The design of the work area of a robotic cell is currently an iterative process of trial and improvement, where, in the best cases, the user places the workstations and robotic manipulators in a 3D virtual environment to then semi-automatically verify variables such as the robot’s reach, cycle time, geometric interferences, and collisions. This article suggests using an evolutionary computation algorithm (genetic algorithm) as a tool to solve this optimization problem. Using information about the work areas and the robot’s reach, the algorithm generates an equipment configuration that minimizes the cell area without interference between the stations and, therefore, reduces the distances the robotic manipulator must travel. The objective is to obtain an optimized layout of the workstations and to validate this optimization by comparing the transfer times between stations with the actual times of an existing screwdriving cell. As a result, the transfer time was reduced by 9%. It is concluded that the algorithm can optimize the layout of a robotic cell, which can lead to significant improvements in efficiency, quality, and flexibility.
The optimization of robotic cells is a key challenge in the manufacturing industry due to the need to maximize efficiency in limited spaces and minimize operation times. Traditional cell design methods often face challenges due to the high complexity and dynamic nature of real-world applications. In response, this study presents a dual approach to optimize both spatial design and traversal time in robotic cells, using bioinspired evolutionary algorithms. Initially, a genetic algorithm is employed to optimize the layout of the cell elements, reducing space usage and avoiding interferences between workstations. Subsequently, an ant colony optimization algorithm is used to optimize the robots’ trajectories, minimizing cycle time. Through simulations and a digital model of the cell, key metrics such as total space reduction, operational time improvement, and productivity increase are evaluated. The results demonstrate that the combination of both approaches achieves significant improvements, enabling an average reduction of 21.19% in the occupied area and up to 20.15% in operational cycle time, consistently outperforming traditional methods. This approach has the potential to be applied in various industrial configurations, representing a relevant contribution in the integration of artificial intelligence techniques for the enhancement of robotic systems.
The optimization of robotic cells is a key challenge in the manufacturing industry due to the need to maximize efficiency in limited spaces and minimize operation times. Traditional cell design methods often face challenges due to the high complexity and dynamic nature of real-world applications. In response, this study presents a dual approach to optimize both spatial design and traversal time in robotic cells, using bioinspired evolutionary algorithms. Initially, a genetic algorithm is employed to optimize the layout of the cell elements, reducing space usage and avoiding interferences between workstations. Subsequently, an ant colony optimization algorithm is used to optimize the robots’ trajectories, minimizing cycle time. Through simulations and a digital model of the cell, key metrics such as total space reduction, operational time improvement, and productivity increase are evaluated. The results demonstrate that the combination of both approaches achieves significant improvements, enabling an average reduction of 21.19% in the occupied area and up to 20.15% in operational cycle time, consistently outperforming traditional methods. This approach has the potential to be applied in various industrial configurations, representing a relevant contribution in the integration of artificial intelligence techniques for the enhancement of robotic systems.