Fatemeh Davoudi Kakhki

Musculoskeletal disorders remain a leading source of occupational injury in manufacturing and logistics industries, often resulting from repetitive lifting and load transfer tasks. While back-support exoskeletons have shown promise in alleviating physical strain, their biomechanical impact under realistic task conditions and their integration with intelligent monitoring systems require further exploration. This study investigates the efficacy of a passive lumbar-support exoskeleton in reducing muscle fatigue during manual material handling tasks by combining high-resolution surface electromyography (sEMG) data with a deep learning-based classification framework. Ten participants performed lifting and rotational (twisting) tasks with and without exoskeleton assistance, while EMG signals were collected from the lower back and thigh muscles. A total of 32 EMG features, extracted via a rolling window approach, were used to train a feedforward neural network (FNN) to classify four task-exoskeleton conditions. The model achieved a classification accuracy of 99% with perfect class separability (AUC = 1.00), out-performing traditional statistical techniques in detecting nuanced biomechanical differences. Physiological analyses confirmed significant reductions in RMS muscle activation during exoskeleton-assisted lifting and twisting tasks, particularly in the lumbar region, demonstrating the exoskeleton’s role in fatigue mitigation and postural stabilization. These findings highlight the advantages of combining wearable biosensors with interpretable AI to support real-time fatigue monitoring and adaptive ergonomic interventions. The proposed framework contributes to the development of intelligent, human-centered exoskeletal systems that align with Industry 4.0 goals for safer, more sustainable industrial work environments.

Understanding and estimating latent demand for active transportation, such as walking and cycling, is essential for designing infrastructure and policies that promote sustainable mobility. Unlike traditional demand models that focus on observed trips, latent demand estimation seeks to quantify the unrealized potential for active travel due to barriers such as inadequate infrastructure, safety concerns, or lack of connectivity. This study develops a comprehensive latent demand model tailored for California, integrating geospatial analysis and multimodal accessibility assessments. The methodology employs a GIS-based corridor analysis approach, utilizing spatial accessibility metrics and distance decay functions to evaluate potential demand. It incorporates employment and population data, school and university enrollments, and park and trail accessibility to estimate the likelihood of walking and cycling trips. To better capture behavioral patterns, the model also classifies cyclists into four categories: strong and fearless, enthused and confident, interested but concerned, and no way, no how. Case studies in Douglas City, El Centro, and downtown San Jose illustrate the model’s application across urban, suburban, and rural contexts. Findings highlight that employment centers and commercial areas significantly contribute to bicycle demand, while schools and recreational spaces influence pedestrian activity. The results emphasize the need for targeted infrastructure improvements to convert latent demand into realized active transportation. This study provides a data-driven framework to guide investments in pedestrian and cycling infrastructure to ensure equitable and effective transportation planning in California.

Repetitive lifting tasks in occupational settings often result in shoulder injuries, impacting both health and productivity. Accurately assessing the biomechanical risk of these tasks remains a significant challenge in occupational ergonomics, particularly within manufacturing environments. Traditional assessment methods frequently rely on subjective reports and limited observations, which can introduce bias and yield incomplete evaluations. This study addresses these limitations by generating and utilizing a comprehensive dataset containing detailed time-series electromyography (EMG) data from 25 participants. Using high-precision wearable sensors, EMG data were collected from eight muscles as participants performed repetitive lifting tasks. For each task, the lifting index was calculated using the revised National Institute for Occupational Safety and Health (NIOSH) lifting equation (RNLE). Participants completed cycles of both low-risk and high-risk repetitive lifting tasks within a four-minute period, allowing for the assessment of muscle performance under realistic working conditions. This extensive dataset, comprising over 7 million data points sampled at approximately 1259 Hz, was leveraged to develop deep learning models to classify lifting risk. To provide actionable insights for practical occupational ergonomics and risk assessments, statistical features were extracted from the raw EMG data. Three deep learning models, Convolutional Neural Networks (CNNs), Multilayer Perceptron (MLP), and Long Short-Term Memory (LSTM), were employed to analyze the data and predict the occupational lifting risk level. The CNN model achieved the highest performance, with a precision of 98.92% and a recall of 98.57%, proving its effectiveness for real-time risk assessments. These findings underscore the importance of aligning model architectures with data characteristics to optimize risk management. By integrating wearable EMG sensors with deep learning models, this study enables precise, real-time, and dynamic risk assessments, significantly enhancing workplace safety protocols. This approach has the potential to improve safety planning and reduce the incidence and severity of work-related musculoskeletal disorders, ultimately promoting better health and safety outcomes across various occupational settings.

In this research, we investigate an advanced methodology for process monitoring in manufacturing that employs synthetic data with deep learning to enhance production and process quality monitoring. Traditionally, process supervision in the manufacturing industry has depended on manual checks and physical sensors, which are costly, time-consuming, and prone to errors. The study introduces a new approach that utilizes synthetic data to train deep learning models for precise defect detection in 3D-printed objects. We further recommend diverse data augmentation and fine-tuning techniques to improve model efficacy. Our evaluation demonstrates an impressive 90% overall accuracy in real-object defect detection, with a 93% precision rate in identifying defects. These results suggest that models trained on synthetic data are effectively equivalent to those trained on real-world data, indicating that the technique used to generate synthetic data is pivotal to model performance. This innovative strategy mitigates the challenges associated with conventional process monitoring methods, and addresses the practical challenges of applying such technological advancements in actual manufacturing environments. By enabling the creation of extensive training datasets and applying deep learning algorithms for object and pattern recognition, this approach boosts the precision and efficiency of operational supervision within the manufacturing process. The approach and findings of this study significantly enhances efficiency, accuracy, and the overall effectiveness of quality assurance in manufacturing.

Manual material handling (MMH) significantly impacts worker health and productivity, often leading to musculoskeletal disorders (MSDs) primarily in the lower back. As a novel assistive technology, exoskeletons may serve as ergonomic tools to mitigate these work-related MSDs. It is essential to examine exoskeletons from the users’ perspectives before their widespread implementation in occupational settings. This study investigates the effectiveness of a passive back-support exoskeleton (BExo) in reducing perceived physical exertion and improving ergonomic safety in a manufacturing context. Twenty-two college students were recruited to perform MMH tasks in a controlled lab environment, both with and without the BExo, followed by completing a survey questionnaire on various aspects of the BExo. Using ANOVA, the study analyzed biomechanical exertion across various body parts and tasks. The findings indicate that the BExo substantially alleviated discomfort and physical exertion in the low back, shoulders and knees, thereby enhancing an ergonomic posture and reducing fatigue. These results underscore the potential of passive exoskeletons to boost workers’ safety and efficiency, providing valuable insights for future ergonomic strategies in industrial settings.

Manual material handling (MMH) significantly impacts worker health and productivity, often leading to musculoskeletal disorders (MSDs) primarily in the lower back. As a novel assistive technology, exoskeletons may serve as ergonomic tools to mitigate these work-related MSDs. It is essential to examine exoskeletons from the users’ perspectives before their widespread implementation in occupational settings. This study investigates the effectiveness of a passive back-support exoskeleton (BExo) in reducing perceived physical exertion and improving ergonomic safety in manufacturing context. Twenty-two college students were recruited to perform MMH tasks in a controlled lab environment, both with and without the BExo, followed by completing a survey questionnaire on various aspects of the BExo. Using ANOVA, the study analyzed biomechanical exertion across various body parts and tasks. The findings indicate that the BExo substantially alleviates discomfort in the low back, shoulders and knees, thereby, enhancing ergonomic posture and reducing fatigue. These results underscore the potential of passive exoskeletons to boost workers’ safety and efficiency, providing valuable insights for future ergonomic strategies in industrial setting.