Analyzing the Effects of Remote Work on Productivity and Work-Life Harmony using Federated Learning Models

The change to remote working has heightened the need to look at performance and working hours while not being invasive of the privacy of the employees. The model described in this paper is a federated learning solution to assess the effects of remote work while avoiding collecting all participants’ data in one place to improve data privacy. The model incorporates decentralized data processing and differential privacy approaches to measure productivity and work-life balance. Outcome findings show an enhancement of 17.3 % in task accomplishing percentages, 20% in break usage, 58.3% in after-work communication, and a 20.6% increase in the work-life balance index. The addresses for privacy protection increased by 100% while model performance rose by 9% and data transportation requirements decreased by 75% which led to better security and operation efficiency. The findings suggest that federated learning could help improve outcomes regarding remote work, enabling relying organizations to support their employees while respecting their data privacy.