The execution of Belief-Desire-Intention (BDI) agents in a Multi-Agent System (MAS) can be practically implemented on top of low-level concurrency mechanisms that impact on efficiency, determinism, and reproducibility. We argue that developers should specify the MAS behaviour independently of the execution model, and choose or configure the concurrency model later on, according to their target domain's specific needs, leaving the MAS specification unaffected. We identify patterns for mapping the agent execution over the underlying concurrency abstractions, and investigate which concurrency models are supported by some of the most commonly used BDI platforms. Although most frameworks support multiple concurrency models, we find that they tend to hide them under the hood, making them opaque to the developer, and effectively limiting the possibility of fine-tuning the MAS.

翻译：信念-欲望-意图（BDI）智能体在多智能体系统（MAS）中的执行可以实际构建于影响效率、确定性和可重复性的底层并发机制之上。我们主张开发者应独立于执行模型来规范MAS行为，随后根据目标领域的具体需求选择或配置并发模型，同时保持MAS规范不受影响。我们识别了将智能体执行映射到底层并发抽象上的模式，并研究了最常用BDI平台所支持的并发模型。尽管大多数框架支持多种并发模型，但我们发现它们往往将其隐藏在底层机制中，使得开发者难以获知其细节，从而实质上限制了MAS精细调优的可能性。