This blog makes the case that higher frequency of data collection and data analysis is more important than more data related to supplier evaluation.
Most of the procedures supplier quality engineers write are comprehensive. We all want more data because we believe it will help us make better decisions, but I believe simple systems are often more successful.
For example, the following system uses just two metrics: 1) the percentage of orders delivered on-time, and 2) the percentage of nonconforming lots. The reason why this works well is that companies can generate automated reports quickly and frequently. More frequent reviews of suppliers are more effective than annual comprehensive reviews. This is the same argument used for planning internal audits. Most companies favor monthly internal process audits instead one annual full-quality system audit.
If you want to add more inputs to the supplier evaluation process, think carefully about how difficult it is to collect data and how frequently reports should be generated. Aging of Supplier Corrective Action Requests (SCARs) is a great metric to track, but if a supplier only receives one SCAR per year—how frequently would you run this report?
To be useful, you need metrics that provide almost real-time feedback. Statistical data on process capability for critical dimensions, or final testing results is available for every production lot. These metrics are also likely to change from lot-to-lot. Asking suppliers for this type of data involves weekly or monthly interaction with the supplier. As I stated earlier, higher frequency of data collection and data analysis is more important than more data.
Lot-specific data can be used to justify sampling plans for incoming inspection. You might even ask the supplier to graph the data and provide a brief analysis of the results. These are value-added services that suppliers can provide—and should already be performing in an ISO 13485 Quality System. Regular review of in-process and final inspection data also forces regular interaction with suppliers. The combination of real-time data, applying statistical techniques and communicating regularly with suppliers is the key to successful supplier management.