In future power systems different agents will take responsibility for different parts of the system. Since any decision made by an agent affects decisions made by other agents, the agents need to collaborate with each other for reliable operation of the whole grid. Having a central management entity that gathers all information may not be viable as agents are not willing to share their information with other parties. Meanwhile, formulating a centralized optimization leads to a complex model that is difficult to solve. Hence, the existing centralized energy management functions may no longer be appropriate for management of such a multi-agent distributed system. This panel will discuss challenges and potentials of distributed optimization and multi-agent systems for management of power systems, including: challenges and applications of distributed computing algorithms for power transmission and distribution systems management; challenges and potential solutions for implementation of multi-agent systems in smart grids; and distributed algorithms for collaborative management and market regulation of TSOs and DSOs