A growing attention has been paid to building envelope features for achieving lower energy consumption especially in large office buildings and hot climate zones, since these features and their variables are affecting energy consumption widely and with different sensitivity. Therefore, this paper conducts simulation-based comparative analyses between main envelope features with their internal variables; the selected features for this study are building geometry ratios, orientations and common envelope finishing materials (FMs). Two applications have been conducted (comparing cases with either a same or different building volumes), and more than 500 cases/simulations have been conducted and studied in total. Accordingly, sensitive features and variables have been determined to enrich design decisions for different cases, along with best variables' integrations that achieve best energy consumption through the proposed applications and cases. Cubic office buildings in Egypt have been used to demonstrate the study, and energy simulations have been achieved using eQuest (DOE-2). Results show that lower height with wider roof achieves best energy consumption if building volume is fixed via comparisons, and vice versa. Gravel and galvanized steel represent best studied roof and walls' FMs, while roofing shingles is the worst one. If building volume is varied via comparisons, horizontal dimensions are the most sensitive feature that affects energy consumption per m², while FMs and height represent lowest sensitivity among studied features. Ranking of cases, features, variables along with sensitive features in details have been analyzed and discussed through the paper.

Utilizing polymers for asphalt concrete (AC) mixture modification has many drawbacks that hinder its wide implementations for roadway construction. Recently, research on employing complementary materials, such as nanomaterials, to balance negative impacts of polymers while enhancing the AC mixture’s performance has received great attention. This study aimed to investigate the effect of incorporating nanoclay (NC) particles on the performance of a high-density polyethylene (HDPE)-modified AC mixture. A 60/70 asphalt binder was first modified with HDPE, and then NC particles were gradually added at a concentration of 1–4% by weight of the asphalt binder. The binders’ physical characteristics, storage stability, and chemical change were scrutinized. AC mixture performance, including pseudo-stiffness, moisture damage resistance, stripping susceptibility, and rutting tendency, was investigated. A statistical analysis on the experimental results was conducted using Kruskal–Wallis and Dunn tests. Test results showed that employing NC/HDPE significantly increased penetration index and thereby enhanced binder temperature sensitivity. Moreover, it prevented oxidation action and separation and, therefore, enhanced binder storage stability. Furthermore, incorporating NC amplified pseudo-stiffness and significantly improved resistance against moisture damage and stripping of HDPE-modified mixtures. Moreover, it improved both elastic (recoverable) and plastic (unrecoverable) deformations of mixtures. The most satisfactory results were attained when incorporating 3% of NC.

The dynamic modulus (E*) of hot-mix asphalt mixtures is one of the most tedious and time-consuming laboratory testing material properties. It requires costly, advanced equipment and skills that are not easily accessible in the majority of laboratories yet. Thus, many studies have been dedicated to developing E* predictive models. Unfortunately, it is a complex task due to the many input variables and their non-linear effect on the E*. This study applies a deep residual neural networks (DRNNs) technique for the first time to the problem to enhance the E* prediction capabilities. The proposed DRNNs architecture utilizes residual connections (i.e., shortcuts) that bypass some layers in the deep network structure in order to alleviate the problem of training with high accuracy. An intensive laboratory database is employed in the DRNNs model development considering all influential input parameters such as; mixture gradation, volumetric properties, binder characteristics, and testing conditions parameters. Moreover, a brute force enumeration is integrated in the model to reduce the number of needed input variables and identify the best combinations of them. Then, the proposed DRNNs performance, with the best combination of inputs, is evaluated using representative performance indicators and compared with the well-known E* predictive models, namely; Witczak 1-37A, Witczak 1-40D, and Hirsch models. Finally, a variance-based global sensitivity (VB-GS) analysis is conducted with the Monte Carlo simulation aid to highlight each input variable effect on the E* magnitude in real practice while removing the potential distortion of results due to the input variables correlations. Performance evaluation indicators reveal that the DRNNs model outperforms other E* prediction ones. Furthermore, VB-GS analysis shows that, among all feasible inputs, binder stiffness characteristics and testing temperature are the most significant ones.