Aided by the capabilities of additive manufacturing in building a part with multiple materials, dynamic sub-components, and complex geometries, the number of parts that are feasible for consolidation has increased drastically. However, to decide which components to consolidate is difficult. Therefore, to identify these potential candidates out of a complex product is highly demanded. We define this issue as a part consolidation candidate detection (PCCD) problem. To solve this problem, we proposed three principles that rationalize the PCCD process with regard to the maximum number and the priority of parts to be consolidated. Based on which, we developed a modularity-based PCCD (MPCCD) framework which is featured by the need for module division and community detection as well as two PCCD algorithms [i.e., strength-based numerical PCCD (NPCCD) and community-based PCCD (CPCCD)]. Two case studies of a throttle pedal and an octocopter are given to demonstrate the effectiveness of the proposed CPCCD algorithm and the MPCCD framework, respectively. In the end, this paper is wrapped up with important conclusions and future research.

中文翻译：

---

一种考虑模块化的基于数值的零件合并候选检测方法

借助增材制造在构建具有多种材料、动态子组件和复杂几何形状的零件方面的能力，可进行整合的零件数量急剧增加。但是，很难决定要合并哪些组件。因此，非常需要从复杂的产品中识别出这些潜在的候选者。我们将此问题定义为零件合并候选检测 (PCCD) 问题。为了解决这个问题，我们提出了三个原则，在最大数量和要合并的零件的优先级方面使 PCCD 过程合理化。在此基础上，我们开发了一个基于模块化的 PCCD (MPCCD) 框架，其特点是需要模块划分和社区检测以及两种 PCCD 算法[即，基于强度的数值 PCCD (NPCCD) 和基于社区的 PCCD (CPCCD)]。给出了油门踏板和八轴飞行器的两个案例研究，分别证明了所提出的 CPCCD 算法和 MPCCD 框架的有效性。最后，本文总结了重要的结论和未来的研究。