Performance Evaluation on the Application of MAST and RCC on a Cambodian Rural Road: A Case Study

Abstract

Cambodia’s transportation sector has exhibited tremendous growth in the past decades together with its economic advancement. However, these improvements are only focused on the national and provincial roads, leaving the rural roads underdeveloped. A large percentage of the rural roads in Cambodia are unpaved while those paved roads are in deteriorated condition, making these rural roads prone to structural failure when subjected to heavy truck traffic and massive rainfall. Therefore, an innovative pavement material shall be used in the construction and rehabilitation of Cambodian rural roads. In this study, a composite pavement testbed consisting of a roller compacted cement (RCC) base layer and a modified double bituminous surface treatment (DBST), namely multi-layered bituminous surface treatment (MAST), was constructed in Cambodia and its performance was evaluated. To have a point of comparison, the testbed was divided into three sections with varying combinations of surface and base layer: (1) MAST surface with an RCC base, (2) MAST surface with an aggregate base, and (3) DBST surface with an aggregate base. Initially, a visual inspection was conducted to investigate the surface condition of the testbed. To assess the structural capacity, a light-falling weight deflectometer device was used. Meanwhile, the surface roughness was evaluated through the sand patch test. Lastly, the pass-by test was performed to quantify the noise level of the pavement. Based on the results, the proposed composite pavement performed significantly superior to the typical pavement in Cambodia, which is the third section. In summary, the composite pavement with MAST as the surface layer and RCC as the base layer was observed to be suitable for heavy truck traffic loading and the environmental conditions of Cambodia.

1. Introduction

The economic growth of a developing country, such as Cambodia, greatly relies on its national and rural roads that connect the whole country for the exchange of goods. Over the past few decades, Cambodia has made a significant improvement in its transportation system [1,2]. As of 2022, the percentages of pavements covered with Portland cement concrete (PCC), asphalt concrete (AC), or double bituminous surface treatment (DBST) for the national and provincial roads are 84% and 39%, respectively [3]. However, these developmental projects were merely focused on the connectivity of economically strategic areas, such as major cities, in the country. Moreover, national and provincial roads are only a quarter of the whole road network of Cambodia, leaving rural roads, which include the majority of Cambodia’s road network, underdeveloped.

Out of the 48,000 km of rural roads in Cambodia, around 45,600 km, which amounts to 95% of the rural road network of Cambodia, is unpaved [3]. These roads have soil or laterite as a surface layer. Being situated along the Mekong River and Tonle Sap basin, Cambodia’s geographical location makes it prone to periodic river floods and massive rainfall [4]. These unpredictable and inevitable natural hazards that subject the unpaved rural roads to a high volume of water cause a decrease in the road’s structural capacity, making it prone to deformation [5]. In the presence of deformation, the road’s serviceability and ride quality or comfort of the road users are reduced.

Furthermore, the rise of the economy also implies an increase in the traffic on its road network. In the case of rural roads, the number of heavy truck traffic that delivers products from the rural areas to the cities is also increasing [6]. Due to the overloading of these heavy trucks, longitudinal unevenness and permanent deformations may occur on unpaved rural roads [5]. In the long run, rural roads will not be able to support the heavy truck traffic load it experiences. Therefore, improving the unpaved rural roads of Cambodia is necessary.

The improvement that needs to be applied on the unpaved roads shall be able to support the heavy traffic loading and overcome the weakening of the soil due to the high volume of rainfall and floods. Furthermore, being a developing country, the budget is limited to the Cambodian government’s capacity for the construction; thus, choosing a cost-effective material for paving the rural roads is necessary [7,8]. Typical pavement materials, such as reinforced concrete or asphalt, are designed to support heavy traffic loading and extreme environments. However, in the case of Cambodia, paving a lengthy span of rural roads with the typical pavement materials is costly [9,10,11]. Thus, an alternative pavement material with a lower cost than the typical pavement materials but also able to withstand heavy traffic under a tropical environment is needed.

Among the pavement materials, rigid pavement, a specifically reinforced concrete pavement, is most effective in withstanding heavy truck traffic loading [12]. However, due to the restrictions in the construction budget, the construction of reinforced concrete pavement would not be feasible. Therefore, a cheaper alternative to reinforced concrete pavement shall be used, which is the roller compacted concrete (RCC).

Unlike reinforced concrete pavement, RCC is a non-reinforced type of concrete pavement and, as stated in its name, is compacted using a vibratory roller [13]. Having no reinforcements needed, RCC is definitely cheaper, simpler, and less time-consuming to construct than the typical reinforced concrete pavement without compromising its structural strength. Moreover, the absence of reinforcements makes the RCC more appropriate for high moisture environmental conditions [14]. Furthermore, RCC has a lower water-to-cement ratio, which gives it the advantage of lowering the amount of Portland cement compared to the typical concrete pavement [15]. The reduction in the Portland cement in the mix design makes RCC more cost-efficient than the typical concrete pavement. However, due to RCC’s lower water-to-cement ratio, the surface of the pavement tends to be rougher than that of the typical concrete pavement, which makes it unsuitable for general roads [16]. Hence, a pavement material suitable as a surface course for RCC shall be applied.

The typical surface course for a composite pavement with cement concrete as the base course is AC [17]; although, using this material for Cambodian rural roads will amount to high construction expenses. A cheaper alternative for the surface course is DBST, which is commonly used in Cambodia [3]. DBST promotes the appropriate skid resistance for a typical surface course and serves as a layer for waterproofing the layers underneath [18]. However, general DBST alone is prone to early stages of wearing, and if there is an insufficient adherence in the binder–aggregate interface, aggregate stripping may occur. Improvements in the general DBST shall be applied to make it more suitable for Cambodian rural roads.

In 2005 [19], North Carolina DOT observed that the enhancement in the gradation of aggregate used in the bituminous surface treatment from finer-graded to coarser and more uniformly graded minimizes aggregate loss. That study suggests an aggregate gradation of larger than 2.36 mm with less than 1.5% of fine aggregates less than 0.075 mm. Moreover, there is a study regarding the modification of asphalt emulsion used in the bituminous surface treatment by adding polymer to increase the mixture’s resistance to rutting [20]. It was found that polymer-modified emulsions enhance the rutting resistance of the mixture at high temperatures, reduce the loss of aggregate coverage, and prolong the life by at least two years more compared to the unmodified emulsion. For the improvement of DBST for Cambodia, these factors were considered and the final mixture was coined multi-layered asphalt surface treatment (MAST).

Before the implementation of the MAST-RCC composite pavement, the cost-effectiveness of the proposed pavement system was validated in 2022 [21]. A simple life-cycle cost analysis was conducted on the MAST-RCC composite pavement and the typical AC pavement. A developed analysis program by the Asphalt Paving Association was used, and the theories and calculations were based on FHWA procedures. The analysis was set to a 30-year service life for both pavements, and the prices for construction, maintenance, rehabilitation, and user costs were considered. The periods at which the rehabilitation and maintenance will be implemented on the MAST-RCC pavement were set to 10 years and 4 years, respectively, while for the AC pavement, the periods were set to 15 years and 5 years, respectively. Furthermore, the costs of the rehabilitation and maintenance of the MAST-RCC pavement were determined to be 42% and 5.1% of the initial construction cost, respectively. Meanwhile, the costs of the rehabilitation and maintenance of the typical AC pavement were set to 42% and 5.1% of the initial construction cost, respectively. After running the analysis, it was found that the 30-year life-cycle cost of MAST-RCC pavement is lower by more than 30% compared to that of the typical AC pavement. This finding showed that the MAST-RCC composite pavement is a viable option for a more cost-effective pavement material. However, the performance of the proposed composite pavement on the actual field condition of the rural roads of Cambodia must be investigated.

In this study, the feasibility of the MAST-RCC composite pavement system for rural roads in Cambodia was evaluated. Initially, laboratory tests were conducted to achieve the best mix design for RCC and MAST. To observe the RCC and MAST’s in situ performance, a 200-m testbed was constructed on a typical rural road in Phnom Penh, Cambodia. The testbed was divided into three different sections: MAST surface layer with an RCC base course, MAST surface layer with a general aggregate base course, and normal DBST with a general aggregate base course. Three months after opening the testbed to traffic, tests were conducted to assess each section’s performance. Firstly, a visual inspection was performed to observe the surfaced distresses present in each section. Secondly, a light-falling weight deflectometer (LFWD) was used to evaluate each section’s structural condition. Thirdly, a sand patch test was conducted to assess the surface roughness of the pavement. Lastly, since rural roads are also located near residential areas in Cambodia, the noise level produced by the tire–pavement interaction on MAST was compared to the typical concrete pavement present in most of the residential areas in Cambodia using the pass-by test. The general flow of this research is presented in Figure 1.

2. Materials and Methods

2.1. Materials

2.1.1. Multi-Layered Asphalt Surface Treatment (MAST)

Bituminous surface treatment, also known as chip seal, is a low-cost surfacing material commonly used in road maintenance and is a low-budget surface layer with a structurally reliable base layer. Basically, it is a layer of aggregates coated with emulsified asphalt as a binding agent [22]. In Cambodia, two layers of bituminous surface treatment, also known as DBST, is commonly used. However, DBST in Cambodia is prone to early surface wearing due to insufficient adherence between the binder–aggregate interface. Therefore, a modified asphalt emulsion is used as a binding agent for the DBST, called the modified mixture MAST.

In this study, to improve the performance of MAST, latex and glass fibers were used as an additive to the mixture. Latex is a type of polymer additive that improves the elasticity of the asphalt emulsion, making it more resistant to aggregate stripping [23]. Meanwhile, glass fibers increase MAST’s resistance to moisture damage [24]. MAST is developed to improve DBST’s resistance to surface wearing and aggregate stripping.

The gradation for the aggregate used in MAST is shown in

Table 1. Percent by weight passing of MAST aggregates.

Nominal Size of Material	Percent by Weight Passing AASHTO Sieve Sizes
	37.5 mm	25.0 mm	19.0 mm	12.5 mm	9.5 mm	4.75 mm	2.36 mm	1.18 mm
25.0 mm	100	90–100	0–45	0–10	0–5	-	0–2	0–0.05
19.0 mm		100	90–100	0–30	0–8	-	0–2	0–0.05
12.5 mm			100	90–100	0–40	0–8	0–2	0–0.05
9.5 mm				100	90–100	0–30	0–2	0–2.00

. The amount of glass fibers is between 60 g/m² and 12 g/m² while the amount of latex is between 3% and 5% of the weight of the emulsified bitumen used. The amount of asphalt emulsion and aggregates on the first layer of application was set to 1.15 kg/m² and 15.2 kg/m², respectively. Meanwhile, for the second application, the amount of asphalt emulsion and aggregates was set to 1.55 kg/m² and 6.15 kg/m², respectively. These values on the application of MAST were in accordance with the construction standards of the Kingdom of Cambodia [25].

To assess the MAST’s resistance to aggregate stripping in the laboratory, a simple sweep test was conducted and additional DBST specimens were included for comparison [26]. At 3-h curing time, the percent mass loss of MAST was 1.4% while the percent mass loss of DBST was 2.9%, which means that the aggregate loss on the surface of MAST specimen was 50% less than the aggregate loss of DBST. Figure 2 shows the comparison of the appearance of the surface of MAST and DBST after the sweep test.

2.1.2. Roller Compacted Concrete (RCC)

Roller compacted concrete (RCC) is basically a traditional concrete mixture with a slightly adjusted mixture design to achieve a zero-slump. The components of RCC are the same as the traditional concrete mixtures but with reduced water content, making it stiff enough to support itself while compacting it using a roller. Unlike typical concrete structures, RCC is constructed in the absence of forms, dowels, joints, or any reinforcement. In recent years, RCC has been used in highly loaded infrastructures and the most notable is its use in dams. Its usage in dams significantly reduced the time of construction, making it more efficient and economical. Moreover, RCC has been present in the construction of horizontal infrastructures that include heavy loading, such as haul roads, loading docks, intermodal port facilities, parking lots, airports, and, recently, highways [27].

In this study, the aggregates used are a combination of coarse and fine aggregates to achieve a uniform particle size distribution with close proximity to the Fuller curve. The aggregate gradation is shown in Figure 3, which is in accordance with the RCC gradation limit of ACI PRC-327-14 [28]. The target water-to-cement ratio of the mix design is 48%, as calculated from the mix design by unit weight summarized in

Table 2. RCC mix design by unit weight.

	Water	Cement	Fine Aggregate	Coarse Aggregate	Admixture
Unit Weight (kg/m3)	132.0	274.3	913.4	1120.4	1.9

.

Using the gradation and mix design stated, laboratory specimens with a curing age of 28 days were fabricated to determine the RCC’s compressive strength and chloride ion permeation resistance using the laboratory tests shown in Figure 4. Based on the results, the values of the compressive strength and chloride ion permeation resistance were 39.4 MPa and 2813 Coulomb, respectively, which classified the RCC as a normal grade.

2.2. Testbed Details

The testbed was constructed on the rural road just before the boundary of Phnom Penh. Figure 5 shows an aerial view of the testbed location, as well as the span of the road where the testbed was constructed. The testbed’s total length was 200 m and was divided into three sections as shown in Figure 6. The first section with a length of 100 m was the proposed composite pavement, which consisted of a MAST surface layer and an RCC base layer. The second section was a MAST surface layer with an aggregate base layer and had a length of 50 m. Lastly, the third section was a 50-m long typical DBST surface layer and aggregate base layer. The third section served as the point of reference in this study. The thicknesses of the surface and base layers of all the sections were 10 mm and 150 mm, respectively, which were in accordance with the typical road structure in Cambodia [25].

2.3. Methods

2.3.1. Light-Falling Weight Deflectometer

Light-falling weight deflectometer (LFWD) is a non-destructive pavement testing device that is mainly used to determine the stiffness modulus of the pavement or soil from the deflection caused by the application of an impulse load. It is also used to determine the compaction degree and evaluate the bearing capacity of the pavement or soil for road construction and earthworks. The device is derived from the static plate loading test and improved the testing procedure by replacing the static load with dynamic load, which closely represents the traffic loading generated by passing vehicles [29,30,31].

In this study, the test procedure using the LFWD was performed in accordance with ASTM E2583-07 [32], and Figure 7 shows the specific device used in the test, Dynatest LWD 3031. The LFWD device is composed of several parts:

• Guide rod;
• Release handle;
• Drop weight;
• Rubber buffer;
• Loading plate;
• Load cell;
• Velocity transducer.

The guide rod connects all of the parts into one device. In the initial phase of the testing, the release handle holds the drop weight on top of the guide rod with a magnetic plate underneath and a security lock. In the loading phase of the test where the drop weight is released, the drop weight makes an impulse load that is received by the rubber buffer. The presence of a rubber buffer makes the lower part of the device a spring-damper system [30]. The impulse load received by the rubber buffer is transferred to the pavement or soil through the loading plate. The load cell records the impulse load created by the drop weight while the deflection due to the impulse load is obtained through the velocity transducer underneath the loading plate.

To calculate the stiffness modulus of the testing point, the Boussinesq half-space equation is used [33]:

$$E_{LFWD}=\frac{2(1-{\mu }^2)\rho R}{s},$$

where $E_{LFWD}$ is the stiffness modulus of the testing location in MPa, $\mu$ is the Poisson ratio, $\rho$ is the applied stress on the testing surface in kPa, R is the plate radius in mm, and s is the deflection in micron.

The LFWD test can be conducted by only one person, but for ease of testing, three people are highly suggested. While testing, the guide rod should be as perpendicular to the surface as possible to avoid variations in the impulse load due to inclination. Moreover, the device should be as firm as possible to the ground and the surface should be flat to achieve a uniform loading to the whole area of the loading plate. Lastly, the load cell and deflection sensor shall be periodically calibrated to make sure that the results are accurate [33].

2.3.2. Sand Patch Test

Sand patch test was conducted to quantify the roughness or macrotexture of the pavement surface. The pavement’s surface texture correlates to the pavement’s skid resistance, tire-pavement noise, and tire wearing. Moreover, it also affects the road user’s ride quality and safety. A highly deteriorated pavement with a very smooth surface is prone to slippage, splash, and spray, especially after heavy rain or floods [34].

In this study, the sand patch test was conducted according to ASTM E965-15 standards [35], and the test procedure is presented in Figure 8. Basically, the sand patch test determines the pavement surface’s roughness by volumetric approach. A known volume of sand was placed on the clean and dry pavement surface. The sand was gradually spread on the pavement surface in a circular manner using a smooth and flat disk until the entire amount of sand was spread evenly on the pavement surface. Assuming that the sand uniformly fills the full depth of the crevices on the pavement surface, at least four diameters of the formed circle were measured on different angles, and the average diameter was used to calculate the mean text depth (MTD) as follows:

$$MTD=\frac{4V}{\pi D^2},$$

where V is the volume of sand in mm³ and D is the average diameter of the circular area covered by the sand in mm.

In this procedure, it is important that the pavement surface is thoroughly cleaned and dried before testing. Moreover, the sand patch test is not applicable on pavement surfaces with visible cracks. The presence of debris, water, and cracks on the pavement surface affects the volume of space the sand needs to fill on the pavement surface, resulting in values not representative of the pavement’s roughness [36].

2.3.3. Pass-By Test

Pass-by test aims to measure the noise level of a passing vehicle at a certain point at the roadside. This testing can be utilized to analyze the sound propagation generated by a certain vehicle at a specific speed. Through this, the comparison of sound levels from different types of vehicles, tire properties, and speed can be conducted [37]. Furthermore, it can also be used to compare the noise mitigation properties of different types of pavement materials.

For this study, the pass-by test was conducted according to EN ISO 11819-1 standard [38]. As shown in Figure 9, the pass-by test consisted of a microphone that was placed 7.5 m from the middle of the vehicle path with a height of 1.2 m and a vehicle driven at a uniform speed [39]. The test should be repeated at least three times for a single location and vehicle speed to observe precision among the results. Due to the pass-by test’s simplicity, this test is the most commonly used among the noise test methods.

In conducting the pass-by test, a controlled environment is needed to minimize the unnecessary noise from different sources other than the interaction of the vehicle to the pavement and the vehicle engine. Moreover, there shall be no obstructions around the microphone that may hinder the travel of sound from the road to the microphone, such as trees, bushes, and traffic signs. The maximum sound level in one vehicle pass is the representative result of each test. Under the assumption of a constant speed with the passing vehicle, a log-linear trend shall be observed in the maximum sound level with respect to the vehicle speed [40].

3. Results and Discussion

After three months of opening the testbed to traffic, the performance of MAST and RCC was evaluated. The surface condition of the pavement was observed through visual inspection and the sand-patch test. Moreover, the structural condition of the pavement was assessed using the LFWD test. Lastly, the noise level of the pavement was determined by conducting the pass-by test.

3.1. Visual Inspection

To assess the current surface condition of the testbed, a visual inspection was conducted. The three sections of the testbed were thoroughly inspected and Figure 10 shows the surface condition of each section.

It can be observed that a significant amount of depression is present in Section 3 while Sections 1 and 2 still maintain an evenly leveled surface. These depressions were mainly caused by the heavy truck traffic, shown in Figure 11 that passes on the testbed. The location of the testbed is on the outskirts of Phnom Penh, which is commonly the route of the heavy truck that carries goods from rural areas to the city and vice versa. The structural capacity of Section 3 was not sufficient to support the high frequency of the heavy loading of the truck. Moreover, the tropical climate of Cambodia weakened the aggregate base layer of the pavement. Frequent rainfall and poor drainage promote a long duration of water retention in the aggregate base, thus decreasing its structural capacity.

The RCC base layer in Section 1 increased the stability of the underlying layer, making it less prone to deformation. Meanwhile, even though Section 2 has the same aggregate base layer as Section 3, the MAST layer on top of the aggregate base layer assisted the whole pavement system in making it intact due to a stiffer binder than that of the general DBST. This proves that replacing the typical DBST with a MAST surface layer improves the performance of the pavement. It served as an effective waterproofing layer to the other pavement layers underneath. Moreover, the RCC base layer is more suitable for heavy truck traffic than the typical aggregate base layer, especially in an environment with frequent heavy rainfall, such as tropical countries.

Furthermore, edge deterioration is present in all of the sections of the testbed, as shown in Figure 12. As observed in Figure 10, the whole span of the testbed has an unbounded edge. The lack of a concrete curb or any stable edge that prevents the deformation of the pavement on its side makes the pavement prone to edge deterioration. Moreover, the settlement at the edge of the pavement was also intensified by the absence of a proper drainage system at the testbed site. As mentioned earlier, Cambodia is a tropical country with a tremendous amount of rainfall. The retention of water on and under the pavement weakens the edge of the pavement, thus forming a settlement. As the deteriorated edge of the pavement is subjected to cycles of rainfall and water retention, its structural condition worsens making it easier to be damaged.

3.2. LFWD Test

Three testing points were chosen for each section, and these points are evenly spaced to represent the whole span of the section. In Section 3, one point was intentionally located on the depression to investigate the structural condition of the deteriorated location. At each point, three repetitions of LFWD testing were conducted. The stiffness modulus per repetition was calculated using the Boussinesq half-space equation, and the average value of the three stiffness moduli was considered as the representative stiffness modulus of the testing point. Figure 13 shows the values of the representative stiffness modulus of the testing points in all the sections.

As observed in the LFWD test results, all the testing points in Section 1 exhibit a significantly high stiffness modulus compared to Sections 2 and 3, ranging from 450 MPa to 650 MPa. This is expected due to the structural strength contributed by the RCC on the pavement system. Moreover, the dominant stiffness modulus of Section 1 correlates to the absence of noticeable surface distress or deformation observed in the visual inspection.

Meanwhile, the stiffness moduli for Sections 2 and 3 are almost similar, ranging from 100 MPa to 150 MPa, with the exception of the point in Section 3 with the depression having a stiffness modulus of less than 100 MPa. Having the aggregate base, it is expected that Sections 2 and 3’s stiffness moduli are in the same range.

Moreover, the lowest value of the stiffness modulus on the point of depression shows the reason behind the deterioration in that area. With a low stiffness modulus and a high traffic volume to support, the pavement is most likely to fail given those conditions.

In summary, it is evident that the main factor for the stiffness modulus in this testbed is the base layer. A base layer with a higher stiffness modulus, like the RCC, is more suitable for heavy traffic than the base layer with a lower stiffness modulus, like the aggregate base.

3.3. Sand Patch Test

To assess the surface roughness of each section of the testbed, the sand patch test was performed. One testing point per section was chosen, and a uniform volume of sand amounting to 5000 mm³ was used for the test. Four diameter readings were obtained, and the average diameter was calculated to compute the MTD. The results of the sand patch test are summarized in Figure 14.

According to the principle of the sand patch test, the circular area covered by the sand on a surface with low roughness is bigger since the sand has fewer crevices to fill in. Meanwhile, surfaces with high roughness tend to form circular areas with smaller diameters since more spaces should be filled in by the sand. In the results of the sand patch test, it is evident that Section 3, having the highest average diameter, has the lowest surface roughness among the three sections since it is the only section with a general DBST surface layer. Meanwhile, Sections 1 and 2 have average diameters closer to each other with measurements of 198.8 mm and 227.5 mm, respectively, since both sections have the same MAST surface.

Based on [41], the acceptable MTD for bituminous surface treatment is at least 0.7. Anything lower than that would not be sufficient to provide appropriate traction to the tire of the vehicle. As shown in the results, Sections 1 and 2 performed well in the sand patch test with values higher than 0.7. Meanwhile, the surface roughness in Section 3 failed the sand patch test, with a MTD value of 0.4. It is implied that the surface roughness of the general DBST declines abruptly in just a few months of exposure to traffic.

Both the results on average diameter and MTD suggest that MAST is significantly superior to the general DBST in terms of surface roughness after being subjected to several months of traffic loading. The poor surface roughness of DBST was caused by the degradation due to the insufficient strength of the pavement system, as well as the low resistance to aggregate stripping based on the sweep test conducted in the laboratory. Therefore, the latex and glass fibers additive of the MAST surface layer effectively increased the material’s resistance to surface wear. Given that information, it is more appropriate to use the MAST surface layer on Cambodian rural roads since it can withstand heavy truck traffic loading without compromising its surface roughness.

3.4. Pass-By Test

The future utilization of MAST in residential areas of Cambodia is also being taken into consideration. Thus, the noise level that it produces should be taken into account. In this study, the noise level of a typical concrete pavement in the residential area of Cambodia was compared to that of the MAST pavement. Figure 15 shows the MAST pavement and concrete pavement where pass-by tests were conducted.

For each location, three vehicle speeds were tested: 40 kph, 50 kph, and 60 kph. The ambient noise was also measured for each location and deducted from the testing results to isolate the noise produced by the tire–pavement interaction and the vehicle engine. The final values of the noise level at different vehicle speeds are summarized in Figure 16.

Based on the results, the noise level of the MAST pavement is lower than that of the concrete pavement by at least 4 dB in all of the testing vehicle speeds. Therefore, the MAST pavement is an appropriate replacement for concrete pavement in residential areas of Cambodia in terms of noise level.

4. Conclusions

This study focused on determining the performance of MAST as a surface layer and RCC as a base layer for Cambodian rural roads. To evaluate the proposed pavement material, a testbed on a rural road in Phnom Penh was constructed. The testbed consisted of three sections with varying surface layers and base layers: (1) MAST surface layer with an RCC base layer, (2) MAST surface layer with an aggregate base layer, and (3) DBST surface layer with an aggregate base layer. The third section served as the control in the experiment since it is the typical pavement material used in Cambodia. The testbed was subjected to traffic load for several months and the condition of the pavement after opening it to traffic was assessed. A visual inspection was performed to investigate the surface condition of the pavement and take into account the notable surface distresses present on the pavement. For the structural capacity of the pavement, the stiffness modulus of each section was determined using the LFWD device. Moreover, the surface roughness of the pavement was evaluated using the sand patch test. Lastly, the noise level of MAST as a surface layer was compared to the typical concrete pavement in residential areas of Cambodia using the pass-by test. The following are the conclusions drawn from this study:

• Through the visual inspection of the sections, it was found that the section with the typical rural road design in Cambodia, which is the DBST surface layer with an aggregate base layer, is susceptible to depression. Meanwhile, the other two sections with MAST as the surface layer preserved its evenly leveled surface. For Section 1, the RCC base layer sustained the pavement system’s structural stability under heavy truck traffic loading. Moreover, its resistance to moisture damage is exhibited in this study with the help of MAST as a waterproofing layer on the surface.
• Edge deterioration was also observed in all the sections of the testbed. This is due to the unbounded nature of the testbed all throughout its span. The lack of concrete curbs or any edge stabilizing structures causes the sideways deformation of the pavement, resulting in settlement and water retention at the edge. This problem must be addressed in future constructions of testbed or actual roads in Cambodia.
• The LFWD test results confirmed the structural strength of the RCC base layer compared to the aggregate base layer. The results showed that all the testing results in Section 1 are significantly higher than that of Sections 2 and 3, with values ranging from 450 MPa to 650 MPa. Meanwhile, having the same aggregate base layer, the stiffness modulus of the points obtained in Sections 2 and 3 are in close proximity to each other, with values ranging from 100 MPa to 150 MPa.
• Moreover, the depression in Section 3 was also included in the LFWD testing points to investigate the reason behind the failure. The results correlated with the visual inspection with the depression point having the lowest stiffness modulus among all the testing points. The stiffness modulus at the depression was under 100 MPa. Given these findings, the combination of low structural capacity, heavy traffic loading, and massive rainfall was found to be the most probable reason for the occurrence of the depressions in Section 3.
• The sand patch test was conducted on the three sections to compare each section’s surface roughness after experiencing traffic for several months. Based on the results, Sections 1 and 2, both with a MAST surface layer, sustained their surface roughness and are still suitable for traffic. However, the DBST surface layer of Section 3 was deemed to be susceptible to surface wearing with an MTD lower than 0.7. This implies that the addition of latex and glass fiber to the MAST effectively elevated the material’s resistance to surface wearing.
• The feasibility of MAST being an alternative pavement material for Cambodian residential areas in terms of noise levels was investigated. The pass-by test was performed on both the MAST pavement and concrete pavement to compare the noise levels. It was found that in all three vehicle speeds tested, the noise level of the MAST pavement is lower by at least 4 dB compared to the noise level produced by the concrete pavement. Therefore, it can be concluded that MAST is an appropriate alternative in terms of noise levels for pavements in Cambodian residential areas.
• In summary, the MAST surface layer and RCC base layer showed promising results in the evaluation of the testbed in the rural roads of Cambodia. Further tests are suggested to fully investigate the feasibility of the proposed composite pavement as an alternative for Cambodian rural roads.