Sample Size

Sample size is the number of units observed or respondents included in a study or inspection. In supply chain management, it represents the count of inspected lots, items, transactions, or personnel records used to make inferences about a larger population with acceptable error and confidence. This definition underlies acceptance sampling, cycle counting, and supplier quality audits, where the sampled subset must statistically represent the whole.

In receiving inspection, sample size determines how many incoming items or spare parts are checked before a lot is accepted or rejected; a supply-chain sampling study reported that for a defective rate of 10%, the minimum sample needed is 139 at 95% confidence and 98 at 90% confidence. On the warehouse floor, cycle-count audits apply the same logic: a statistically chosen sample of bins, SKUs, or transaction records represents the full stock population, and heterogeneous operations (multiple shifts, product families, or supplier tiers) require larger samples to capture variation. In material tracking, sample size also applies to barcode scans and ERP records, ensuring sampled transactions reflect actual inventory movement. Procurement-quality studies sample purchase orders or inspection lots to estimate defect rates and supplier reliability. Survey-based supply chain research uses formulas like Cochran, Krejcie & Morgan, power analysis, or Yamane; structural equation models need at least 200 observations, with 300–500 preferred.