer under load. Since the wheel track of MMLS3 was not long enough to load three types of structures at the same time, 7PA-CTB was loaded separately at wheel track 1 under dry conditions. As shown in Figure 7c, the range in the transverse resilient tensile strain of 7PA-CTB-D was between 30 με and 50 με, which was significantly less than that of 7PA-CTPB-D. That was because the base layer of 7PA-CTB-D was a dense cement-treated base with a larger modulus. Figure 7b,d, respectively, demonstrate the variation in the resilient strains of the three types of structures on the first day. The longitudinal resilient tensile strain of 4PA-CTPB-D had a more significant increase than the transverse resilient strain. On the contrary, the transverse tensile resilient strain of 7PA-CTPB-D and 7PA-CTB-D increased significantly, while the longitudinal tensile strain of 7PA-CTPB-D did not change significantly.
In general, the longitudinal strain of the bottom of the surface layer of asphalt pavement is chosen to evaluate its fatigue performance, and it can be used to design the thickness of pavement structures. The specified thickness can be calculated by use of a mechanistic–empirical method based on accelerated pavement testing. In this study, the results of strain responses indicated the transverse strain of 7PA-CTPB-D or 7PA-CTB-D was higher than its longitudinal strain, whereas the transverse strain of 4PA-CTPB-D was lower than its longitudinal strain. Therefore, a higher strain response was chosen to compare the performance of different structures. From Figure 7, it can be seen that 4PA-CTPB-D produced a minimum resilient tensile strain, which indicated that the structure with a thinner surface layer could obtain better anti-fatigue performance, whereas a thinner surface was more likely to produce shear failure (depending on the nature of the supporting layer). This was consistent with the results of some studies; the asphalt layer was subjected to the alternating effect of tensile and compressive strain, and the thicker asphalt layer had a lower tensile stress and better fatigue resistance [40,41]. Moreover, the strain of 7PA-CTB-D was far below the strain of 7PA-CTPB-D, which indicated the higher modulus of the base layer induced a much lower strain response. When using the same thickness of surface and base layer, a porous base layer with a lower modulus induced a little higher strain at the bottom of the surface layer. Materials with a relatively high modulus and a thicker surface could be beneficial to fully permeable pavement. However, the construction costs and economic benefits should be considered.

3.2. Wet Conditions

Rainwater can infiltrate into natural soil for fully permeable pavement, which leads to the saturation status of the subgrade for long time during its service life. Therefore, when loading 7PA-CTPB-W and 7PA-CTB-W, water was added to maintain the water immersion status of the subgrade. However, it was found that water flowed out of the structures after a few minutes, and the subgrade of 4PA-CTPB was thicker than the other two structures. In addition, the waterproof geotextile was used to separate 4PA-CTPB, 7PA-CTPB, and 7PA-CTB. Therefore, water could not flow into 4PA-CTPB. Water was poured into 7PA-CTPB and 7PA-CTB every hour to maintain the wet state of the top of the subgrade. The above operation could keep the top of the subgrade in a wet state or saturated state. Figure 8 and Figure 9 present the results of the resilient tensile strain of 7PA-CTPB-W and 7PA-CTB-W. As shown in Figure 8a, the longitudinal strain of 7PA-CTPB-W and 7PA-CTB-W consisted of both compression and tension, whereas the transverse strain showed tension only. In addition, 7PA-CTPB-W presented a typical longitudinal strain curve (compression–tension–compression), whereas 7PA-CTB-W had a variation from compression to tension only. From Figure 8a, the longitudinal resilient strain of 7PA-CTB-W recovered slowly from the ending of one loading cycle to the beginning of the next loading cycle whereas 7PA-CTPB-W recovered rapidly. The reason could be that the cement-treated base had a higher modulus or rigid nature.
The transverse strains of the two types of structures showed higher resilient tensile strains compared with the longitudinal strains from Figure 9a. 7PA-CTPB-D also followed the same rule. The transverse resilient tensile strain of both 7PA-CTPB-W and 7PA-CTB-W presented tension only, and 7PA-CTPB-W showed a higher transverse resilient strain caused by the lower modulus of its base and the saturated subgrade. With increasing loading cycles, the longitudinal resilient tensile strain of 7PA-CTPB-W changed little, while 7PA-CTB-W’s increased significantly.
Figure 7a and Figure 9a demonstrate resilient tensile strain versus cycles; these strain histories show an initial high rate of increase (primary phase), followed by a lower rate of increase (secondary phase). In general, the curve of fatigue life had a primary and secondary phase and finally a sudden high rate of increase (tertiary phase) just prior to the failure. Since the testing was intermittent, the strain recovered during relaxation. Therefore, the three types of structures produced no failures in this study. However, an increase in resilient strain in a day could indicate the damage development of the porous asphalt mixture under continuous loading.
Moreover, as shown in Table 6, typical resilient tensile strains for the initial loading stage were chosen. The initial loading stage meant the corresponding structure just began to be loaded. The longitudinal resilient tensile strains of 7PA-CTPB-D and 7PA-CTPB-W were close to each other, whereas the transverse resilient tensile strain of 7PA-CTPB-D under dry conditions was nearly twice as large as that of 7PA-CTPB-W under wet condition. The tested results proved that the water sensitivity of subgrade soil could be an important factor because of a significant reduction in resilient tensile strain under wet condition. Therefore, this would suggest that the subgrade should comprise high-quality materials for the successful performance of permeable pavements. It might be necessary to treat the natural or low-compacting-level subgrade for specific requirements. However, treated subgrade will increase costs. Therefore, investors and operators should choose the types of structures flexibly. Moreover, the transverse resilient tensile strain of 7PA-CTB-D was a little higher than that of 7PA-CTB-W from Table 6, which could be caused by the better compaction of the subgrade, leading to a higher modulus. Before the paving base materials, the compaction level of the subgrade was only 90%, and the soil could be compacted again under dry conditions. Moreover, the water content of the soil increased in the short term when water was poured into the inner pavement. The subgrade soil gradually reached the optimal moisture content, leading to an increased compaction level and modulus. The increase in the modulus of subgrade soil was beneficial, leading to a lower tensile strain. Lu used APT to study the mechanical response of permeable pavement under different saturation conditions. The results showed that the mechanical response of permeable pavement was obviously affected by water content, and an increase in water content led to an increase in vertical total stress, which might lead to an increase in deformation. With a decrease in water content, the horizontal strain at the bottom decreased [42]. Compared with 7PA-CTB-W with a dense cement-treated base, the increase in modulus of the subgrade soil produced a more significant beneficial effect on 7PA-CTPB-W with a cement-treated permeable base. The water produced no obviously adverse impact due to the better compaction of the subgrade. As is well known to all, the low strength of permeable pavement makes it difficult to apply it in heavy traffic. It could be better to increase the strength of the subgrade, where a suitable permeability coefficient should be considered at the same time. In addition, the limitations of objective test conditions should be realized, such as uncontrollable environmental conditions, differences due to different construction technology, and instrument and equipment error, which could mean that some results and conclusions were not ideal.
In addition, the stripping of the surface was absent during the whole loading duration. In this study, water was poured into pavement to maintain the wet condition of the subgrade. The research indicated that the deterioration of pavement slabs is accelerated by water, and the fatigue life is dramatically shortened compared to dry situations [43,44]. It was found that water flowed out of the structures after a few minutes. The base layer and surface layer were not affected by moisture distribution. Therefore, stripping, pumping, and other properties were not considered. Generally, a faster speed can lead to the higher pore–water pressure ratio and mud pumping. The increasing class of roadways and loading cycles also induces strong pumping [45,46,47]. In this study, the filtration geotextile was laid on the top of the subgrade of 4PA-CTPB and 7PA-CTPB, which could decrease the pumping risk.

3.3. Results of Rut Depth

After each tens of thousands of loading cycles, the average rut depth was measured. Figure 10a shows the average rut depth of the three types of structures under dry conditions. In the primary phase of loading, the average rut depth increased significantly. With increasing cycles, the surface of the asphalt pavement was compacted, and the average rut depth had a lower rate of increase. In addition, after 0.3 million cycles, 4PA-CTPB-D produced the highest rut depth, and 7PA-CTB-D produced the lowest rut depth, which indicated the increasing thickness of the surface layer and modulus of the base layer could decrease rut depth. 7PA-CTB-D had a thicker surface layer and higher modulus of base layer, which could be the reason why 7PA-CTB-D produced the lowest rut depth. Generally, using less binder or a stiffer binder both improved pavements’ rutting performance. In addition, pavements with a thin surface layer had less permanent compression than pavements with a thick surface layer, mostly because the thicker layer has more material to deform [48]. The rut depth of 4PA-CTPB-D was higher than that of 7PA-CTPB-D. In other words, the thinner surface layer produced a larger rut depth. The results could be affected by the scale and other environmental and test conditions.
Figure 10b presents the average rut depth of 7PA-CTPB-W and 7PA-CTB-W under wet conditions. Compared with dry conditions, it was similar in that the average rut depth increased significantly in the primary phase of loading, and the average rut depth increased slowly in the secondary phase. In addition, the average rut depth had a more significant difference between 7PA-CTPB-W and 7PA-CTB-W, and the average rut depth of 7PA-CTPB-W was higher than that of 7PA-CTB-W. Compared with the rut depth under dry conditions, the average rut depth of 7PA-CTPB-W increased to 2.65 mm only after 0.85 million cycles, whereas 7PA-CTPB-D increased to 2.77 mm after 0.3 million. Similarly, the average rut depth of 7PA-CTB-D was higher than 7PA-CTB-W. Fully permeable pavements allow rainwater to infiltrate into the subgrade, which results in reduced stability due to the infiltration and scour of rainwater. The subgrade is mostly affected by rainwater [49]. Under the influence of water and repetitive load, the subgrade will easily lose its strength and stability, resulting in damage to the entire pavement structure. In this study, the initial permeability was 2.3 × 10−5 cm/s, and the initial compacting level was 90%. The research indicated increasing the strength or compacting level could effectively improve the deformation resistance of the surface layer. The subgrade with a high modulus was able to change the stress–strain behavior of the pavement structure and decreased fatigue and rutting damage [50,51]. Finite element and mechanical analyses indicated increasing the elastic modulus of the subgrade could decrease the deflection of surface and tensile stress or strain at the bottom of the surface layer [52,53]. In this study, the initial compacting level of the subgrade was 90%; the subgrade could be recompacted after water immersion and being loaded. After the subgrade was soaked, the average rut depth of the surface layer decreased. This might be caused by re-compaction of the subgrade, which increased the strength of the subgrade. Moreover, from Table 6, the recoverable strain of 7PA-CTPB-D is significantly higher than that of 7PA-CTPB-W. The increase in the modulus of the subgrade soil was beneficial, leading to a lower tensile strain. This is consistent with the results of strain in this study. Moreover, the strength of CTB and CTPB might have increased after loading under dry conditions.
The results of average rut depth confirmed that the thickness of the surface layer and the base materials were critical parameters. The rutting resistance of the dense asphalt mixture and porous asphalt mixture was compared; the thicker dense asphalt mixture may also lead to early rutting fatigue. This showed that surface thickness was one of the important factors affecting rutting deformation. And the study has shown that thinner asphalt surfaces are prone to rutting earlier [54,55]. A thinner surface layer could produce a more permanent deformation. However, pavements with a thin surface layer could also produce a less permanent deformation, mostly because the thicker layer has more material to deform [56]. Moreover, the denser or stiffer base provides a better rutting resistance performance. Therefore, a thicker surface layer or a stiffer base could be better for a fully permeable pavement; this conclusion was consistent with that obtained by the resilient tensile strain.

3.4. Results of Temperatures of Bottom Layer

In this study, the temperature at the bottom of the surface layer was measured, and the temperature differences from the initial temperature are demonstrated in Figure 11. The temperature differences of different structures during the entire loading are shown in Figure 11a–c. The temperature differences presented the same variation trend as strain versus cycles. The temperature differences increased with increasing loading cycles during the daytime, and the temperature difference dropped to a minimum during the nighttime. Therefore, the strain variation was caused by both load and temperature. For outdoor experiments, there exist many factors, especially temperature. Therefore, it was extremely difficult to maintain an exact target temperature for the entire duration of the test; the plots show that the variations were within 6 °C. As shown in Figure 11a, 4PA-CTPB-D produced a slightly higher temperature at the bottom of the surface layer due to a thinner surface layer than 7PA-CTPB-D. From Figure 11b, the temperature differences of 7PA-CTPB-D were a little higher than those of 7PA-CTB-D, which could be because porous base materials had a lower heat conductivity, leading to more heat accumulating on top of the porous base layer. Figure 11c demonstrates that the temperature differences of 7PA-CTPB-W were a little higher than those of 7PA-CTB-W. The denser cement-treated base had a higher heat conductivity, which resulted in the fact that heat could transfer more easily into the base and subgrade. Moreover, after a relaxation of one night, the temperature difference of 7PA-CTB-W was a little lower than that of 7PA-CTPB-W, which could be caused by porous base material with a low heat conductivity. Compared with a drainage pavement structure, fully permeable pavement with a porous base could produce a higher temperature at the bottom of the surface layer, potentially leading to more damage to the surface layer, because porous base materials have a lower heat conductivity compared with dense base materials. The heat transfer principle is mainly affected by the heat conductivity of materials. Heat transfer efficiency reduces when heat transfers from the surface layer to the base layer. Heat will accumulate between the two layers.
In this study, the accelerated pavement testing was stopped every night, which resulted in typical relaxation behavior; the strain and temperature differences at the beginning of each day of loading are presented in Figure 11d–f. The transverse strain obviously varied with the temperature differences, while the variation in longitudinal strain with temperature difference was not obvious. From Figure 11d, 4PA-CTPB-D induced a little less variation in strain that 7PA-CTPB-D, which indicated that the thicker surface layer had greater fluctuations in temperature and strain. From Figure 11e, compared with the other two structures, 7PA-CTB-D presents a similar change law. Especially, both the transverse and longitudinal strains of 7PA-CTB-W obviously varied with temperature differences and present a similar change trend from Figure 11f. After the relaxation of a night, the strain recovered, and the remaining unrecovered strain left. From Figure 11d–f, both the transverse and longitudinal strains of the three structures varied with temperature, and no obvious permanent unrecovered strain was produced, which indicated that temperature was a critical factor affecting the strain. In this study, the environmental condition was uncontrollable, and further study should consider this factor. However, continuous full-scale accelerated pavement testing should be conducted on a field pavement. Therefore, if allowed, conducting continuous down-scaled accelerated pavement testing would obtain better results.
The above results indicated that the temperature produced some effect on the viscoelastic asphalt layer. In some other studies, the resilient strain history was strongly dependent on the temperature and temperature gradient [15,25]. The gradient of resilient strain is a function of temperature and the temperature gradient. If the temperature or temperature gradient is very high, the resilient strain can reach a high value very quickly, and this can increase the fatigue damage process. MMLS has been identified as an effective, economic, and reliable trafficking tool to characterize the rutting and fatigue performance of pavement materials. However, due to relaxation at night in this study, the viscous behavior and fatigue damage accumulation were absent; only the resilient strain or elastic behavior was significant. The elastic behavior has been presented in longitudinal and transverse strain results under dry and wet conditions in this study. From the master curve of Figure 3, increasing temperatures resulted in the reduced modulus of the porous asphalt mixture. From the results of this study and other studies [15,57,58], a higher temperature could result in a higher resilient strain, which could increase the fatigue damage process. This might be related to the modulus reduction of the surface layer.

4. Conclusions

In this study, a piece of one-third scale accelerated loading equipment, MMLS3, was used to investigate the mechanical properties of permeable asphalt pavement under different moisture conditions of the subgrade. Three types of permeable asphalt pavement structures were designed to study the effects of surface thickness and base type on the strain responses and rutting of permeable asphalt pavement. Based on the results of this study, the following conclusions can be drawn:
  • The transverse resilient tensile strain significantly increased with increasing loading cycles with a decreasing rate, and the increase could be affected by temperature. The thickness of the surface layer and base materials had effects on the longitudinal strain, leading to different strain change patterns;
  • Under dry conditions, a structure with a thinner surface layer was more prone to produce a larger permanent deformation and a lower tensile strain in the surface layer. In addition, the base materials had a significant effect on tensile strain at the bottom of the surface layer;
  • Compared with dry conditions, the resilient strain of the structures with different base materials (7PA-CTPB-D and 7PA-CTB-D) presented a similar change trend under wet conditions. Compared with dry conditions, both the longitudinal and transverse resilient tensile strain decreased under wet conditions. In addition, compared with the rut depth under dry conditions, the average rut depth was slightly lower after 0.85 million cycles under wet conditions, which could be caused by the better compaction of the subgrade soil and the increased strength of the base materials after loading for dry conditions;
  • The temperature differences at the bottom of the surface layer presented the same variation trend as the strain cycles. The temperature differences increased with increasing loading cycles during each day of loading, which indicated the strain variation was caused by both temperature and load. In addition, fully permeable asphalt pavement with a permeable base produced a higher temperature at the bottom of the surface layer, potentially leading to more damage to the surface layer.
This study demonstrated the non-linear accumulation of transverse strain with loading cycles, while proposing optimized design strategies through the dual consideration of subgrade strength enhancement and hydraulic conductivity requirements. And the test conditions were closer to the field environment, which provides a reference for evaluating the long-term performance of different permeable asphalt pavement structures. Accelerated loading tests are usually loaded hundreds of thousands or even millions of times to study the long-term performance of different structures, requiring the construction of circuit roads. In this study, three different pavement structures were designed, and accelerated loading tests were carried out at the same time. The application of scaled APT with rigorous similitude principles provides a methodological advancement for an economical yet mechanistically valid pavement performance assessment. In addition, this study also pointed out the impact of temperature and load continuity on the experimental results; the temperature and the saturation degree of subgrade should be precisely controlled in future studies.

Author Contributions

Conceptualization, B.Y. and H.L.; methodology, H.L.; formal analysis, Y.L.; investigation, all authors; writing—original draft preparation, B.Y.; writing—review and editing, Y.S.; supervision, Y.H.; project administration and funding acquisition, H.L. and M.C. All authors have read and agreed to the published version of the manuscript.

Funding

The research is supported by the National Key R&D Program of China (2023YFE0121400 & 2023YFB2604000), the Science and Technology Commission of Shanghai Municipality (Grant No. 23210711400), Gansu Provincial Science and Technology Plan (23JRRA1375), and the research project of Jiangsu Communications Holding Co., Ltd. (No. 2024JKY13).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data available on request due to restrictions.

Conflicts of Interest

The authors declare that this study received funding from Jiangsu Communications Holding Co., Ltd. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication. Authors Bing Yang, Yingtao Li and Murong Cheng were employed by the company Jiangsu Expressway Engineering Maintenance Technology Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Turco, M.; Kodešová, R.; Brunetti, G.; Nikodem, A.; Fér, M.; Piro, P. Unsaturated hydraulic behaviour of a permeable pavement: Laboratory investigation and numerical analysis by using the HYDRUS-2D model. J. Hydrol. 2017, 554, 780–791. [Google Scholar] [CrossRef]
  2. Min, S.K.; Zhang, X.; Zwiers, F.W.; Hegerl, G.C. Human contribution to more-intense precipitation extremes. Nature 2011, 470, 378–381. [Google Scholar] [CrossRef]
  3. Scholz, M.; Grabowiecki, P. Review of permeable pavement systems. Build. Environ. 2007, 42, 3830–3836. [Google Scholar] [CrossRef]
  4. Meiarashi, S. Researches on low noise pavement in Japan. J. Acoust. Soc. Jpn. (E) 1999, 20, 19–27. [Google Scholar] [CrossRef]
  5. Li, H.; Jones, D.; Harvey, J. Development of Mechanistic-Empirical Design Procedure for Fully Permeable Pavement Under Heavy Traffic. Transp. Res. Rec. 2012, 2, 83–94. [Google Scholar] [CrossRef]
  6. Dessouky, S.; Pothuganti, A.; Walubita, L.F.; Rand, D. Laboratory Evaluation of the Workability and Compactability of Asphaltic Materials prior to Road Construction. J. Mater. Civ. Eng. 2013, 25, 810–818. [Google Scholar] [CrossRef]
  7. Kazmee, H.; Tutumluer, E.; Beshears, S. Using Accelerated Pavement Testing to Evaluate Reclaimed Asphalt Pavement Materials for Pavement Unbound Granular Layers. J. Mater. Civ. Eng. 2017, 29, 04016205. [Google Scholar] [CrossRef]
  8. Saeedzadeh, R.; Romanoschi, S.A.; Akbariyeh, N.; Khajeh-Hosseini, M.; Abdullah, A.Q. Sustainability Assessment of Recycled Asphalt Mixtures Based on Performance in Full-Scale Testing. J. Transp. Eng. B-Pavements 2018, 144, 04018024. [Google Scholar] [CrossRef]
  9. Kohler, E.; Plessis, L.D.; Harvey, J.; Motumah, L. Comparison of precast and cast-in-place concrete pavements responses under heavy vehicle simulator loads. In Proceedings of the 8th International Conference on the Bearing Capacity of Roads, Railways & Airfields International Conference, Champaign, IL, USA, 29 June–2 July 2009. [Google Scholar] [CrossRef]
  10. Iida, A.; Shimeno, S.; Kaneda, M.; Satou, M. Evaluation of Plastic Flow Rutting Resistance of Dense-Graded Asphalt Pavements on Expressways. J. Pavement Eng. JSCE 1999, 4, 173–178. [Google Scholar] [CrossRef]
  11. Kim, S.-M.; Hugo, F. Small-Scale accelerated pavement testing. J. Transp. Eng. 1998, 124, 117–122. [Google Scholar] [CrossRef]
  12. Huang, Y.; Wang, L.; Xiong, H. Evaluation of pavement response and performance under different scales of APT facilities. Road Mater. Pavement Des. 2017, 18, 159–169. [Google Scholar] [CrossRef]
  13. Al-Qadi, I.L.; Wang, H. Full-depth Pavement Responses under Various Tire Configurations: Accelerated Pavement Testing and Finite Element Modeling. Presented at the Asphalt Paving Technology 2009, Minneapolis, MN, USA, 15–18 March 2009; pp. 721–760. Available online: https://www.researchgate.net/publication/288714532 (accessed on 1 January 2009).
  14. Frederick, H.P.E. Comparative scaled MMLS3 tests versus full-scale MLS10 tests in Mozambique. Presented at the Transportation Research Board Meeting, Washington, DC, USA, 21–25 January 2007. [Google Scholar]
  15. Bhattacharjee, S.; Mallick, R.B. Effect of temperature on fatigue performance of hot mix asphalt tested under model mobile load simulator. Int. J. Pavement Eng. 2012, 13, 166–180. [Google Scholar] [CrossRef]
  16. Bhattacharjee, S.; Gould, J.S.; Mallick, R.B.; Hugo, F. An Evaluation of use of Accelerated Loading Equipment for Determination of Fatigue Performance of Asphalt Pavement in Laboratory. Int. J. Pavement Eng. 2004, 5, 61–79. [Google Scholar] [CrossRef]
  17. Smit, A.; Hugo, F.; Rand, D.; Powell, B. Model mobile load simulator testing at National Center for Asphalt Technology test track. Transp. Res. Rec. 2003, 1832, 182–190. [Google Scholar] [CrossRef]
  18. Bhattacharjee, S.; Mallick, R.B. Determination of Damage Development in Asphalt Concrete Using Small-Scale Accelerated Pavement Testing via Frequency Domain Analysis Approach. J. Transp. Eng. ASCE 2012, 138, 723–731. [Google Scholar] [CrossRef]
  19. Lee, J.; Kim, Y.R.; Lee, J. Rutting performance evaluation of asphalt mix with different types of geosynthetics using MMLS3. Int. J. Pavement Eng. 2014, 16, 894–905. [Google Scholar] [CrossRef]
  20. Liu, Y.; Huang, Y.; Sun, W.; Nair, H.; Lane, D.S.; Wang, L. Effect of coarse aggregate morphology on the mechanical properties of stone matrix asphalt. Constr. Build. Mater. 2017, 152, 48–56. [Google Scholar] [CrossRef]
  21. Cong, L.; Wang, T.J. Effect of fine aggregate angularity on skid-resistance of asphalt pavement using accelerated pavement testing. Constr. Build. Mater. 2018, 168, 41–46. [Google Scholar] [CrossRef]
  22. Kermani, B.; Stoffels, S.M.; Xiao, M.; Qiu, T. Experimental Simulation and Quantification of Migration of Subgrade Soil into Subbase under Rigid Pavement Using Model Mobile Load Simulator. J. Transp. Eng. B-Pavements 2018, 144, 04018049. [Google Scholar] [CrossRef]
  23. Harvey, J.; Smith, D.R.; Jones, D.J.; Killingsworth, B.; Willis, R.; Li, H. Permeable Pavement Road Map Workshop and Proposed Road Map for Permeable Pavement. UC Davis: National Center for Sustainable Transportation. 2018. Available online: https://escholarship.org/uc/item/1vz9p6gc (accessed on 1 August 2018).
  24. Kim, H.; Sokolov, K.; Poulikakos, L.D.; Partl, M.N. Fatigue Evaluation of Porous Asphalt Composites with Carbon Fiber Reinforcement Polymer Grids. Transp. Res. Rec. 2009, 2116, 108–117. [Google Scholar] [CrossRef]
  25. Arraigada, M.; Pugliessi, A.; Partl, M.N.; Martinez, F. Effect of full-size and down-scaled accelerated traffic loading on pavement behavior. Mater. Struct. 2014, 47, 1409–1424. [Google Scholar] [CrossRef]
  26. Zhang, Y.; Druta, C.; Wang, L.; Xiong, H.; Zhang, W. Dynamic responses of asphalt concrete slab under cyclic wheel loading using acceleration spectrum analysis. Constr. Build. Mater. 2017, 152, 134–144. [Google Scholar] [CrossRef]
  27. Zhou, W.; Huang, X.; Wang, L. Study on the void reduction behaviour of porous asphalt pavement based on discrete element method. Int. J. Pavement Eng. 2017, 18, 285–291. [Google Scholar] [CrossRef]
  28. Sun, Y.; Guo, R.; Gao, L.; Wang, J.; Wang, X.; Yuan, X. Study on Dynamic Response Characteristics of Saturated Asphalt Pavement under Multi-Field Coupling. Materials 2019, 12, 959. [Google Scholar] [CrossRef]
  29. Li, H.; Jones, D.; Wu, R.; Harvey, J. Development and HVS Validation of Design Tables for Permeable Interlocking Concrete Pavement: Final Report; UC Davis, UC Berkeley UCPRC-RR-2014-04.2. 2014. Available online: https://escholarship.org/uc/item/6f97n4tm (accessed on 1 December 2014).
  30. Al-Qadi, I.L.; Dessouky, S.; Tutumluer, E.; Kwon, J. Geogrid mechanism in low-volume flexible pavements: Accelerated testing of full-scale heavily instrumented pavement sections. Int. J. Pavement Eng. 2011, 12, 121–135. [Google Scholar] [CrossRef]
  31. Gokhale, S.; Byron, T.; Iyer, S.; Choubane, B. Evaluation of Pavement Strain Gauge Repeatability. Transp. Res. Rec. 2009, 2094, 30–40. [Google Scholar] [CrossRef]
  32. JTG E20-2011; Standard Test Methods of Bitumen and Bituminous Mixtures for Highway Engineering. Ministry of Transport of the People’s Republic of China: Beijing, China, 2011. Available online: https://xxgk.mot.gov.cn/2020/jigou/glj/202006/t20200623_3312352.html (accessed on 2 December 2011).
  33. JTG/T F20-2015; Technical Guidelines for Construction of Highway Roadbases. Ministry of Transport of the People’s Republic of China: Beijing, China, 2015. Available online: https://xxgk.mot.gov.cn/2020/jigou/glj/202006/t20200623_3312274.html (accessed on 28 May 2015).
  34. ASTM C1701; Standard Test Method for Infiltration Rate of In Place Pervious Concrete. ASTM: West Conshohocken, PA, USA, 2023.
  35. Wu, J.; Ye, F.; Hugo, F. Strain response of a semi-rigid base asphalt pavement based on heavy-load full-scale accelerated pavement testing with fibre bragg grating sensors. Road Mater. Pavement Des. 2015, 16, 316–333. [Google Scholar] [CrossRef]
  36. Zhao, Z.; Xu, L.; Li, X.; Guan, X. Comparative analysis of pavement performance characteristics of flexible, semi-flexible and rigid pavement based on accelerated pavement tester. Constr. Build. Mater. 2023, 387, 131672. [Google Scholar] [CrossRef]
  37. Garcia, G.; Thompson, M.R. Strain and Pulse Duration Considerations for Extended-Life Hot-Mix Asphalt Pavement Design. Transp. Res. Rec. 2008, 2087, 3–11. [Google Scholar] [CrossRef]
  38. Nilsson, R.N.; Oost, I.; Hopman, P.C. Viscoelastic Analysis of Full-Scale Pavements: Validation of VEROAD. Transp. Res. Rec. 1996, 1539, 81–87. [Google Scholar] [CrossRef]
  39. Muraya, P.M.; van Dommelen, A.E. APT testing and visco-elastic analysis of asphalt motorway pavements. Presented at the 2nd International Conference on Accelerated Pavement Testing, Minneapolis, MN, USA, 25–29 September 2004; 27p. Available online: https://trid.trb.org/view/1157018 (accessed on 23 August 2012).
  40. Zhuang, C.; Guo, H.; Zhao, S.; Shu, S.; Ye, Y.; Xing, B. Study on fatigue performance of asphalt mixture in service life based on accelerated loading test. Case Stud. Constr. Mater. 2024, 20, e03055. [Google Scholar] [CrossRef]
  41. Ritter, J.; Rabe, R.; Wolf, A. Analysis of the Long-Term Structural Performance of Flexible Pavements Using Full-Scale Accelerated Pavement Tests. Procedia—Soc. Behav. Sci. 2012, 48, 1244–1253. [Google Scholar] [CrossRef]
  42. Lu, G.; Liu, P.; Törzs, T.; Wang, D.; Oeser, M.; Grabe, J. Numerical analysis for the influence of saturation on the base course of permeable pavement with a novel polyurethane binder. Constr. Build. Mater. 2020, 240, 117930. [Google Scholar] [CrossRef]
  43. Nguyen, H.Q.H.; Maekawa, K. Multi-Scale Simulation for Fatigue Life Evaluation of Concrete Pavement subjected to Moving Load under Dry and Wet Conditions. J. Adv. Concr. Technol. 2020, 18, 95–115. [Google Scholar] [CrossRef]
  44. Hu, M.; Lyu, L.; Pahlavan, F.; Han, P.; Sun, D.; Fini, E.H. Toward Sustainable Non-Emitting Asphalts: Understanding Diffusion–Adsorption Mechanisms of Hazardous Organic Compounds. Adv. Sustain. Syst. 2025, 9, 2400868. [Google Scholar] [CrossRef]
  45. Kermani, B.; Xiao, M.; Stoffels, S.M.; Qiu, T. Measuring the migration of subgrade fine particles into subbase using scaled accelerated flexible pavement testing—A laboratory study. Road Mater. Pavement Des. 2017, 20, 36–57. [Google Scholar] [CrossRef]
  46. Alobaidi, I.; Hoare, D.J. Mechanisms of Pumping at the Subgrade-Subbase Interface of Highway Pavements. Geosynth. Int. 1999, 6, 241–259. [Google Scholar] [CrossRef]
  47. Wang, W.; Wu, Y.; Yang, C.; Feng, Q. Mechanism and influencing factors of mud pumping in railway subgrade. J. Traffic Transp. Eng. 2019, 19, 54–64. [Google Scholar] [CrossRef]
  48. Wu, R.; Harvey, J.; Buscheck, J.; Mateos, A. Mechanistic-Empirical (ME) Design: Mix Design Guidance for Use with Asphalt Concrete Performance-Related Specifications. University of California Paveemnt Research Center, UC Davis, UC Berkeley, CA, USA. 2018. UCPRC-RR-2017-12. Available online: http://www.ucprc.ucdavis.edu/PDF/UCPRC-RR-2017-12.pdf (accessed on 1 August 2018).
  49. Zhao, L. Study on Materials and Structure Design of Urban Permeable Pavement. Master’s Thesis, Chang’an University, Xi’an, China, 2010. [Google Scholar]
  50. Bejarano, M.O.; Thompson, M.R. Subgrade Damage Approach for the Design of Airport Flexible Pavements. In Advancing Airfield Pavements, Presented at the 27th International Air Transportation Conference, Chicago, IL, USA, 5–8 August 2001; American Society of Civil Engineers: Reston, VA, USA, 2001. [Google Scholar]
  51. Lin, D.F.; Luo, H.L.; Wang, H.Y.; Hung, M.J. Successful Application of CLSM on a Weak Pavement Base/Subgrade for Heavy Truck Traffic. J. Perform. Constr. Facil. 2007, 21, 70–77. [Google Scholar] [CrossRef]
  52. Selvi, P. Fatigue and rutting strain analysis on lime stabilized subgrades to develop a pavement design chart. Transp. Geotech. 2015, 2, 86–98. [Google Scholar] [CrossRef]
  53. Tang, X.; Stoffels, S.M.; Palomino, A.M. Mechanistic-empirical approach to characterizing permanent deformation of reinforced soft soil subgrade. Geotext. Geomembr. 2016, 44, 429–441. [Google Scholar] [CrossRef]
  54. Coleri, E.; Harvey, J.T.; Yang, K.; Boone, J.M. Micromechanical investigation of open-graded asphalt friction courses’ rutting mechanisms. Constr. Build. Mater. 2013, 44, 25–34. [Google Scholar] [CrossRef]
  55. Stuart, K.; Mogawer, W.; Aapt, A. Validation of the superpave asphalt binder fatigue cracking parameter using the FHWA’s accelerated loading facility. In Technical Sessions of the Association-of-Asphalt-Paving-Technologists; Association Asphalt Paving Technologists: Colorado Springs, CO, USA, 2002; pp. 116–146. [Google Scholar]
  56. Wu, R.; Harvey, J.; Signore, J. Cracking and Rutting Performance of Composite Pavements Under Testing with Heavy Vehicle Simulator Rubberized Hot-Mix Asphalt with Gap-Graded Aggregates over Portland Cement Concrete. Transp. Res. Rec. 2012, 2304, 177–184. [Google Scholar] [CrossRef]
  57. Cheng, H.; Liu, L.; Sun, L. Determination of Layer Modulus Master Curve for Steel Deck Pavement using Field-Measured Strain Data. Transp. Res. Rec. 2019, 2673, 617–627. [Google Scholar] [CrossRef]
  58. Cheng, H.; Liu, J.; Sun, L.; Liu, L. Critical position of fatigue damage within asphalt pavement considering temperature and strain distribution. Int. J. Pavement Eng. 2020, 22, 1773–1784. [Google Scholar] [CrossRef]
Applsci 15 04359 g001
Figure 1. The model mobile load simulator MMLS3.
Figure 1. The model mobile load simulator MMLS3.
Applsci 15 04359 g001
Applsci 15 04359 g002
Figure 2. Structures of 3 types of permeable pavement.
Figure 2. Structures of 3 types of permeable pavement.
Applsci 15 04359 g002
Applsci 15 04359 g003
Figure 3. The master curve of the porous asphalt mixture.
Figure 3. The master curve of the porous asphalt mixture.
Applsci 15 04359 g003
Applsci 15 04359 g004
Figure 4. Summary of work sequence: (a) steel tank; (b) compacted base; (c) strain gauge; (d) compacted asphalt mixture; (e) the loading device.
Figure 4. Summary of work sequence: (a) steel tank; (b) compacted base; (c) strain gauge; (d) compacted asphalt mixture; (e) the loading device.
Applsci 15 04359 g004
Applsci 15 04359 g005
Figure 5. Plane graph of positions of wheel track and strain gauges.
Figure 5. Plane graph of positions of wheel track and strain gauges.
Applsci 15 04359 g005
Applsci 15 04359 g006
Figure 6. Dry condition: (a) longitudinal and (b) transverse strains versus time (t).
Figure 6. Dry condition: (a) longitudinal and (b) transverse strains versus time (t).
Applsci 15 04359 g006
Applsci 15 04359 g007
Figure 7. Resilient strain under dry conditions: (a) 7PA-CTPB-D and 4PA-CTPB-D; (b) 7PA-CTPB-D and 4PA-CTPB-D in a day; (c) 7PA-CTB-D; (d) 7PA-CTB-D in a day.
Figure 7. Resilient strain under dry conditions: (a) 7PA-CTPB-D and 4PA-CTPB-D; (b) 7PA-CTPB-D and 4PA-CTPB-D in a day; (c) 7PA-CTB-D; (d) 7PA-CTB-D in a day.
Applsci 15 04359 g007
Applsci 15 04359 g008
Figure 8. Wet conditions: (a) longitudinal and (b) transverse strains versus time (t).
Figure 8. Wet conditions: (a) longitudinal and (b) transverse strains versus time (t).
Applsci 15 04359 g008
Applsci 15 04359 g009
Figure 9. Resilient strain under wet conditions: (a) 7PA-CTPB-W and 7PA-CTB-W; (b) 7PA-CTPB-W and 7PA-CTB-W in a day.
Figure 9. Resilient strain under wet conditions: (a) 7PA-CTPB-W and 7PA-CTB-W; (b) 7PA-CTPB-W and 7PA-CTB-W in a day.
Applsci 15 04359 g009
Applsci 15 04359 g010
Figure 10. Rut depth: (a) dry conditions and (b) wet conditions.
Figure 10. Rut depth: (a) dry conditions and (b) wet conditions.
Applsci 15 04359 g010
Applsci 15 04359 g011
Figure 11. Temperature differences and strain: (a) 4PA-CTPB-D and 7PA-CTPB-D; (b) 7PA-CTB-D; (c) 7PA-CTPB-W and 7PA-CTB-W; (d) 4PA-CTPB-D and 7PA-CTPB-D at the beginning of each day; (e) 7PA-CTB-D at the beginning of each day; (f) 7PA-CTPB-W and 7PA-CTB-W at the beginning of each day.
Figure 11. Temperature differences and strain: (a) 4PA-CTPB-D and 7PA-CTPB-D; (b) 7PA-CTB-D; (c) 7PA-CTPB-W and 7PA-CTB-W; (d) 4PA-CTPB-D and 7PA-CTPB-D at the beginning of each day; (e) 7PA-CTB-D at the beginning of each day; (f) 7PA-CTPB-W and 7PA-CTB-W at the beginning of each day.
Applsci 15 04359 g011
Table 1. Details of MMLS3.
Table 1. Details of MMLS3.
MMLS3Specification
Evaluated track length (mm)1260
Tire diameter (mm)300
Tire width (mm)80
Tire pressure (MPa)0.7
Load per tire (kN)2.7
Speed (km/h)7.5
Number of cycles per hour6000
Table 2. Different types of structures and materials.
Table 2. Different types of structures and materials.
TypesSurface LayerBase LayerSubgradeAbbreviation
4PA-CTPBPACement-treated permeable base (CTPB)Dry4PA-CTPB-D
Wet4PA-CTPB-W
7PA-CTPBPACement-treated permeable base (CTPB)Dry7PA-CTPB-D
Wet7PA-CTPB-W
7PA-CTBPACement-treated base (CTB)Dry7PA-CTB-D
Dry7PA-CTB-W
Table 3. Properties of materials.
Table 3. Properties of materials.
SubgradeBaseSurface
TypeClayTypeCTBCTPBTypePA
Compaction (%)90Cement content (%)4.508Asphalt content (%)4.66
Maximum dry density (g/cm3)1.914Optimum water content (%)5.403.8Total porosity (%)17.90
Optimum moisture content (%)9.6Maximum dry density (g/cm3)2.402.104Stability (KN)8.10
Hydraulic conductivity (10−5 cm/s)2.37 days unconfined compressive strength (MPa)4.865.37Raveling loss (%)8.00
--Porosity (%)-20.43Porosity (%)19.21
Table 4. Gradations of surface and base materials.
Table 4. Gradations of surface and base materials.
Sieve (mm)Percent Passing (%)
CTBCTPBPA
31.5100.0100.0100.0
26.5-100.0100.0
1975.592.0100.0
16--100.0
13.2--90.0
9.548.627.079.6
4.7529.013.029.2
2.3620.37.016.0
1.18-5.012.3
0.68.9-7.7
0.3-2.04.6
0.15--3.6
0.0753.51.05.4
Table 5. Model parameters of master curve.
Table 5. Model parameters of master curve.
ParametersδαβγC1C2
Fitting value−2.6717.062−1.719−0.34215.600147.564
Table 6. Typical resilient strain of different types of structures.
Table 6. Typical resilient strain of different types of structures.
TypesLongitudinal Strain (με)Transverse Strain (με)
4PA-CTPB-D2810
7PA-CTPB-D1294
7PA-CTB-D-42
7PA-CTPB-W1050
7PA-CTB-W2836
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Yang, B.; Li, H.; Li, Y.; Cheng, M.; Sun, Y.; Han, Y. Experimental Investigation into Permeable Asphalt Pavement Based on Small-Scale Accelerated Testing. Appl. Sci. 2025, 15, 4359. https://doi.org/10.3390/app15084359

AMA Style

Yang B, Li H, Li Y, Cheng M, Sun Y, Han Y. Experimental Investigation into Permeable Asphalt Pavement Based on Small-Scale Accelerated Testing. Applied Sciences. 2025; 15(8):4359. https://doi.org/10.3390/app15084359

Chicago/Turabian Style

Yang, Bing, Hui Li, Yingtao Li, Murong Cheng, Yang Sun, and Yuzhao Han. 2025. "Experimental Investigation into Permeable Asphalt Pavement Based on Small-Scale Accelerated Testing" Applied Sciences 15, no. 8: 4359. https://doi.org/10.3390/app15084359

APA Style

Yang, B., Li, H., Li, Y., Cheng, M., Sun, Y., & Han, Y. (2025). Experimental Investigation into Permeable Asphalt Pavement Based on Small-Scale Accelerated Testing. Applied Sciences, 15(8), 4359. https://doi.org/10.3390/app15084359

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop