This paper presents a structured methodology for the development of agent-based modeling and simulation (ABMS) systems, aimed at improving modularity and reusability in complex system representations such as Lego® pieces. The methodology addresses the challenge of translating real-world problems into executable ABMS models by defining standardized tables for key model components: environment, agents, interactions, and decision logic. Each element is described using modular structures that facilitate both conceptual validation (by domain experts) and computational validation (through simulation testing). A case study focused on a simplified hospital Emergency Department (ED) illustrates the practical application of this framework, showcasing how agent behaviors and system dynamics can be clearly represented and adapted. The approach emphasizes reusability and automation potential, bridging the gap between conceptual design and code-level implementation.