Task Allocation in Human-Robot Collaboration (HRC) Based on Task Characteristics and Agent Capability for Mold Assembly

Abstract The involvement of repetitive motion with force and heavy component handling during the manual mold assembly causes an ergonomic problem that harms the musculoskeletal system on the human worker. This ergonomic problem can be eliminated using a robot. However, the traditional automation system is impractical to be implemented in the mold assembly because of the complexity of the customized mold structure. Hence, human-robot collaboration (HRC) systems became the potential solution to improve the working conditions with the assistance of a robot while adapting to the frequent change over in the mold assembly operation. In this paper, we propose a task allocation model for the HRC system that consists of three agents who are one human and two robots to cope with heavy part handling and the diversity of tasks in the mold assembly. First, we decompose the assembly operation into functional actions. Second, we evaluate the action assignment using the Analytic Network Process (ANP) based on the action characteristics and agent capability. Third, we allocate the functional action to the agents in the HRC work cell using the Genetic Algorithm (GA) to minimize the operation cycle time and maximize the agent capability. The outcome of this paper is to contribute to the expansion of the HRC system with one human and two robots in the mold assembly operation. In addition to the improvement in the ergonomic condition, the proposed task allocation model can achieve a wider variety of tasks assigned to robots with minimum tool change.