Automated guided vehicles (AGVs) are a key technology to facilitate flexible production systems in the context of Industry 4.0. This paper investigates an optimization model and a solution using a decentralized multi-agent approach for a new capacitated multi-AGV scheduling problem with conflicting products (CMASPCP) to take full advantage of AGVs. The novelty of the problem and our model lies in the introduction of AGV capacity constraints and constraints arising from conflicting products, i.e. products that cannot be transported together. As the new I4.0 paradigm tends towards decentralized control, we also present a decentralized multi-agent approach in which AGVs autonomously coordinate to solve the task. The performance of the proposed decentralized approach is compared to a mixed-integer linear programming model on a set of 110 problem instances with different sizes and degrees of complexity. The obtained results show that the proposed decentralized multi-agent approach is effective and competitive in terms of the solution quality and computational time.

自动导引车 (AGV) 是在工业 4.0 背景下促进灵活生产系统的关键技术。本文研究了一种优化模型和一种使用分散式多代理方法解决具有冲突产品 (CMASPCP) 的新容量多 AGV 调度问题的解决方案，以充分利用 AGV。问题和我们的模型的新颖之处在于引入了 AGV 容量限制和冲突产品（即不能一起运输的产品）引起的限制。由于新的 I4.0 范式倾向于分散控制，我们还提出了一种分散的多代理方法，其中 AGV 自主协调以解决任务。所提出的分散方法的性能与混合整数线性规划模型在一组 110 个具有不同大小和复杂程度的问题实例上进行了比较。获得的结果表明，所提出的分散式多代理方法在解决方案质量和计算时间方面是有效且具有竞争力的。