Integrating scenarios, i*, and AspectT in the context of multi-agent systems