中文内容由 Google Cloud Translation 自动生成，仅供参考；请以英文原文为准。问答资料来自欧盟化学品管理局（ECHA）官方网站。
答案（中文）
ECHA 利用所有可用信息对物质进行分组。该算法使用 IUCLID 档案中所有注册、成分和参考物质的所有可用标识符。因此，在提交信息允许的范围内，所有结构相关的物质都很有可能被识别。然而，对于某些物质，例如无机 UVCB，物质分组可能不太准确，可能需要额外的人工工作来微调方法并手动验证结果的准确性。ECHA 的物质分组旨在尽可能地包容和稳健。例如，分子结构上被认为对分组无关紧要的细微差异可能具有毒理学相关性。对物质组的监管需求评估可能会考虑这些因素，在这种情况下，可能会将物质组拆分为更小的子组，或者对物质进行单独评估。因此，物质组的稳健性取决于我们所处的流程/阶段以及该阶段所需的确定性水平。因此，在迭代的物质组评估过程中，会重新考虑物质组成员资格。
Answer (EN)
ECHA utilises all available information for grouping substances. The algorithms use all available identifiers in all registrations, compositions and reference substances in IUCLID dossiers. Hence, it is quite likely that all structurally related substances are identified, to the extent that the submitted information allows to do so. However, for certain substances, such as inorganic UVCBs, substance grouping may be less accurate and may require additional manual effort for fine tuning the approach and manually verifying the accuracy of the results.
Substance grouping by ECHA is designed to be inclusive and is robust to the extent that is possible. E.g. small differences in molecular structure that are considered insignificant for grouping can be toxicologically relevant. The assessment of the regulatory needs of the group may include such considerations, in which case it is possible that the group will be split in smaller subgroups, or the substances will be assessed individually. Hence, the robustness of the group depends on the process/stage we are at and at the level of certainty this stage requires. Therefore, group membership is reconsidered throughout the iterative group assessments.