Abstract The Resource Constrained Project Scheduling Problem (RCPSP) is a combinatorial optimization problem which is non-deterministic polynomial-time (NP)-hard in nature. Due to the diversified applications of RCPSPs, they have been commonly used as scheduling procedures in real-world problems. Since, in practice, project data are prone to changes or disruptions, this paper introduces a mathematical model for a reactive scheduling approach, called the Event Based Reactive Approach (EBRA). This proposed EBRA approach is employed to examine its recovery performance under both a single disruption and a series of independent resource disruptions. Several simulated disruption data are hypothesized to represent real-world disruption scenarios and, without loss of generality, the proposed reactive approach is proved to be efficient in reducing the number of variables and computational complexity and also to be resilient in realistic changes, such as duration inflation and dynamic resource usages. Along with employing an exact method by LINGO software, this paper also proposes an enhanced iterated greedy (EnIG) approach to meta-heuristically solve larger and computationally expensive benchmark instances taken from the Project Scheduling Library (PSPLIB).

中文翻译：

---

基于事件的反应式调度方法解决资源不可靠的资源受限项目调度问题

摘要 资源约束项目调度问题（RCPSP）是一个组合优化问题，它本质上是非确定性多项式时间（NP）难题。由于 RCPSP 的多样化应用，它们已被普遍用作现实世界问题中的调度程序。由于在实践中项目数据很容易发生变化或中断，因此本文介绍了一种反应式调度方法的数学模型，称为基于事件的反应式方法 (EBRA)。这种提议的 EBRA 方法用于检查其在单个中断和一系列独立资源中断下的恢复性能。假设几个模拟的中断数据来表示真实世界的中断场景，并且不失一般性，事实证明，所提议的反应式方法在减少变量数量和计算复杂性方面是有效的，并且在现实变化中具有弹性，例如持续时间膨胀和动态资源使用。除了采用 LINGO 软件的精确方法外，本文还提出了一种增强的迭代贪婪 (EnIG) 方法，用于元启发式地解决来自项目调度库 (PSPLIB) 的更大且计算成本高的基准实例。