运输验证的样本量怎么定

结论：样本量取决于失效的严重程度和可接受的风险

运输验证的目的是证明产品经过运输后仍然完好——包装未破损、无菌屏障未失效、产品未损坏、功能正常。

样本量的问题是：测几个才能得出可靠的结论？

答案取决于几件事：失效的后果有多严重（无菌屏障失效对使用者的风险）、可接受的失效率是多少、以及产品和包装的变异程度。

后果越严重、要求的置信度越高，需要的样本量越大。 这是统计上的必然，不是可以商量的。

样本量的考虑因素

失效后果。 无菌产品的包装失效后果严重，样本量要求高；非无菌产品的外观损伤后果较轻。

可接受质量水平。 允许多大的失效比例。

置信度要求。 希望以多大的把握得出结论。

产品与包装的变异。 变异大则需要更多样本才能反映真实情况。

试验的破坏性。 某些检验是破坏性的，样本要预留。

多种包装规格。 如果有多种规格，是否都要测、能否用代表性规格。

成本与时间。 现实约束，但不应当单凭成本就降低样本量。

「代表性规格」的选择要有依据。 通常选最不利的——最大最重的（惯性大、跌落冲击大）、最小最轻的（容易在堆码中被压）、或者结构最脆弱的。选择理由要写下来。

试验的设计

模拟实际运输环境。 试验条件要对应实际的运输方式和环境。

试验序列。 通常按实际运输的顺序组织：搬运—运输振动—堆码—跌落—环境条件。

环境预处理。 高温高湿或低温条件下的预处理，模拟运输中的环境暴露。

振动。 按运输方式对应的振动条件，公路、铁路、航空、海运的谱不同。

堆码。 模拟仓储和运输中的堆叠压力。

跌落。 按包装的重量和尺寸确定跌落高度，各个面、棱、角都要测。

多式联运。 实际运输可能涉及多种方式，试验序列要对应。

「试验序列」比单项试验更重要。 实际运输中损伤是累积的——先经历振动再跌落，与只跌落的结果不同。按序列做才能反映真实情况。

失效判据的确定

试验之后检验什么，要事先定好：

包装外观。 破损、变形、开裂。

密封完整性。 无菌产品的关键项，用完整性检验方法。

密封强度。 运输后是否下降。

产品外观。 产品本身有无损伤。

产品功能。 功能是否正常。

内部检查。 必要时拆开检查内部结构和部件。

「密封完整性」是无菌产品运输验证的核心判据。 包装外观完好不等于密封完好，微小的通道从外观看不出来，必须用完整性检验方法确认。

与实际运输的对应

试验条件要对应实际情况，需要了解：

运输方式。 公路、铁路、航空、海运，以及组合。

运输距离与时长。 影响累积的振动量。

气候条件。 沿途可能经历的温度和湿度极值。

装卸次数。 装卸环节是跌落风险的来源。

堆码方式。 仓储和运输中的堆叠高度。

外包装情况。 是否有外箱、托盘。

建议收集实际运输的数据。 在实际货物中放置记录仪，记录振动、冲击、温湿度，这些数据可以用来验证试验条件的合理性，也可能发现意料之外的严苛条件。

结果的处理

全部通过。 结合样本量给出结论，说明结论的适用范围。

出现失效。 分析失效模式和原因——是包装设计问题、材料问题、还是试验条件过严。

改进后重做。 改进包装或产品后重新验证。

文件记录。 试验方案、样本信息、试验条件、检验结果、结论。

「结论的适用范围」要说清楚。 验证是基于特定的包装、特定的运输条件做的，更换包装或者改变运输方式，结论可能不再适用。

再验证的时机

包装材料或结构变更；产品重量或尺寸变化；运输方式变更；运输路线变更（气候条件不同）；以及定期的再验证。

「运输路线变更」容易被忽略。 原来只走国内公路，现在增加了海运出口，运输的时长、环境、振动条件都不同，原验证不覆盖新情况。

试验项目与对应的运输环节

「低气压」这一项在航空运输中适用，包装内外的压差可能使密封受力甚至破坏，含气体或密封的包装尤其要注意。

包装设计上的应对

运输验证发现问题后，包装改进的常见方向：

增加缓冲。 吸收冲击能量。

改变结构。 避免应力集中在密封处。

加外箱。 增加一层保护。

改变摆放方式。 使产品在箱内受力更合理。

固定产品。 减少箱内的移动和碰撞。

改进密封。 提高密封的强度和一致性。

「固定产品」往往是性价比较高的改进，因为箱内的移动碰撞是很多损伤的直接原因，而固定的成本通常不高。

与稳定性研究的配合

运输验证与稳定性研究可以结合安排：

货架期末的样品做运输验证。 老化后的包装性能可能下降，这时的运输考验更严。

运输后的样品做稳定性观察。 看运输是否影响后续的储存稳定性。

共用样品和取样点。 减少样品消耗。

「货架期末做运输验证」是较严的做法，但它对应的是真实情况——产品可能在仓库存放较久之后才发运。只用新产品做验证，得到的结论偏乐观。

我们的做法

做运输验证时，我们建议按试验序列组织而不是单项试验，因为实际运输中的损伤是累积的。 单项试验合格，序列试验可能失败。

对于无菌产品，我们建议运输后的检验以密封完整性为主，辅以强度检验。外观检查不足以发现微小通道，而微小通道对无菌屏障是实质性的失效。

如果你在规划运输验证，不确定样本量和试验序列怎么定，可以把产品、包装和运输情况发过来一起讨论，或者直接联系：132 4819 8029。能力范围见服务介绍，送检要求见送检要求，其他问题见常见问题。

English version

Conclusion. Transport validation demonstrates that product remains intact after distribution: packaging undamaged, sterile barrier maintained, product undamaged and function normal. The sample size question is how many units must be tested for a reliable conclusion. The answer depends on the severity of failure, meaning the risk to the user from a compromised sterile barrier; the acceptable failure rate; and the variability of product and packaging. The more severe the consequence and the higher the confidence required, the larger the sample must be. That is a statistical necessity rather than a matter for negotiation.

Factors in sample size. Failure consequence, since packaging failure on a sterile product is serious and demands a larger sample while cosmetic damage on a non-sterile product is less so. Acceptable quality level, meaning the permitted proportion of failures. Confidence required in the conclusion. Product and packaging variability, since greater variability needs more samples to reflect reality. Destructiveness of the tests, since destructive checks consume samples that must be allowed for. Multiple pack sizes, raising the question of whether all must be tested or a representative size suffices. And cost and time, which are real constraints but should not be the sole reason for reducing sample size. Choosing a representative size needs a basis, usually the least favourable, being the largest and heaviest with greater inertia and drop impact, the smallest and lightest and so vulnerable in stacking, or the structurally weakest. Record the reasoning.

Test design. Simulate the actual distribution environment, with conditions corresponding to the real mode and environment. Organise a test sequence following the real order: handling, transport vibration, stacking, drop and environmental conditioning. Precondition environmentally, at high temperature and humidity or low temperature, to represent environmental exposure in transit. Apply vibration according to the mode, since road, rail, air and sea have different spectra. Apply stacking to simulate warehouse and transport compression. Apply drops from heights determined by pack weight and size, covering faces, edges and corners. And reflect multimodal transport, since real distribution may involve several modes and the sequence should follow. The sequence matters more than individual tests: damage in real distribution accumulates, and vibration followed by drop gives a different result from drop alone. Only a sequence reflects reality.

Determining failure criteria. Decide in advance what is inspected afterwards. Packaging appearance, covering damage, deformation and cracking. Seal integrity, the critical item for sterile products, assessed by integrity test methods. Seal strength, examined for decline after transport. Product appearance, covering damage to the product itself. Product function, confirming normal operation. And internal inspection, opening the product where necessary to examine internal structure and components. Seal integrity is the core criterion for sterile products: intact packaging appearance does not mean an intact seal, fine channels are invisible externally, and integrity test methods are needed to confirm.

Correspondence with real distribution. Test conditions must match reality, so establish the modes of transport, whether road, rail, air, sea or a combination. Establish distance and duration, which govern accumulated vibration. Establish climatic conditions, meaning the temperature and humidity extremes likely en route. Establish the number of handling operations, which are the source of drop risk. Establish stacking arrangements in warehouse and transport. And establish outer packaging, whether cartons or pallets are used. Collect real distribution data where possible: placing recorders in actual shipments to log vibration, shock and climate validates the test conditions and may reveal unexpectedly severe conditions.

Handling results. All passing, state the conclusion together with the sample size and the scope within which it applies. Where failures occur, analyse the failure mode and cause, whether packaging design, material or overly severe test conditions. Revalidate after improving packaging or product. And document the protocol, sample information, conditions, results and conclusion. State the scope of the conclusion clearly: validation rests on particular packaging and particular distribution conditions, and changing either may render it inapplicable.

When to revalidate. After packaging material or structural changes. After changes in product weight or dimensions. After changes in transport mode. After changes in route, where climatic conditions differ. And at the scheduled interval. Route change is easily overlooked: a product previously shipped only by domestic road and now exported by sea faces different duration, environment and vibration, and the original validation does not cover it.

How we handle it. For transport validation we suggest organising a sequence rather than individual tests, because damage accumulates in real distribution and a product passing individual tests may fail a sequence. For sterile products we suggest leading post-transport inspection with seal integrity supported by strength testing, since visual inspection cannot find fine channels and a fine channel is a substantive failure of the sterile barrier.

Send us the product, packaging and distribution details and we will work out the sample size and sequence. Phone or WeChat: +86 132 4819 8029.