Human–Robot Collaboration (HRC) plays a key role in improving efficiency and quality in industrial environments. Collaborative robots (cobots) assist operators with specific tasks, making operator safety and well-being a priority, especially when working without physical barriers. The assessment of mental workload is important for the development of effective human–machine interfaces that promote comfort, satisfaction, efficiency and safety. Despite limited research in this area, this study investigates tools for assessing mental workload during collaborative tasks with cobots. Five volunteers were observed during human–human and human–robot interactions in standard and time-constrained assembly tasks using a portable EEG device to monitor cognitive load. Alpha and beta wave averages were analysed for each participant across both phases. This study presents a framework for assessing how human–robot collaboration impacts cognitive load during tasks that exceed safety distances, providing parameters for accurately evaluating cognitive states. However, it is essential to also assess interactions between operators, robots, and the environment to identify anomalies. Reliable data is crucial for understanding workflow dynamics and enabling timely preventive or corrective measures. A more comprehensive parameter matrix, larger participant pools, and more complex tasks will strengthen future studies and improve the assessment of mental workload.