This paper proposes a multi-agent based differential evolution algorithm for solving economic dispatch problem considering ramp rate limit and prohibited operating zones. In this work, multiagent system (MAS) concept is integrated with differential evolution (DE) to form a multiagent based differential evolution (MADE) algorithm. In MADE, a member in the population is treated as an agent. They all live in a lattice like environment. Neighborhood competition and cooperation among agents and differential evolution approach make this algorithm reach optimum value quickly. The proposed MADE has been applied to three test systems namely 3unit, 6 unit and 15 unit systems and the results are compared with DE and existing approaches. Simulation results show that MADE gives better value and converges quickly to optimum values than earlier reported methods. Copyright © JES 2009.