Publications

You can also find my articles on my Google Scholar profile.

Journal Articles

Dynamic hazard zone modeling for proximity detection in work zones

Published in Automation in Construction, 2026. DOI: 10.1016/j.autcon.2026.107132 | PDF

A dynamic, probabilistic hazard zone model for construction equipment that continuously adjusts its size and orientation from real-time equipment kinematic data, representing risk as a bivariate generalized Gaussian probability field with multi-level hazard boundaries. In simulation and field experiments, it significantly reduced false positives compared with a conventional static circular hazard zone while maintaining stable detection across varying equipment speeds.

Recommended citation: Demeke, A. Y., Jang, Y., & Jeong, I. (2026). Dynamic hazard zone modeling for proximity detection in work zones. Automation in Construction, 190, 107132. https://doi.org/10.1016/j.autcon.2026.107132

Vehicle intrusion detection in highway work zones using inertial sensors and lightweight deep learning

Published in Automation in Construction, 2025. DOI: 10.1016/j.autcon.2025.106291 | PDF

Inertial measurement unit (IMU) sensors attached to traffic cones, combined with a lightweight LSTM deep learning model, distinguish real vehicle intrusions into highway work zones from non-hazardous events such as manual handling and wind displacement. The model reached 96% accuracy and 97% recall for actual intrusions and runs in real time on edge devices, reducing false alarms and alarm fatigue.

Recommended citation: Younesi Heravi, M., Demeke, A. Y., Dola, I. S., Jang, Y., Jeong, I., & Le, C. (2025). Vehicle intrusion detection in highway work zones using inertial sensors and lightweight deep learning. Automation in Construction, 176, 106291. https://doi.org/10.1016/j.autcon.2025.106291

Advancing Highway Work Zone Safety: A Comprehensive Review of Sensor Technologies for Intrusion and Proximity Hazards

Published in Transportation Research Record: Journal of the Transportation Research Board, 2025. DOI: 10.1177/03611981241265845 | PDF

A systematic review, following the PRISMA protocol, of sensor technologies for preventing vehicle intrusion and worker–equipment proximity hazards in highway work zones, covering GPS, radar, laser, infrared, RFID, Bluetooth, and ultrasonic sensors. It compares them by accuracy, range, power consumption, cost, and user-friendliness, and concludes that sensor selection must fit the specific application and that effective deployment also requires integrated, cost-effective, user-friendly, and secure systems and supporting regulatory frameworks.

Recommended citation: Demeke, A. Y., Younesi Heravi, M., Dola, I. S., Jang, Y., Le, C., Jeong, I., Lin, Z., & Wang, D. (2025). Advancing highway work zone safety: A comprehensive review of sensor technologies for intrusion and proximity hazards. Transportation Research Record, 2679(2), 1066–1089. https://doi.org/10.1177/03611981241265845

New Findings on Existing Resilient Modulus Constitutive Models through Performance Comparison on LTPP Data

Published in Journal of Transportation Engineering, Part B: Pavements, 2024. DOI: 10.1061/JPEODX.PVENG-1457 | PDF

A performance comparison of ten constitutive models for predicting the resilient modulus (MR) of unbound pavement materials, using repeated load triaxial test data from the Long-Term Pavement Performance (LTPP) database. The NCHRP model adopted by the MEPDG, Puppala’s model, and Ni’s model performed best, with Puppala’s slightly ahead, while the Rahim and George, Kim, and Moossazadeh and Witczak models performed poorly; models of the NCHRP form expressed in terms of confining pressure and deviator stress performed better.

Recommended citation: Demeke, A. Y., Sachpazis, C. I., Assefa, E., & Pantelidis, L. (2024). New findings on existing resilient modulus constitutive models through performance comparison on LTPP data. Journal of Transportation Engineering, Part B: Pavements, 150(2), 04024012. https://doi.org/10.1061/JPEODX.PVENG-1457

Modeling the resilient modulus of subgrade soils with a four-parameter constitutive equation

Published in Modeling Earth Systems and Environment, 2023. DOI: 10.1007/s40808-023-01735-0 | PDF

A new four-parameter constitutive model predicts the resilient modulus (MR) of subgrade soils from bulk stress, confining pressure, and deviatoric stress. Fitted to Long-Term Pavement Performance (LTPP) test data, it achieved a coefficient of determination of 0.94 (RMSE 2.20, MAE 1.76) and generally outperformed both the bulk stress model and the universal model used by the MEPDG.

Recommended citation: Demeke, A. Y., Sachpazis, C. I., Assefa, E., & Pantelidis, L. (2023). Modeling the resilient modulus of subgrade soils with a four-parameter constitutive equation. Modeling Earth Systems and Environment, 9(4), 3795–3800. https://doi.org/10.1007/s40808-023-01735-0

Conference Papers

Adaptive Hazard Zone Model for Improved Safety in Highway Work Zones

Published in Proceedings of the 42nd International Symposium on Automation and Robotics in Construction (ISARC 2025), 2025. DOI: 10.22260/ISARC2025/0016 | PDF (IAARC)

A dynamic hazard zone model for construction equipment in highway work zones that uses a bivariate Gaussian distribution and real-time GPS and IMU data to resize the hazard zone as equipment speed and movement change. In Webots simulations it detected hazards with 93% accuracy, compared with 60% for a static circular hazard zone.

Recommended citation: Demeke, A. Y., Younesi Heravi, M., Dola, I. S., Jeong, I., & Jang, Y. (2025). Adaptive hazard zone model for improved safety in highway work zones. In Proceedings of the 42nd International Symposium on Automation and Robotics in Construction (ISARC 2025) (pp. 113–121). IAARC. https://doi.org/10.22260/ISARC2025/0016

Ayenew Yihune Demeke