Infrastructures

Acoustic emission (AE) measurements have many uses to evaluate the integrity of materials. AE is often used to detect leakage in pipelines. It has also been used to monitor changes in strength properties of fiber-reinforced concrete. In the oil and gas industry, AE is predominantly used to study fracture initiation and propagation. In particular, characterization of samples is key for evaluating subsurface formations for successful underground storage. Research has been performed to understand the behavior of AE in uniaxial compression and single-stage triaxial compression tests. However, the validity of this method has not been documented in a multistage triaxial test. This characterization is required to understand the stability of the host rock under the related stress changes and potential mineralogical changes that may occur. Typically, there is a shortage of geologic samples. A single multistage triaxial test eliminates the need for twin samples and provides an economic and time-saving protocol compared to conventional methods. A single multistage triaxial (MST) test allows a constitutive model to be developed for a host rock. This work establishes a protocol for performing these tests with minimal corrections to the measurements. Acoustic emissions were measured on five different samples undergoing multistage triaxial tests. Two different behaviors were observed. For the coarse grained samples, designated Group 1 (Miocene sandstone, Wilcox Formation, and Cambrian sandstone), the number of AE events did not show a strong dependence on confining stress. They did show an exponential increase in AE events with increasing deviatoric stress during each stage. In contrast, the Group 2 samples (Niobrara Marl and Niobrara Chalk) exhibited significantly different stress-dependent AE behaviors. The amplitude of the AE events is significantly smaller than in the quartz-dominated samples, indicating a more ductile and diffuse failure mechanism. The correlation between maximum compressive strength and the point of positive dilatancy is 1.2 for both groups of samples, even though a different pattern of AE events is observed.

Urban road networks in Sub-Saharan Africa diverge from the sensor-rich, lane-disciplined environments conventional models assume. This paper establishes an empirical baseline of network efficiency and velocity decay along the 8.0 km Mile 17 to Governor’s Roundabout corridor in Buea, Cameroon, marked by a 25 m/km inbound gradient, unsignalized intersections, and a taxi share of 59.55% inbound and 65.27% outbound. Fifteen-minute PCU counts were recorded at three points across three peak windows over three consecutive weekdays (1–3 April 2026), yielding 18 h of counts and 36 floating-car GPS runs. Volume-to-capacity ratios (v/c), Peak Hour Factors, and space-mean speeds were derived for both directions, with SD, coefficient of variation, and 95% confidence intervals computed to quantify sampling uncertainty. Results show near over-saturation (v/c: 0.98–1.06) at Bonduma, particularly during the afternoon outbound window. However, the 95% confidence intervals span v/c = 1.00, indicating significant uncertainty in Level of Service classifications. Mean Travel Time Ratios ranged from 1.10 to 2.04. At Bonduma in the evening, an elevated Informal Transport Disruption Factor (ITDF = 0.96) co-occurred with only moderate volume (v/c = 0.87), consistent with, though not proof of, informal-transport-related delay. Finally, the paper introduces two new indicators, the Directional Flow Asymmetry Index (DFAI) and Corridor Velocity Decay Rate (CVDR), to quantify directional imbalances that standard v/c analysis misses. Alongside the ITDF, these metrics serve as empirical calibration inputs for a future reinforcement learning traffic control framework (BACATC). Because no AI controller was built or evaluated in this study, the findings establish a quantified field baseline rather than a demonstration of AI-based control.

Existing railway obstacle detection methods primarily focus on object localization and classification, while lacking the capability to directly support operational risk assessment. To address this limitation, this paper proposes a risk-aware railway obstacle detection framework that integrates track segmentation and lateral distance estimation. The proposed framework jointly optimizes obstacle detection and track segmentation through shared feature learning, and estimates the lateral distance between detected obstacles and track boundaries based on the extracted track geometry. Railway clearance constraints are further incorporated for obstacle risk-level determination, thereby establishing an end-to-end pipeline from obstacle detection to operational risk assessment. Consequently, an end-to-end pipeline is established to classify obstacle risk levels in railway environments. To validate the proposed framework, a dedicated RN-rail-Object dataset is constructed, and comprehensive ablation studies, comparative experiments, and edge deployment evaluations are conducted. Experimental results demonstrate that the proposed method achieves 89.2% mAP for obstacle detection, 93.7% mIoU and 97.1% Dice for track segmentation, while maintaining an inference speed of 214.6 FPS on the RTX A4000 platform. Furthermore, edge deployment demonstrates the feasibility of integrating obstacle detection, track extraction, and geometry-based risk assessment on an edge computing platform, with representative scenarios illustrating the identification of different obstacle risk states. The proposed framework provides an interpretable geometry-based approach for extending conventional obstacle detection toward railway risk-aware perception, showing potential for intelligent railway inspection, early warning, and safety-oriented operation and maintenance.