Deep Learning-Based Thermal Image Analysis for Pavement Defect Detection and Classification Considering Complex Pavement Conditions

Automatic damage detection using deep learning warrants an extensive data source that captures complex pavement conditions. This paper proposes a thermal-RGB fusion image-based pavement damage detection model, wherein the fused RGB-thermal image is formed through multi-source sensor information to achieve fast and accurate defect detection including complex pavement conditions. The proposed method uses pre-trained EfficientNet B4 as the backbone architecture and generates an argument dataset (containing non-uniform illumination, camera noise, and scales of thermal images too) to achieve high pavement damage detection accuracy. This paper tests separately the performance of different input data (RGB, thermal, MSX, and fused image) to test the influence of input data and network on the detection results. The results proved that the fused image’s damage detection accuracy can be as high as 98.34% and by using the dataset after augmentation, the detection model deems to be more stable to achieve 98.35% precision, 98.34% recall, and 98.34% F1-score.

EVALUASI KERUSAKAN PERKERASAN LENTUR DENGAN METODE PCI DAN SDI (STUDI KASUS: JALAN JATISARI, KARAWANG)

Large vehicles that repeatedly pass a road cause damage to the pavement of the Jatisari National Road, Karawang. Various pavement damage that occurs such as holes, patches, crocodile skin cracks, groove cracks, sungkur, roadside cracks, and subsidence. Pavement Condition Index (PCI) is a method commonly used to indicate the condition of road pavement, so that it can be known good handling to maintain the pavement. The Surface Distress Index (SDI) method can also be used to indicate the condition of the road surface. With the PCI method, the results of the calculation in the Pamanukan direction are classified as perfect at 78%, very good 14%, good 4% and moderate 4%. while the Cikampek direction is classified as perfect at 74%, very good 12%, good 8%, moderate 4%, and bad 2%. Using the SDI method, good results were obtained for both directions. Based on the results of the analysis, research using the PCI and SDI methods showed different results, because the PCI method observed all the damage that occurred on the pavement, while the SDI method only observed 4 elements of damage, so the results displayed were different. ABSTRAKKendaraan besar yang berulang kali melewati sebuah jalan menyebabkan kerusakan pada perkerasan Jalan Nasional Jatisari, Karawang. Berbagai Kerusakan perkerasan yang terjadi seperti, lubang, tambal, retak kulit buaya, retak alur, sungkur, retak tepi jalan, dan amblas. Pavement Condition Index (PCI) merupakan metode yang biasa digunakan untuk menunjukkan kondisi perkerasan jalan, sehingga bisa diketahui penanganan yang baik untuk memelihara perkerasan jalan tersebut. Selain itu, digunakan metode Surface Distress Index (SDI) untuk menunjukkan kondisi permukaan jalan. Dengan Metode PCI, hasil perhitungan pada arah Pamanukan digolongkan sempurna sebesar 78%, sangat baik 14%, baik 4% dan sedang 4%. sedangkan pada arah Cikampek digolongkan sempurna sebesar 74%, sangat baik 12%, baik 8%, sedang 4%, dan buruk 2%. Dengan metode SDI, diperoleh hasil Baik untuk kedua arah jalan. Berdasarkan hasil analisis, penelitian menggunakan metode PCI dan SDI menunjukkan hasil yang berbeda, dikarenakan dalam metode PCI mengamati semua kerusakan yang terjadi pada perkerasan jalan, sedangkan untuk metode SDI hanya mengamati 4 unsur kerusakan, sehingga hasil yang ditampilkan berbeda.

THE EFFECT OF SEA WATER IMMERSION ON BUTON ASPHALT (AS-BUTON) MIXTURE

One of the causes of flexible pavement damage is being submerged in water with a high acidity, such as seawater. To overcome this condition, a study on the use of buton asphalt (as-buton) was conducted. The advantages of as-buton are that it has higher adhesion and is more resistant to temperature changes. As-buton modification by mixing Lawele as-buton Southeast Sulawesi with oil asphalt pen 60/70. This study focuses on determining the effect of the addition of as-buton to the modified asphalt mixture and the effect of seawater immersion on as-buton mixture based on the Marshall characteristic test. The percentage by weight of as-buton used is 5%, 7.5%, and 10%. The immersion was carried out in freshwater and seawater immersion for up to 24 hours at a temperature of 30°C. The result of this study show that the addition of as-buton increase the value of stability, void in mixture, Marshall quotient (MQ), and void in mineral aggregate, but reduce the flow, void filled with asphalt, and density. The increase of stability and MQ value in seawater-immersion due to the addition of as-buton was higher than that in freshwater. The stability value of the 10% modified as-buton mixture in freshwater-immersion increased by 169.338 kg (14.61%) and in seawater-immersion increased by 1261.669 kg (55.65%). The flow value in freshwater-immersion decreased by 1.9 mm (25.33%) while in seawater-immersion decreased by 1 mm (14.08%). For further research, variations in the addition of as-buton will be carried out to produces a modified asphalt mixture.

Truck-Platooning Impacts on Flexible Pavements: Experimental and Mechanistic Approaches

Truck platoons are expected to improve safety and reduce fuel consumption. However, their use is projected to accelerate pavement damage due to channelized-load application (lack of wander) and potentially reduced duration between truck-loading applications (reduced rest period). The effect of wander on pavement damage is well documented, while relatively few studies are available on the effect of rest period on pavement permanent deformation. Therefore, the main objective of this study was to quantify the impact of rest period theoretically, using a numerical method, and experimentally, using laboratory testing. A 3-D finite-element (FE) pavement model was developed and run to quantify the effect of rest period. Strain recovery and accumulation were predicted by fitting Gaussian mixture models to the strain values computed from the FE model. The effect of rest period was found to be insignificant for truck spacing greater than 10 ft. An experimental program was conducted, and several asphalt concrete (AC) mixes were considered at various stress levels, temperatures, and rest periods. Test results showed that AC deformation increased with rest period, irrespective of AC-mix type, stress level, and/or temperature. This observation was attributed to a well-documented hardening–relaxation mechanism, which occurs during AC plastic deformation. Hence, experimental and FE-model results are conflicting due to modeling AC as a viscoelastic and the difference in the loading mechanism. A shift model was developed by extending the time–temperature superposition concept to incorporate rest period, using the experimental data. The shift factors were used to compute the equivalent number of cycles for various platoon scenarios (truck spacings or rest period). The shift model was implemented in AASHTOware pavement mechanic–empirical design (PMED) guidelines for the calculation of rutting using equivalent number of cycles.

METHODS AND CRITERIA FOR ASSESSING THE CONDITION OF RIGID PAVEMENT CONTAINING CRACKS

Road agencies around the world note the aging of the national road network and the accelerated destruction of pavement during operation due to changes in the composition of the traffic. Many scientific studies have noted that cracks of various configurations, depths, origins are the most characteristic type of pavement damage, make up a significant part of the total area of destruction and are the main danger as the initial stage of more serious damage leading to loss of pavement distribution. Timely detection of hidden cracks in the coating layers and the base layers allows to prevent the destruction of pavement and significantly reduce the cost of road repairs. The purpose of the article is to analyze the methods and criteria for assessing the condition of non-rigid pavement, which contains hidden defects in its structural layers. The results of the analysis make it possible to take into account a particular type of defect when choosing a calculation scheme and the order of calculation of pavement. The analysis proved that the calculated schemes of pavement design do not allow to take into account such indicators of pavement condition as the number of cracks, their position, geometric parameters of hidden cracks, sizes of cracked base blocks, volumetric damage of the layer material. The assumptions and simplifications accepted in the normative documents actually make it impossible to assess the condition and predict the failure of the design of paved roads with cracks. Therefore, the assessment of the condition of non-rigid pavement should be based on probabilistic methods and criteria for assessing the condition of pavement, which allow to take into account the heterogeneity of pavement design, due, inter alia, the presence of cracks in layers of monolithic materials. The next stage of research is the analysis of existing probabilistic methods of designing and assessing the condition of pavement.

Design and Mechanical Properties of Steel-UHPC Lightweight Composite Decks

Orthotropic steel decks (OSDs) have been widely used in long-span bridges due to their advantages of being lightweight, having high capacity, and allowing rapid construction. However, due to the insufficiency of local stiffness of OSD, fatigue cracking and pavement damage have been common problems of OSDs worldwide. It seriously affects the safety and durability of long-span bridges. Therefore, to solve this problem, this paper introduces an innovative steel ultrahigh-performance concrete (steel-UHPC) lightweight composite deck (LWCD). LWCD can reduce the fatigue stress of the conventional OSD by up to 80% and extend the fatigue life to twice the design requirements. Furthermore, engineering practices in China have proven that LWCD can effectively reduce manufacturing costs and maintenance costs throughout the whole life cycle of the structures. Thus, to provide references for design and maintenance of long-span bridges, this paper introduces the structural design, construction techniques, joint construction design, repair methods, and economic benefits of LWCD in detail. Furthermore, numerical simulations and laboratory tests are introduced in this paper to validate the superiority of LWCD.

Heat Transfer Characteristics of Asphalt Mixtures with Different Thicknesses

Generally, the main causes of asphalt pavement damage are repeated loads and increased temperatures. The surrounding environment and material factors influence the degree of damage. In this study, the heat transfer characteristics with a change in the asphalt pavement thickness were analyzed. The sample types were the conventional dense asphalt mixture (WC-2), dense asphalt mixture containing 30% recycled aggregate (ReWC-2), and drainage asphalt mixture (PA-13). The mixtures were compacted using a rotary compactor. Thermal accumulation tests on samples with thicknesses of 3, 5, and 10 cm were performed. The values for the 10-cm-thick samples were 1.27 for WC-2, 1.28 for ReWC-2, and 0.91 for PA-13. The thermal conductivity tended to decrease as the pavement thickness increased. An experimental study on the heat transfer characteristics with varying thicknesses of asphalt pavement suitable for energy harvesting was conducted.