Adaptive Micro-Break Scheduling Using Reinforcement Learning and Multi-Modal Fatigue Detection

View at publisher Summarise with VIRAT
Authors: Jones Mescarenhas, Julia Punitha Malar Dhas, Deva Priya Isravel
Cited by: 0
References: 19

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Modern digital workplaces demand intelligent solutions, while prolonged computer usage leads to cognitive and physical exhaustion. A framework for adaptive learning is presented in this research, which can determine the intervals for micro-breaks that are most beneficial for user ergonomics and wellness. The proposed method involves the use of multimodal sensing to monitor posture, typing rhythm, mouse movement, and noise levels in the environment, thus allowing very precise identification of mental load and fatigue. Reinforcement learning acts as the decision-making engine that alters its actions in real-time based on continuous user feedback and thus gives personalized and context-aware break recommendations. A hybrid structure merges the current ergonomic standards with adaptive, data-driven refinement to achieve a kind of compromise between scientific reasoning and personal experience. The model considers environmental and behavioral inputs and, while supporting user productivity, it also prevents the occurrence of physical discomfort. A large language model that can explain fast cognitive state assessments based on the surrounding context is integrated into the system and thus increases transparency. The entire framework operates as a continuous feedback loop that customizes itself according to user behavior and workspace mobility, hence ensuring that performance and wellness outcomes are continuously increased.

Fields of Study & Subjects

Ergonomics and Musculoskeletal Disorders Occupational Health and Safety Management Context-Aware Activity Recognition Systems

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
1575–1581
Publication date
21 January 2026
Language
EN