[论文解读] A Hazard Analysis Technique for Additive Manufacturing

Gregory Pope, Mark Yampolskiy|arXiv (Cornell University)|May 29, 2017

Manufacturing Process and Optimization被引用 11

一句话总结

本文提出一种基于系统理论过程分析（STPA）的危险分析技术，用于识别增材制造（AM）中的网络物理风险，重点关注软件漏洞、知识产权盗窃、零件可靠性及网络安全性。研究展示了STPA在识别软件密集型AM系统中潜在危险方面的有效性，为在关键任务应用中部署前主动缓解风险提供了前瞻性框架。

ABSTRACT

The promise of Additive Manufacturing (AM) includes reduced transportation and warehousing costs, reduction of source material waste, and reduced environmental impact. AM is extremely useful for making prototypes and has demonstrated the ability to manufacture complex parts not possible (or prohibitively expensive) with conventional machining. Scientists and manufactures are finding increased uses for AM in creation of all types of finished products including those built from polymers, biological material, and metals. Although companies such as GE have been using 3D printing for Additive Manufacturing for over thirty years to make mandrels for light bulb manufacturing, application areas of Additive Manufacturing have increased substantially in recent years, particularly due to the reduction in cost of 3D printers. Like most emergent technologies, there are bound to be growing pains with AM. This paper looks at the software that supports AM and 3D printing and their vulnerability to cyber-attacks, intellectual property theft, defect rates of AM software (which can cause undesired consequences themselves and also create vulnerabilities that a hacker may exploit), part reliability and safety of devices incorporating 3D printed parts (when making mission critical parts), and security/throughput issues of computer networks. Literature searches, consulting with technical experts and a relatively new hazard analysis technique will be used, one especially developed for software intensive systems called Systemic Theoretic Process Analysis (STPA). The purpose of this white paper is to identify risks (or hazards for mission critical parts) for AM in this emergent stage so that mitigations can be applied before accidents occur. A second purpose of this white paper is to evaluate the effectiveness of STPA as a hazard analysis technique in a field that is still relatively new.

研究动机与目标

• 识别增材制造系统中的潜在危险，特别是由软件漏洞和网络威胁引发的危险。
• 评估系统理论过程分析（STPA）作为新兴AM环境中危险分析方法的适用性与有效性。
• 通过在故障发生前发现隐藏的系统级风险，支持关键任务AM应用中的主动风险缓解。
• 应对增材制造工作流程中日益增长的零件可靠性、知识产权盗窃及网络安全问题。

提出的方法

• 应用系统理论过程分析（STPA），一种专为复杂、软件密集型系统设计的危险分析技术。
• 识别AM流程中的控制结构，以分析故障传播和不安全控制行为。
• 结合文献综述与专家咨询，以支持危险识别并验证研究发现。
• 利用STPA的分层控制结构，对AM系统中软件、硬件和网络组件之间的交互进行建模。
• 分析AM软件中的漏洞，包括缺陷率和可利用的弱点，这些弱点可能导致零件失效或网络攻击。
• 评估STPA在识别与知识产权盗窃、零件可靠性及网络吞吐量问题相关的危险方面的表现。

实验结果

研究问题

• RQ1 STPA如何有效应用于具有复杂软件依赖关系的增材制造系统中，以识别危险？
• RQ2 AM软件缺陷和网络漏洞会引发哪些特定的网络物理风险？
• RQ3 STPA在多大程度上能发现传统危险分析方法可能遗漏的AM系统中的潜在危险？
• RQ4 知识产权盗窃和零件可靠性问题在关键任务AM应用中如何表现为系统性风险？
• RQ5 STPA能否作为新兴技术（如增材制造）中可行且全面的危险分析框架？

主要发现

• STPA成功识别出AM系统中的多个潜在危险，特别是与软件控制失效和网络攻击向量相关的危险。
• AM工作流程中的软件缺陷被发现显著增加了零件可靠性风险，且可能被恶意行为者利用。
• 由于增材制造设计文件的数字化特性及其在网络中的传输，知识产权盗窃被识别为一种系统性风险。
• 网络吞吐量和安全问题被证明会影响AM过程的完整性和安全性，尤其是在实时或关键任务环境中。
• STPA的应用揭示了AM系统中此前未被考虑的故障模式，证明其在发现复杂、相互依赖的风险方面具有重要价值。
• STPA在分析软件密集型系统方面表现出色，提供了一种超越传统方法的结构化危险识别途径。