Dual task scheduling strategy for personalized multi-objective optimization of cycle time and fatigue in human-robot collaboration

Abstract

The increased adoption of collaborative robots into human dominated work cells improves production efficiency of hard-to-automate manufacturing tasks. System-centric task scheduling objectives aim to fully utilize all available agents, often physically burdening workers with schedules designed maximized efficiency. Therefore, the modeling and optimization of human fatigue, as one of the main contributors to efficiency decline and adverse health conditions, is significant to task scheduling in Human-Robot Collaboration (HRC). Furthermore, HRC teams often involve drastically different workers with differing capabilities and muscle strengths – with direct implication toward their personalized fatigue responses. As such, we present a dual scheduling strategy for personalized multi-objective optimization of cycle time and fatigue considering recovery in HRC. The scheduling strategy involves two fatigue minimization objectives which either minimize the team’s fatigue state or an individual worker’s fatigue state. This balances fatigue accumulation between workers, providing targeted rest and recovery to specific workers while maintaining production efficiency. We designed a custom NSGA-III genetic algorithm for simultaneously minimizing cycle time and fatigue. Our model and algorithm are applied to an HRC assembly case and show promising results in redistributing tasks between agents to minimize personalized fatigue. © 2023 The Author(s)