Over the last several years, a technological convergence-increasing processing power and memory capacities, component software infrastructure developments, and adaptive agent architectures-has made possible the development of a class of simulation and optimization tools that was unimaginable earlier. With careful design, these tools can be easy to use for nonexpert users, and they can provide a level of decision support and insight that can potentially have tremendous impact in terms of problem understanding and performance. This technology is particularly relevant to the electricity enterprise as it undertakes the difficult transition toward deregulation and open competition. Our current work, as realized in SEPIA, constitutes little more than a first step in this context, but we hope it can serve as an initial demonstration of the power of a complex adaptive systems approach for challenging practical applications.