Background: In Advanced Manufacturing and Supply Chain (AMSC) environments, workers are subjected to highly dynamic and demanding conditions that often impose significant physical strain. This strain frequently leads to increased fatigue, which in turn impacts worker safety, productivity, and overall well-being. Specifically, prolonged or repetitive activities in these settings can cause muscle fatigue, evident in changes in electromyography (EMG) signal characteristics. Therefore, monitoring muscle condition to prevent overexertion is crucial, particularly in tasks involving sustained or repetitive forearm movements.

EMG is a key tool used to monitor the muscle condition of workers, helping to determine their maximum workload, and the number of repetitions they can perform before experiencing fatigue. However, measuring the activity of specific forearm muscles, such as the Flexor Carpi Radialis, Extensor Carpi Radialis Longus, Flexor Digitorum, and Extensor Digitorum, presents unique challenges. These challenges stem primarily from the muscles' smaller size, complex anatomical structures, and their role in fine motor control. The muscles' close proximity to each other often leads to signal crosstalk, complicating the isolation of individual muscle signals, especially when using surface EMG. Additionally, the depth and overlapping nature of these muscles make accurate signal acquisition more challenging. In light of these complexities, this study aims to explore the feasibility of using a wearable EMG sensor to measure the electrical impulses produced by workers’ forearm muscles. The goal is to continuously evaluate muscle fatigue in a non-intrusive manner that does not interfere with ongoing tasks, thereby enhancing worker safety and productivity in AMSC environments.

Research Objectives and Plans:

1. Objective 1: Address Challenges in EMG Signal Acquisition

   • To tackle the specific challenges associated with measuring EMG signals from smaller, closely packed forearm muscles, including Flexor Carpi Radialis, Extensor Carpi Radialis Longus, Flexor Digitorum, and Extensor Digitorum.

   • To develop methods to reduce signal crosstalk and improve the accuracy of individual muscle signal isolation.

2. Objective 2: Optimize EMG Sensor Design and Placement

   • To design and optimize the placement of wearable EMG sensors for effective monitoring without hindering the workers’ movements or task performance.

3. Objective 3: Develop Advanced Signal Processing Techniques

   • To create advanced signal processing algorithms that can accurately interpret the nuanced activities of the forearm muscles and differentiate between normal activity and onset of fatigue.

4. Objective 4: Assess Muscle Fatigue in Real-Time

   • To develop and implement a system using wearable EMG sensors that can continuously monitor the electrical activity of forearm muscles in workers, thereby assessing muscle fatigue in real-time.

5. Objective 5: Determine Maximum Workload and Repetition Limits

   • To utilize EMG data to ascertain the maximum workload, optimal lifting height, and the safe number of repetitions that workers can perform without reaching a fatigue threshold.

REU Student Outcomes:

This project is ideal for a student with interests in machine learning, deep learning, biomedical engineering, human factors, and data analysis. The REU student will:

• Gain experience in the development and implementation of wearable technology.

• Learn about the intersection of human physiology and AMSC systems.

• Contribute to data analysis and machine learning algorithm development.

• Work under the supervision of Dr. Yunmei Liu (Industrial Engineering) and participate in the preparation of conference and journal papers.

Required Skills:

• Python and R

• Machine learning and deep learning

• Background or interest in Biomedical Engineering or related fields.

• Proficiency in data analysis and familiarity with physiological data.

• Basic knowledge of wearable technology and its applications.