Search

Pavement friction has been identified as crucial in traffic safety. Since the Highway Safety Manual prediction algorithm is often based on crash frequency, the crash severity distribution might be assumed unchanged before and after the countermeasure. However, pavement surface treatments can improve the friction to different levels, by which crash severity outcomes may vary greatly. To explore the implicit effects of pavement friction on vehicle crash severity, this paper first validates the extreme gradient boosting model performance and then the Shapley additive explanations interaction values are employed to interpret individual features and the nonlinear interactions among predictors. Under various scenarios, the extreme gradient boosting (XGBoost) output probability is utilized to convert into dynamic crash severity distributions. Results also indicate that friction becomes more significant when the friction number is less than 38, and immediate corrective actions are needed when the friction number is below 20. Key words: pavement friction, crash severity, highway safety, extreme gradient boosting (XGBoost), Shapley additive explanations (SHAP). Le frottement des chaussees a ete juge crucial pour la securite routiere. Etant donne que l&apos;algorithme de prediction du Manuel de Securite routiere est souvent fonde sur la frequence des collisions, la distribution de la gravite des collisions peut etre consideree comme inchangee avant et apres la contre-mesure. Par contre, les traitements de surface de chaussee peuvent ameliorer le frottement a differents niveaux, par lesquels les resultats de gravite des collisions peuvent varier considerablement. Explorer les effets implicites du frottement de la chaussee sur la gravite des collisions, cet article valide d&apos;abord les performances du modele d&apos;amplification de gradient extreme (<< XGBoost >>), puis les valeurs d&apos;interaction des explications additives de Shapley (SHAP) sont utilisees pour interpreter les caracteristiques individuelles et les interactions non lineaires entre les predicteurs. Dans divers scenarios, la probabilite de sortie XGBoost est utilisee pour convertir en distributions dynamiques de gravite des collisions. Les resultats indiquent egalement que le frottement devient plus important lorsque le numero de frottement est inferieur a 38 et que des mesures correctives immediates sont necessaires lorsque le numero de frottement est inferieur a 20. Mots-cles : frottement des chaussees, gravite des collisions, securite routiere, amplification de gradient extreme (<< XGBoost >>), explications additives de Shapley (<< SHAP >>)., Introduction Pavement friction plays an important role in highway safety. Poor pavement friction tends to increase the possibility of vehicle skidding or hydroplaning, leading to increased safe stop distance and [...]

Pavement crack detection has always been an important problem in pavement maintenance. With rapid development in computer vision technology, crack detection using deep learning has become one of the major tasks in automatic pavement condition detection and assessment. This paper aims to develop a method of crack grid detection based on convolutional neural network. First, an image denoising operation is conducted to improve image quality. Next, the processed images are divided into grids of different scales (10 x 10, 20 x 20, 30 x 30), and each grid is fed into a convolutional neural network for detection. The pieces of the grids with cracks are marked and then returned to the original images. Finally, on the basis of the detection results, threshold segmentation is performed only on the marked grids. Information about the crack parameters is obtained via pixel scanning and calculation, which realizes complete crack detection. The experimental results show that 30 x 30 grids perform the best with the accuracy value of 97.33%. The advantage of automatic crack grid detection is that it can avoid fracture phenomenon in crack identification and ensure the integrity of cracks. Thresholding smaller grids allows better combinations of the differences between objects and backgrounds, which makes segmentation results more accurate. Key words: convolutional neural network, pavement crack detection, maximum entropy threshold segmentation, crack type judgment, parameter calculation. La detection de fissures dans la chaussee a toujours ete un probleme important dans l'entretien de la chaussee. Avec le developpement rapide de la technologie de vision par ordinateur, la detection des fissures par l'application de l'apprentissage en profondeur est devenue l'une des taches principales de la detection et de l'evaluation automatiques de l'etat des chaussees. Cet article vise a developper une methode de detection de fissures par grille, et ce, basee sur un reseau neuronal convolutif. Tout d'abord, une operation de debruitage d'images est effectuee pour ameliorer la qualite d'image. Ensuite, les images traitees sont divisees en grilles differentes (10 x 10, 20 x 20, 30 x 30), et chaque grille est alimentee dans un reseau neuronal convolutif pour la detection. Les morceaux des grilles avec des fissures sont marques puis retournes aux images originales. Enfin, sur la base des resultats de detection, la segmentation des seuils n'est effectuee que sur les grilles marquees. L'information sur les parametres des fissures est obtenue par le balayage et le calcul des pixels, qui realisent la detection complete de fissures. Les resultats experimentaux montrent que les grilles 30 x 30 sont les plus performantes avec une precision de 97,33%. L'avantage de la detection automatique des fissures par grille est qu'elle peut eviter le phenomene de rupture lors de l'identification des fissures et assurer l'integrite des fissures. Des grilles plus petites permettent une meilleure combinaison des differences entre les objets et les arriere-plans, ce qui rend les resultats de segmentation plus precis. [Traduit par la Redaction] Mots-cles: reseau neuronal convolutif, detection de fissures dans la chaussee, segmentation maximale du seuil d'entropie, jugement du type de fissure, calcul de parametres., Introduction Pavement maintenance has always been a very important problem, which is inseparable from the detection of pavement distress. After protracted use, any pavement, whether made of asphalt or cement, [...]

Previous analytical studies have demonstrated that low-density lightweight cellular concrete (LCC) subbase pavements can support up to 20 times more traffic loads than unbound granular subbase pavements while protecting the pavement subgrade from adverse freeze–thaw effects in cold regions. This study examines the possibility of providing better-performing pavements on the field through the construction, instrumentation, and monitoring of flexible pavement sections incorporating three LCC densities (400, 475, and 600 kg/m³) as subbase material compared with unbound granular material in Canada. The effects of daily and seasonal temperatures on pavement critical responses to stress and strains were evaluated. The findings showed that these LCC pavements reduced asphalt concrete tensile strain by over two times compared with unbound granular pavements, and that strain increased with a daily temperature increase. Daily subgrade pressure (stress) change was reduced by up to 68%. The study concluded that longer life pavements could be achieved with LCC subbase thicknesses ≥250 mm. [ABSTRACT FROM AUTHOR]

In North America, highway agencies have started using Superpave as it incorporates a performance-based asphalt binder specification and a mix design analysis system. However, in a pavement management system (PMS), the performance of the pavement structure significantly influences management decisions. In this regard, accurate prediction and evaluation of performance is a very important aspect. With this in mind, this study investigates the performance of Superpave through the mechanistic-empirical (M-E) approach, field-evaluated performance, and laboratory performance tests. It considers 15 sections of highways from Ontario. The investigation found that the international roughness index (IRI) and permanent deformation are overpredicted in the M-E approach compared with field observations. Additionally, to better understand the performance, the dynamic modulus of asphalt mixtures and binder rheological testing are also performed. The master curve developed for the surface mixtures suggests a lower level of fatigue resistance that justifies the bottom-up fatigue failure in the field-observed scenarios. Key words: Superpave, mechanistic-empirical (M-E) approach, mechanistic-empirical pavement design guide (MEPDG), pavement condition evaluation, pavement performance, dynamic modulus test. En Amerique du Nord, les organismes responsables de voirie ont commence a utiliser Superpave, car ce systeme integre une specification de liant d'asphalte basee sur le rendement et un systeme d'analyse de la conception du melange. Toutefois, dans un systeme de gestion routiere (SGR), le rendement de la structure des chaussees a une incidence considerable sur les decisions de gestion. A cet egard, l'exactitude des previsions et de l'evaluation du rendement sont des aspects tres importants. Dans cette optique, la presente etude se penche sur le rendement de Superpave au moyen de l'approche mecanisteempirique (M-E), evalue le rendement sur le terrain et effectue des essais de rendement en laboratoire. On considere 15 troncons d'autoroute du reseau routier de l'Ontario. Il ressort de cet examen que l'indice de rugosite international (IRI) et la deformation permanente sont surestimes dans l'approche M-E par rapport aux observations sur le terrain. De plus, afin de mieux comprendre le rendement, un module dynamique des melanges d'asphalte et des essais rheologiques de liant sont egalement effectues. La courbe maitresse developpee pour les melanges de revetement suggere un niveau inferieur de resistance a la fatigue qui justifie la rupture par fatigue ascendante dans les scenarios observes sur le terrain. [Traduit par la Redaction] Mots-cles : Superpave, approche mecaniste-empirique (M-E), guide de conception mecanisto-empirique des chaussees (<< mechanistic-empirical pavement design guide (MEPDG) >>), evaluation de l'etat de la chaussee, tenue de la chaussee, essai du module dynamique., 1. Introduction In North America, highway agencies began using the Superior Performing Asphalt Pavements (Superpave) system in the 1990s with consideration of improved performance. Superpave is a comprehensive asphalt mix [...]