Search

Effective winter road maintenance relies on precise road friction estimation. Machine learning (ML) models have shown significant promise in this; however, their inherent complexity makes understanding their inner workings challenging. This paper addresses this issue by conducting a comparative analysis of road friction estimation models using four ML methods, including regression tree, random forest, eXtreme Gradient Boosting (XGBoost), and support vector regression (SVR). We then employ the SHapley Additive exPlanations (SHAP) explainable artificial intelligence (AI) to enhance model interpretability. Our analysis on an Alberta dataset reveals that the XGBoost model performs best with an accuracy of 91.39%. The SHAP analysis illustrates the logical relationships between predictor features and friction within all three tree-based models, but it also uncovers inconsistencies within the SVR model, potentially attributed to insufficient feature interactions. Thus, this paper not only showcase the role of explainable AI in improving the ML interpretability of models for road friction estimation, but also provides practical insights that could improve winter road maintenance decisions. [ABSTRACT FROM AUTHOR]

Moisture-induced damage has been known as a major premature distresses in asphalt pavement. Numerous review papers have been made to provide the progression in study on moisture damage of asphalt pavement. However, mechanical properties of asphalt mixtures used as a parameter for moisture susceptibility evaluation were not reviewed in detail. To make the review of moisture-induced damage in asphalt pavement more comprehensive, this paper reviews research studies on moisture damage of asphalt pavement in view of mechanical properties as a parameter during the past 50 years. These parameters include tensile strength ratio (TSR), cohesion ratio (CR), dynamic modulus ratio (DMR), resilient modulus ratio (RMR), and shear strength ratio (SSR). TSR and CR are only relevant at bottom of asphalt concrete (AC), while SSR can be used to describe moisture damage at different AC depths. DMR and RMR are concentrated on the impact of moisture damage via performance characteristics of asphalt mixtures. [ABSTRACT FROM AUTHOR]

Facility management maintains building service quality through cycles of condition assessments and rehabilitation. Building components, however, differ in their nature, service lives, deterioration patterns, and textual/visual inspection data. This complicates the condition assessment process and subsequent rehabilitation decisions. This paper proposes a smart condition assessment framework that uses different artificial intelligence (AI) techniques that suit the condition data analysis of different building components. The framework has been applied to a dataset of over 2000 maintenance requests for roof and heating, ventilation, and air conditioning (HVAC) systems across a 600-villa portfolio. To address their varying needs, convolutional neural networks were used on images of roof defects, while enhanced data mining was used on textual data of HVAC systems. Accordingly, work packages of deteriorated components were identified, and a 60-day schedule was developed to repair 203 HVAC units. This research shows how AI can assist facility management with respect to condition assessment, rehabilitation planning, and resource allocation. [ABSTRACT FROM AUTHOR]

The road agencies collect and submit weigh-in-motion (WIM) data to the Federal Highway Administration as part of their traffic monitoring program. Therefore, the WIM data should be precise and accurate. One way to evaluate WIM measurement errors is by using the test truck data collected immediately before and after equipment calibration. The limitation of this approach is that the data represent a snapshot in time and may not represent a long-term WIM site performance. This paper presents an approach for estimating WIM system accuracy based on axle load spectra attributes (normalized axle load spectra (NALS) shape factors). This alternative approach allows for characterizing temporal changes in WIM data consistency. The WIM error data collected before and after calibration were related to Class 9 NALS shape factors in the proposed methodology. This paper aims to determine WIM system errors based on axle loading without physically performing WIM equipment performance validation using test trucks. The presented methodology can be used to estimate systematic errors (drift) in the WIM system at any point in time after the equipment calibration. This approach can help highway agencies select optimum timings for routine maintenance and calibration of WIM equipment without compromising its accuracy. The results show that the WIM accuracy for the single axle (SA) and tandem axle (TA) can be estimated with SA and TA NALS shape factors with an acceptable degree of error for bending plate to quartz piezo sensors. Examples are included to demonstrate the application and significance of the developed models. [ABSTRACT FROM AUTHOR]

Bridge infrastructure assets should be planned, designed, constructed, and operated to account for the physical processes that may occur over their service lives. In ice-affected rivers, this includes the interaction of bridge infrastructure with river ice. Advancements in river-ice science and modelling of ice processes in ice-covered rivers have enhanced the abilities of civil/hydrotechnical engineers to plan, design, and construct bridges in and along rivers with seasonal ice covers. In this paper, an overview is presented of ice-related considerations with respect to bridges. The bottom-chord elevation of the bridge superstructure with respect to water and ice levels, pier design to withstand ice forces, the waterway opening for ice passage, ice-induced scour, and research needs are discussed. However, as outlined in the paper, research on ice loads and passage at bridge sites and on hydroclimatic and ice conditions that led to bridge damage and failure remains needed. [ABSTRACT FROM AUTHOR]

Tack coat is a thin asphalt applied between the existing surface and asphalt overlay during road rehabilitation. Its coverage uniformity is vital in providing adhesive bonding between the two layers in the structure. This paper proposes a novel method for automatic uniformity inspection of tack coat coverage using images collected by drones. In this method, color thresholding is first performed to extract coarse region of the tack coat. Then, the boundaries of the tack coat are detected to refine the tack coat region by masking outliers. Next, the region of interest is processed to extract texture features using gray-level co-occurrence. Finally, the texture features are leveraged to grade the tack coat uniformity level using LightGBM. Experiments were conducted to examine the performance of the method. The resulting accuracy signifies the potential of this method in developing high-efficiency and cost-effective solutions to enhance the current tack coat inspection practice. [ABSTRACT FROM AUTHOR]

In construction, the accurate measurements are important in ensuring the quality of work delivered. Different measuring tools have been developed to help workers conduct accurate measuring. However, they may be subject to manipulation difficulties, such as the need for tap/gesture interaction. This paper proposes a novel eye gaze-aided virtual tape measure framework that provides a hands-free manner for conducting the measurements in construction. This framework consists of three components: data collection for point of interest, sensor calibration, and distance calculation. Its effectiveness is tested by measuring the dimensions of 15 common objects in laboratory and on-site environments and achieves the average absolute and relative errors of 2.4 cm and 4.8%. The absolute errors range from 0.3 to 7.3 cm. A comparison study is conducted to demonstrate its superior performance over iPhone's Measure application. The results illustrate the feasibility and potential of using the framework to enable measures for smart construction. [ABSTRACT FROM AUTHOR]

Resilience is a multidisciplinary concept that deals with rapid response and recovery of a system experiencing a disruption. Despite extensive research on this topic, there is a need to clarify resilience concepts in the context of road freight transportation systems (FTSs) from a three-dimensional perspective (i.e., physical infrastructure, users, and managing organizations) and to identify persistent knowledge gaps concerning the characterization and measurement of FTS resilience vis-à-vis disruptive events. This paper addresses these shortcomings through a systematic review of 149 research studies. The synthesis of findings clarifies inconsistencies associated with the characteristics of FTS resilience and in so doing, establishes a unified framework for measuring FTS resilience through the life cycle of disruptive events. Critical data gaps, methodological shortcomings, and a lack of empirical evidence concerning FTS resilience to disruptive events remain. More robust analytical approaches are required to incorporate interdependencies among FTS dimensions into resilience assessments. [ABSTRACT FROM AUTHOR]

This paper presents an effort to address urgent issues associated with semitruck platooning. The weigh-in-motion data were first examined to determine the characteristics of naturally formed semitruck platoons. A safety-based two-vehicle platoon model was then utilized to determine the critical headways for assessing the safety of semitruck platooning in various driving scenarios. The possibilities for passenger vehicles to cut into semitruck platoons were assessed using actual vehicle headway data. Finally, the potential impacts of semitruck platooning on typical interstate highway pavements were assessed using finite element analysis. Most semitruck platoons consist of two to three vehicles. An optimal platoon gap of 0.94 s was identified for connected and automated semitruck platooning. The potential impacts of truck platooning on concrete pavement were negligible for a platoon gap greater than 0.2 s. The maximum possible impact was less than 4% in terms of the service life reduction for interstate asphalt pavement. [ABSTRACT FROM AUTHOR]

It is crucial to evaluate the consistency in weigh-in-motion (WIM) loading over time and quantify the relative accuracy of axle loads based on axle load spectra (ALS) data for different sensor types. This paper presents the temporal evaluation of the ALS from 51 WIM sites and 128 records available in the long-term pavement performance data. Analysis of ALS data over time shows that for single ALS, there is a significant difference in peak loads between the bending plate, and quartz piezo sensor measurements. Also, 100% of the bending plate 86% of the quartz piezo, and 66% of the piezo cables sites exhibited consistent single axle peak loads after 1 year of calibration event. For tandem ALS, significant differences were observed between loaded peaks of bending plate sensor from both quartz piezo and piezo cable sensors. Also, 83% of the bending plate, 78% of the quartz piezo, and 50% of the piezo cable sites exhibited loaded peaks consistently 1 year after calibration. The results show that calibration frequencies longer than 1 year may be acceptable for the bending plate sensors. However, calibration frequencies of at least 1 year for quartz piezo and less than a year for piezo cable sensors are recommended. Pavement designers and analysts should be aware of the changes in WIM data and calibration frequency over time. [ABSTRACT FROM AUTHOR]

The process of icing and the resulting layered ice masses, called aufeis, are caused by the freezing of overflow originating from groundwater or surface water. Aufeis can directly impact infrastructure and property, most commonly through winter ice formation and spring flooding within, against, and on the surface of hydraulic structures and transportation infrastructure. They also represent a safety concern for drivers. This geohazard often needs to be managed proactively and efficiently to mitigate associated risks. This paper provides an overview of the consequences of aufeis in northwestern Canada. A total of 50 existing and novel icing and aufeis mitigation approaches are described and classified. The context of applicability for each approach is identified, considering the source of water, the type of infrastructure, and its role in the formation of aufeis. Finally, future research avenues to support the development or improvement of aufeis risk reduction techniques are presented. [ABSTRACT FROM AUTHOR]

Ice platforms are engineered sheets of floating ice that are designed to support long-term stationary loads, usually related to construction. These loads can include cranes, drill rigs, excavators, girder trucks, and tandem trucks. This paper will discuss general design procedures for ice platforms as well as cases with special considerations such as multiple long-term loads, repetitive long-term loads, objects dropped down onto the ice, loads near large holes in the ice, loads near bridge piers, loads near banks, and loads on access ramps. [ABSTRACT FROM AUTHOR]

The process of icing involves the freezing of overflow layers, on ground or within streams, and results in ice bodies called "aufeis" that are common in most northern landscapes. Knowledge about aufeis is still limited despite the cold region engineering challenge they represent. Understanding the causes of overflow events leading to aufeis development represents a key for the prediction, mitigation, and management of this geohazard and can also support the planning and design of infrastructure in the North. This paper introduces a practical classification for the diverse range of overflow processes that generate aufeis, including under-represented processes, such as the instability of winter streamflow. Importantly, it distinguishes flow conveyance from water supply overflow processes and describes the temporal aspect of icing intensity. Finally, research topics are proposed to improve our understanding of aufeis, including their predictability, the impact of climate change on their occurrence and extent, and stream morphology–aufeis interaction. [ABSTRACT FROM AUTHOR]

This paper presents a method to quantitatively assess the impact of ice-affected hydraulics on hydropower operations using the widely used Hydrologic Engineering Center River Analysis System (HEC-RAS) hydraulic model. The model was used to identify historical periods where downstream river ice growth reduced the net head on a US Army Corps of Engineers dam on the Pend Oreille River in Idaho, USA. The model was used to create curves of projected stage impacts based on border-ice and continuous bank-to-bank ice formation scenarios. The temperature modelling capabilities in HEC-RAS were used to hindcast ice formation periods on the river that were validated with shore-based webcam data. These results provide a framework that can be applied to other rivers with hydropower generation, particularly sites where an open-water HEC-RAS model exists. [ABSTRACT FROM AUTHOR]

A large amount of research has been undertaken on tubular bolted end-plate connections loaded in axial tension, and considerably less on such connections in flexure. This paper studies the combined effects of these two loads and presents the first laboratory tests on these connections under combined tension and flexure. Using validated finite element models, a parametric study with 360 numerical models is performed on these connections under axial and eccentric tension loads. After screening this database of experimental tests and numerical results to isolate only connection failure modes, current design methods for predicting the connection capacity under eccentric tension are evaluated and revised. In addition, based on interaction curves, a novel approach is proposed to design these splices under combined loads. For all methods, the use of the plate ultimate and yield stress was compared for the end-plate material strength parameter. Lastly, a reliability analysis is performed for the application of the recommended design methods to Canadian practice. [ABSTRACT FROM AUTHOR]

Bioretention cells in the Toronto region are usually sized to accommodate runoff from the 90th percentile storm, which has a depth of about 25 mm. This research examines the water quality control performance and cost of bioretention cells sized to satisfy alternative design criteria ranging from 5 to 50 mm. The long-term average runoff-capture efficiencies provided by representative bioretention cells are determined, and their capital as well as operation and maintenance costs are estimated. Results indicate that the current design criterion of 25 mm is probably too high and not cost-efficient. In fact, above some threshold levels, little improvement in runoff-capture and pollutant removal performances may be achieved if the design criterion is increased further, but cost would still linearly increase. Presented here is a methodology that can be used to properly consider both the performance and cost of bioretention cells for establishing a more cost-efficient design criterion. It is shown that a more cost-efficient design criterion for Toronto could be lower than the current one, and significant cost savings can be realized if a lower design criterion is implemented. Some inappropriate ways of quantifying the effective volume of storage provided by bioretention cells are also identified in this paper. [ABSTRACT FROM AUTHOR]

In this paper, the effect of ternary binder containing Portland cement type I (PCI), slag, and limestone (LS) powder in varying proportions on the suction development and the mechanical strength of cemented paste backfill (CPB) was investigated. Mechanical tests, microstructural analyses (thermo-gravimetry analysis (TG/DTG) and mercury intrusion porosimetry), and monitoring (suction, electrical conductivity, and temperature) experiments were conducted on CPB. In the first part of this study, binary binders with two different PCI:slag proportions, 50:50 and 80:20, were examined with no LS, followed by the slag replacement with increasing weight percentages of LS from 0% to 20% by weight of the total binder, with constant cement content. The results indicate that the slow slag hydration kinetics decreased the suction and strength gain rate at a young age in the binary sample with high slag content (50 wt.%); however, its latent hydraulic and pozzolanic properties enhanced the strength gain at a later age, from 28 days. The addition of 5 wt.% and 10 wt.% LS to the ternary binder system improved the early age strength of CPB up to 7 days and 1 day, respectively, compared with the binary control samples, i.e., CPB with slag and without LS. While the addition of up to 10% LS to the ternary binder improved the rate of suction development and increased suction at early ages, beyond 10%, the addition of LS slightly decreased mechanical performance at all ages. Overall, the results obtained reveal that up to 50 wt.% slag with up to 10 wt.% LS with cement could be used as a ternary binder to improve the early age strength and suction development rate of CPB studied without significant loss in compressive strength at advanced ages. The results of this research could be of great benefit to the mining industry through the use of a ternary binder to replace cement, which could improve the overall performance of the CPB system, facilitate a more sustainable CPB design, and improve tailings management practices and overall economics. [ABSTRACT FROM AUTHOR]