Prediction of the crack condition of highway pavements using machine learning models

Departments of Transportation regularly evaluate the condition of pavements through visual inspections, nondestructive evaluations, image recognition models and learning algorithms. The above methodologies, though efficient, have drawn attention due to their subjective errors, uncertainties, noise effects and overfitting. To improve on the outcomes of the shallow learning models already used in pavement crack prediction, this paper reports on an investigation of the use of recursive partitioning and artificial neural networks (ANN; deep learning frameworks) in predicting the crack rating of pavements. Explanatory variables such as the average daily traffic and truck factor, roadway functional class, asphalt thickness, and pavement condition time series data are employed in the model formulation. Overall, it is observed that the recursive partitioning (regression tree – R² > 0.8 and classification tree – R² > 0.6) and ANN (continuous response – R² > 0.8 and categorical response – R² > 0.6) are compelling machine learning models for the prediction of the crack ratings based on their goodness-of-fit statistics, mean absolute deviation (MAD < 0.4) and the root mean square errors (RMSE between 0.30 and 0.65).     

Pull-Down of Metalloproteins in Their Native States by Using Desthiobiotin-Based Probes

One-third of all proteins are estimated to require metals for structural stability and/or catalytic activity. Desthiobiotin probes containing metal binding groups can be used to capture metalloproteins with exposed active-site metals under mild conditions so as to prevent changes in metallation state. The proof-of-concept was demonstrated with carbonic anhydrase (CA), an open active site, Zn²⁺-containing protein. CA was targeted by using sulfonamide derivatives. Linkers of various lengths and structures were screened to determine the optimal structure for capture of the native protein. The optimized probes could selectively pull down CA from red blood cell lysate and other protein mixtures. Pull-down of differently metallated CAs was also investigated.     

The Analysis of the Lubrication Condition and Friction Losses of a Single Acting Cross Head Guide Shoe of a Low Speed Cross Head Diesel Engine: Part II — A Practical Model for the Determination of the Oil Film Thickness

The cross head guide shoe plate of a large diesel engine operates under a large normal force which presses it against the rails between which it executes both axial and transverse motion. In order to ensure the safe operation of an unconventional bearing such as the guide shoe system of the cross head engine, an algorithm that can be used to predict accurately the oil film thickness that can be achieved for all the engine's operating regimes, is invaluable. A solution of the three dimensional form of Reynolds differential equation can provide the requisite oil film thickness. However, with even the availability of very fast computers and large computer storage capacity this solution for a large axial bearing such as the guide shoe becomes impractical in terms of computation time and computer core requirements. This paper describes alternative simplified algorithm for accurate prediction of the oil film thickness developed in the guide shoe bearing at different operating regimes of the engine.     

A review of U.S. Army Corps of Engineers airborne coastal mapping in the Great Lakes

The U.S. Army Corps of Engineers (USACE) Joint Airborne Lidar Bathymetry Technical Center of Expertise (JALBTCX) collected its first airborne coastal mapping data on the Great Lakes in 1995. Since then, the JALBTCX has collected nearly 5 billion elevation and depth measurements and created over 2000 geographic information system (GIS) products for the shorelines of the Great Lakes. With improvements in airborne coastal mapping technology and the introduction of regional management for the USACE, surveys expanded into the regional scale, multi-sensor National Coastal Mapping Program (NCMP). The NCMP was initiated in 2004 to produce recurrent, regional, high-resolution light detection and ranging (lidar) elevation data and hyperspectral and aerial imagery to support regional sediment management, regulatory functions, flood damage reduction, asset management, emergency operations, and environmental stewardship. The JALBTCX is a collaboration among the USACE, the U.S. Naval Oceanographic Office (NAVOCEANO), and the National Oceanic and Atmospheric Administration (NOAA). The partners have worked together on airborne coastal mapping and charting since the late 1980s with the goal of advancing airborne lidar bathymetry and associated technologies. The collaboration has fielded three generations of airborne sensors and has transferred this technology to the commercial sector, supporting an expanding market for bathymetric lidar. The purpose of this paper is to provide an overview of the history of USACE survey efforts in the Great Lakes (1995–2012), an in-depth review of the resulting imagery and lidar data products, and new information product developments and applications to support environmental and coastal engineering throughout the Great Lakes region.     

Short communication: Apparent efficiency of colostral immunoglobulin G absorption in Holstein heifers

Adequate absorption of bovine colostrum correlates with improved neonatal health. The apparent efficiency of absorption (AEA) of immunoglobulins can be measured using a mathematical equation based on serum and colostral IgG concentration levels, as well as calf body weight and the volume of colostrum being fed. Although commonly measured in research projects, little information is available on the normal AEA across a large group of healthy calves on multiple farms. The purpose of this study was to observe how contributing factors (volume of feeding, birth weight, and time of feeding) can alter AEA and establish a reference range for AEA in healthy calves. Study subjects were 100 Holstein heifer calves from 5 different dairies in North Carolina and Colorado. After a normal calving, the heifer received either 4 or 5.6 L of colostrum within 4 h of birth, an aliquot of the fed colostrum was saved, and a blood sample was collected between 24 and 36 h after birth. Birth weights were measured using the same weight tape on each farm. Radial immunodiffusion assay was performed to obtain IgG concentrations in the colostrum and serum samples. From this data, the AEA was calculated. The AEA ranged from 7.7 to 59.9% with mean of 28.1 ± 9.5% and median of 27.5%. The AEA of 69% of the calves fell between 21 and 40%. The AEA varied widely between calves, even when feeding was standardized. Results suggest that serum IgG concentration may potentially be increased by feeding increased volumes of colostrum or genetic selection, given the wide range of AEA values obtained.     

Texture‐modified 3D printed dark chocolate: Sensory evaluation and consumer perception study

This study aimed to assess the preferences and perceptions of texture‐modified three‐dimensional (3D) printed chocolate through three measures: two tasting tests and one survey. In the first test, 30 semitrained panelists ranked their overall preference from among the three samples of chocolate printed in a honeycomb pattern with infill percentages (IPs) of 25, 50, and 100%. The panelists ranked the samples based on appearance and hardness. In the second test, the same panelists nominated one preference between a 3D printed sample (100% IP) and a cast commercial chocolate sample. Friedman test indicated that there was no significant difference in overall preferences for hardness although the panelists significantly preferred the appearance of samples with 25 and 50% over the 100% infill. Furthermore, there was no significant difference in preference between the cast and 100% infill samples. The texture data of the chocolate samples showed that a higher force was required to break the chocolate samples as the IP increased from 25% (20.4 ± 1.1 N) to 100% (54.4 ± 1.5 N). Also, the 3D printed chocolate (printed in 100% IP) was found to be less hard than that of casted chocolate. In the survey of consumer perceptions, a total of 244 participated and assessed the samples for their intricate design and novel technology concept through a questionnaire. While there was a general awareness of 3D printing technology among these participants, many were impressed with the application of 3D printing to chocolate, as this was the first time they had seen this. The results obtained from the sensory tests and consumer survey provided a useful insight into consumers' perception of 3D food printing and the 3D products design. This awareness will be beneficial to promote this technology in the food industry.     

Exploring Variability and Performance in a Sub-200-mV Processor

In this study, we explore the design of a subthreshold processor for use in ultra-low-energy sensor systems. We describe an 8-bit subthreshold processor that has been designed with energy efficiency as the primary constraint. The processor, which is functional below V_dd=200 mV, consumes only 3.5 pJ/inst at V_dd=350 mV and, under a reverse body bias, draws only 11 nW at V_dd=160 mV. Process and temperature variations in subthreshold circuits can cause dramatic fluctuations in performance and energy consumption and can lead to robustness problems. We investigate the use of body biasing to adapt to process and temperature variations. Test-chip measurements show that body biasing is particularly effective in subthreshold circuits and can eliminate performance variations with minimal energy penalties. Reduced performance is also problematic at low voltages, so we investigate global and local techniques for improving performance while maintaining energy efficiency.     

Evidence for intact semantic representations in patients with frontal lobe lesions.

Patients with frontal lobe lesions and control participants were assessed on 2 tests of semantic knowledge. In the triadic comparison task, participants were shown all possible triplets of 12 animal names and judged which 2 of each triplet were most alike. In the ordered similarity task, participants rank ordered animals in terms of their similarity to a target animal. For both tasks, semantic structure-- derived from multidimensional scaling techniques-- revealed similar representations in patients with frontal lobe lesions and control participants. Additional pathfinder analyses also produced networks that did not differ between groups. These patients exhibited intact semantic knowledge despite deficits on tests of free recall and verbal fluency that involved the same semantic category and exemplars. Thus, intact representation of semantic knowledge in frontal patients stands in contrast to their marked deficits in strategic retrieval of semantic knowledge. (PsycINFO Database Record (c) 2010 APA, all rights reserved)