Description

Problem: Obtaining information about the manufacturing processes of human workers in labor-intensive manufacturing systems (LIMSs) presents significant challenges. LIMS are manufacturing systems where human participation is integral to the value chain, such as in the apparel, footwear, and home goods industries. The main challenges arise in two areas: data privacy legislation and the need for appropriate sensors. Tracking relevant data points of workers’ operations involves researching and installing suitable sensors that can capture the necessary data without interfering with processes. These sensors should be adaptable to future changes in processes or scenarios while complying with data privacy legislation. 
However, by enhancing data collection or utilizing existing data, the manufacturing process could be improved in multiple ways. Efficiency and resource allocation could be increased, while worker well-being and motivation could be enhanced. A first step towards better understanding LIMSs could involve extracting information about tasks from accelerometer data, such as task start and end times, worker movement patterns, and ergonomic risks associated with repetitive motions.

Goal: The goal of this master thesis is to create a model that extracts information needed to improve manufacturing processes from accelerometer data. This model aims to identify key metrics such as task start and end times, worker movement patterns, and ergonomic risks associated with repetitive motions.

Required Skills and Knowledge:

• Programming proficiency (preferably in Phyton or R)
• Proficiency in statistical methods and data analysis techniques
• Familiarity with data analysis libraries and frameworks (e.g., Pandas, NumPy, Scikit-learn, TensorFlow)
• Ability to interpret and visualize complex datasets
• Basic understanding of accelerometer functionality