6.4: Distinguishing specification limits from control limits

Quality and Regulatory Compliance in Bioscience · LibreTexts

0%

Contents

    Text

    Size
    100%
    Theme
    Typeface
    Line length
    Spacing

    Read aloud

    How

    Speed: 1×
    Light up
    Highlight

    Symbols: Mulberry Symbols, © 2018–2026 Steve Lee, CC BY-SA 4.0 (licence, the set), the site's own copies.

    From LibreTexts, under CC BY-SA 4.0, as the source states it; its equations are written in TeX, which the PDF shows typeset. This page at LibreTexts.

    6.4: Distinguishing specification limits from control limits

    Specification limits are established acceptance boundaries for a product or measured characteristic. They come from applicable product requirements, such as an approved specification or customer agreement. Control limits are statistical boundaries used to assess process behavior. They are calculated for the particular chart statistic and Specification vs control limits.png monitoring approach.

    A target is the desired value. It is not necessarily the same as the observed process average. A team may aim for 10.00 mL while its historical average is slightly different. The target, specification limits, and control limits answer different questions and must be labeled clearly.

    An individual result can meet its specification and still signal a process change. Conversely, a stable process can produce values outside specification if its usual spread is too wide or its average is poorly positioned. Recalculating control limits does not improve the product. Revising a specification needs separate scientific justification and the applicable approvals.

    Worked example comparing two kinds of limits

    The following numbers are hypothetical teaching values, not regulatory limits. Assume the company has an approved individual-fill specification of 9.50–10.50 mL. An established chart of individual fills has a center line of 10.00 mL, a lower control limit of 9.80 mL, and an upper control limit of 10.20 mL. Those chart limits are assumed to come from an adequate earlier baseline; they are not calculated from the few values shown here.

    Passing fill can signal change.png

    The 10.28 mL result at sample 7 meets the volume specification but signals a possible change in the process. Follow the signal-response procedure and assess affected work before deciding what adjustment, if any, is justified. The separate 10.62 mL result in the comparison table also fails the volume requirement and needs the applicable nonconforming-product response. A passing volume alone does not establish that a bottle meets every other product requirement.

    Process capability describes how well a stable process can meet its specifications. It depends on both the location and spread of results relative to requirements. Formal capability calculations have assumptions and data needs. This chapter introduces the concept without asking you to calculate an index from a small illustrative dataset.

    Compare like with like. Control limits for subgroup averages describe averages, not the expected range of individual bottles. A narrow average chart cannot establish that every individual bottle is acceptable. The unit, sampling plan, and plotted statistic must be clear before comparing results with any limits.


    This page titled 6.4: Distinguishing specification limits from control limits was last modified on Fri, 06 Nov 2026 23:54:20 GMT and is shared under a CC BY-SA 4.0 license and was authored, remixed, and/or curated by Emalee MacKenzie.