Model-based Multi-Attribute Collaborative Production Cell Layout Optimization

Abstract

Designing the layout of a work cell remains a challenging problem due to the large search space of possible layouts which is often subjected to a variety of optimization objectives. This paper proposes a model-based methodology to aid the layout design of a collaborative work cell. This model takes into account feasibility, reachability, safety and ergonomics as constraints in the layout optimization formulation. Determining reachability and static placement feasibility is accomplished using 3D Computer Aided Design (CAD) files as inputs combined with inverse kinematics. Geometric characteristics, like bounding boxes, and assembly requirements are determined from the 3D CAD files and used in the layout optimization. Inverse kinematics is used to calculate the reach of the robots and to reliably estimate ergonomic workload of an operator in the work cell. The optimized layout of the work cell is visualized for every start and end point of an assembly task, which allows to provide task instructions to operators and could potentially be used to program the robots and cobots. The methodology is validated on a practical use case.