In the 21st century, the urgent educational demand for cultivating complex skills in vocational training and learning is met with the effectiveness of the four-component instructional design model. Despite its success, research has identified a notable gap in the address of formative assessment, particularly within computer-supported frameworks. This deficiency impedes student self-awareness of skill mastery and limits effective monitoring of skill learning in the classroom by teachers. To address this gap, the study introduces an enhanced four-component instructional design model that seamlessly integrates formative assessment. Based on this model, an automated system for assessing complex skills was developed, with the aim of formative assessment and improving skill learning. A control experiment involving 54 industrial robot professional participants in vocational colleges has preliminarily verified the feasibility and effectiveness of computer-supported formative assessment. The findings reveal that this approach significantly enhances students' schema construction, knowledge, skill mastery, and transfer ability, thereby improving the overall effectiveness of complex skill learning. In addition, participants who underwent computer-supported formative assessment reported high levels of system satisfaction and usefulness, with no adverse impact on their learning attitudes, motivation, or cognitive load. This study contributes a robust theoretical framework and practical case study for computer-supported formative assessment in complex skill learning, providing empirical support for the advancement of computer-supported teaching. The integration of formative assessment within the four-component instructional design model offers a novel perspective, addressing a critical gap in the existing literature and laying the foundation for future developments in this educational domain.