Cyber-physical systems (CPS) have evolved rapidly, which puts high pressure on project management plans, responding, and, to a certain extent, being intelligent as well in multi-agent engineering-related settings, where system complexity and team dynamics converge. The paper proposes a strategic configuration of an adaptive project management (PM) framework applicable in a multi-agent team working in a cyber-physical product development. As opposed to the traditional static approaches, the suggested framework would incorporate real-time data feedback loops, distributed autonomy, and dynamic role assignment intended to overcome the unpredictability and interdisciplinariness of CPS projects. The framework is based on the ideas of systems engineering, agile management, and artificial intelligence, allowing to maintain the step-by-step coordination of physical and digital parts more effectively, better responding to changes during development. The design encloses expandable agent roles, task priority procedures, and flexible synchronization operations to facilitate transitional flow that occur when the design specifications change or when there are disruptions in the development environment. The performance of the framework is compared to the usual project management techniques as the participant takes part in a simulated case study which involved a distributed team developing a smart embedded device. The findings are evidenced by high gains in task performance efficiency, agent coordination and system flexibility when under dynamic constraints. In addition, the architecture prioritizes strategic decision-making because it directs the project to the overall project goals, as well as the individual agent levels via performance tracking in terms of real-time assessment and predictive analytics. The current study will add a new insight into the handling of CPS product development due to its fill-in the gap in the coordination scenery between the technical design activities and the flexible team coordination methods. The results have beneficial consequences regarding the present day engineering managers who are striving to develop the productivity, resilience, and innovation in distributed multi-agent settings and provide the foundation concerning future improvements that incorporate the functions of machine learning-based decision support and the utilisation of automatic project modification capabilities.
Volume 14 | Issue 1
Pages: 63-70