Fireproofing deterioration is widespread in industrial facilities throughout the country. Spalling concrete has potential to damage equipment and harm personnel. Replacing concrete fireproofing like-in-kind, without consideration for proper anchorage or material durability, does not eliminate the hazard as spalls may potentially occur again over time. However, when properly designed and installed, concrete is a durable option for replacing deficient fireproofing in aggressive environments typically present in industrial processing units. This paper presents the results of a case study on a structure in a Midwest industrial complex. Extensive concrete fireproofing repairs were performed on the structure 12 years ago. Design requirements included normal weight concrete with polypropylene fibers which enhance durability by improving cracking resistance. During a fire, the fibers melt forming relief channels for moisture to escape, thus eliminating explosive spalling. Installation methods included welded wire reinforcement (WWR) with positive anchorage to structural steel. WWR was attached to post-installed adhesive anchors between column flanges where existing fireproofing was sound and difficult to remove. After 12 years in service, repairs exhibit no significant defects. This level of durability is attributed to the design and installation methods utilized. Concrete fireproofing is a durable option for fire protection, provided structures are designed to support its weight, its mixture design is properly proportioned, and it is adequately anchored and reinforced.

Soft computing applications through artificial intelligence (AI) are becoming increasingly popular in civil engineering. From concrete technology to structural engineering, AIs have provided successful solutions to various problems and greatly reduced the computational costs while achieving excellent prediction accuracy. In this study, a review of the main artificial neural network (ANN) types used in civil engineering is presented. Each ANN type is described, and example applications are provided. As a new research contribution, a deep feedforward neural network (FFNN) is developed to predict the load capacities of post-installed adhesive anchors installed in cracked concrete, which is challenging and computationally expensive to achieve with conventional methods. The development of this FFNN is discussed, the influence of several parameters on its performance is demonstrated, and optimum parameter values are selected. In addition, a hybrid methodology that combines 2D nonlinear finite element (NLFE) techniques with the developed FFNN is briefly presented to account for real-life adverse effects in anchor analysis, including concrete cracking, wind-induced beam bending, and elevated temperatures. The results show that the developed network and methodology can rapidly and efficiently predict the load capacities of adhesive anchors installed into cracked concrete, accounting for the damage caused by the cracks, with high accuracies.

The bond between reinforcing bars and concrete is an important property that determines the performance of reinforced concrete structures. Accurate prediction of the bond strength is essential for ensuring the safety and economy of the structures. This paper proposes an artificial neural network for predicting the bond strength between straight deformed reinforcing bars and concrete under tensile load. The neural network was trained using a large dataset of test results from the ACI Committee 408 database. A robust ten-fold cross-validation method was employed for evaluating network performance and finding optimal network parameters. Hyperparameter tuning was carried out to establish the optimal network hyperparameters. The relative impact of the neural network input parameters on the bond strength was evaluated using the SHAP method. The developed neural network with the optimal hyperparameters shows a good agreement with the test results. Its accuracy exceeds the accuracy of the descriptive equations.

This special publication draws inspiration from the Technical Session entitled “The Concrete Industry in the Era of Artificial Intelligence,” held during the ACI Virtual Concrete Convention in spring 2020. To parallel the Technical Session, this special publication is also tailored to showcase the unprecedented potential of leveraging artificial intelligence (AI) methods—including its derivatives of machine learning (ML) and deep learning (DL)—in the concrete industry as a whole. The idea behind this effort started as a thought during an ACI Committee 216 meeting. From there, both ACI Committees 444 (Chair: Thomas Schumacher) and 554 (Chair: Liberato Ferrara) displayed interest in co-sponsoring this special publication. This special publication comprises fifteen papers (five from our panelists and ten received from authors representing academia and the concrete industry). This collection of papers covers the use of various AI techniques at the material level (i.e., concrete performance and mass-scale testing, property predictions, and optimization, etc.), elemental level (e.g., behavioral and capacity prediction of slabs, walls, beams, and anchorages, etc.), as well as system level (viz. damage and crack detection of concrete bridges and concrete composite structures). We are very thankful to ACI, the ACI Technical Activities Committee, as well as all three technical committees. Your kind support and commitment have not only allowed us to explore a new realm of possibilities but have also enabled us to set the stage towards a new and modern future to our industry. Special thanks go to our panelists and contributors who were very kind to share their most recent research and unique ideas pertaining to infusing AI solutions to various problems within our domain. In addition, we send our warm regards to our reviewers, ACI staff, and Ms. Barbara A. Coleman for her help in setting up and editing this effort.

Externally bonded (EB) steel reinforced grout (SRG) composites have the potential to improve the flexural and shear performance of existing concrete and masonry structural members. However, one of the most commonly observed failure modes of SRG-strengthened structures is due to composite debonding, which reduces composite action and limits the SRG contribution to the member load-carrying capacity. This study investigated an endanchorage system for SRG strips bonded to a concrete substrate. The end anchorage was achieved by embedding the ends of the steel cords into the substrate. Nineteen single-lap direct shear specimens with varying composite bonded lengths and anchor binder materials were tested to study the effectiveness of the end-anchorage on the bond performance. For specimens with relatively long bonded length, the end-anchorage slightly improved the performance in terms of peak load achieved before detachment of the bonded region. Anchored specimens with long bonded length showed notable post-detachment behavior. Anchored specimens with epoxy resin achieved load levels significantly higher than the peak load before composite detachment occurred. For specimens with relatively short bonded length, the end-anchorage provided a notable increase in peak load and global slip at composite detachment. A generic load response was proposed for SRG-concrete joints with end anchors.