Abstract The Resource-Constrained Multi-Project Scheduling Problem (RCMPSP) has been an essential topic of study for the past two decades. This paper presents the generalized RCMPSP(ABD) problem, where category A refers to project activities that can be implemented using fixed resources over fixed durations, category B refers to project activities that can be performed using resources as category A but the non-preemptions are relaxed, and category D refers to project activities that can be executed using flexible resources over flexible durations and may or may not be interrupted. The activity types A, B, or D related to resource utilization and interruptability are known for each activity. After blending the activity types, A, B, and D into the generalized problem, solution heuristics for the RCMPSP(ABD) are presented. Experimental testing shows that substantial reductions in the makespan of the entire project are possible, while also freeing up resources to be used on other projects. A mixed integer-linear program (MILP) formulation is also presented to define this combined problem. This problem has widespread application in practice, for example, in One-of-a-Kind Production (OKP) and construction environments.

中文翻译：

---

解决不同活动假设下资源受限的多项目环境

摘要 资源受限的多项目调度问题 (RCMPSP) 在过去的二十年里一直是一个重要的研究课题。本文提出了广义 RCMPSP(ABD) 问题，其中 A 类是指可以使用固定资源在固定期限内实施的项目活动，B 类是指可以使用 A 类资源执行但不可抢占的项目活动D 类是指可以在灵活的期限内使用灵活的资源执行的项目活动，可以中断也可以不中断。与资源利用和可中断性相关的活动类型 A、B 或 D 对于每个活动都是已知的。在将活动类型 A、B 和 D 混合到广义问题中后，提出了 RCMPSP(ABD) 的求解启发式。实验测试表明，可以大幅缩短整个项目的完工时间，同时还可以释放资源用于其他项目。还提出了混合整数线性规划 (MILP) 公式来定义这个组合问题。这个问题在实践中有着广泛的应用，例如，在 One-of-a-Kind Production (OKP) 和建筑环境中。