Maximizing System Flexibility

You use Quality to configure the quality-control environment by identifying each process control and data-collection point. To configure what you want to control and how you want to control the process:

1. Define quality control requirements.

   Quality functions determine which fields are available for data entry or query. They also provides parenting for related process hierarchies and their associated streams. When defining the quality function, you determine the trace and hierarchy fields that are common to that quality application or functional area. When recording inspection data against a sampling, hierarchy fields determine the data that you are entering sampling results against. You use trace fields to enter traceability information, such as the lot number associated with that sampling.

   Measurement plans represent groups of characteristics that form measurement criteria for inspection and control. When defining characteristics for a measurement plan, you determine the properties of the characteristic being measured, the subgroup size, how to display the data on charts, and the quality control rules to apply for each characteristic.

   Control plans bring together application, measurement, and control-and-response criteria for a specific product and process. Up until this point, all elements of configuration that you have established haven't been applied to a specific process. The control plan provides the remaining element of configuration, defining the process and applying methods of inspection and control to that process.

   When defining a control plan, you reference a quality function that indicates the fields to display as hierarchy fields on the plan. Then, for each individual hierarchy field combination, you define the measurement plan that applies.

2. Construct control charts.

   Quality includes standard control charts you use to define specialized charts with formulas, labels, and test zones. Each component of the chart can contain requirements. For example, the point on the chart can be subtracted from the nominal specification value before being plotted, or control limit formulas may contain a special constant to achieve 2.5 sigma control limits.

3. Create diverse control rules.

   If the organization has unique or non-standard procedures for evaluating process control or limit calculations, you can define these same procedures within Quality.

   You can create control tests from Shewhart prototypes and selectively apply the tests to mean or variance charts. The tests are evaluated when performing data-entry functions. You can control the sensitivity of each test by specifying the number of points to be evaluated and the number of points in the pattern required to trigger a violation.

   You can also employ a three-level phased approach for establishing control limits by setting them to automatically calculate at 10, 20, and 30 subgroups. This gives the operation a moderate level of control at 10 to 29 subgroups and fixed control at 30. You can also automatically recalculate at a specified point interval.