0 10 20 30 40 50 (e) Stratification
FIGURE 15.3 Interpreting control chart patterns. (Adapted from Evans, J.R., Statistical Process Control for Quality Improvement: A Training Guide to Learning SPC, 1st ed. Copyright 1991. Reprinted by permission of Pearson Education, Inc., Upper Saddle River, NJ.)
be scientifically determined. In many cases, the appropriate CLs may not be readily apparent or available to HACCP team members. Wedding has listed some sources to consult for this information, including scientific research articles, government documents, trade association guidelines, in-plant studies, university extension publications, and industry experts . If outside sources are used to establish CLs, they should be documented and become part of the HACCP plan.
Once CLs based on scientific data have been determined for each CCP, a capability study must be conducted on the HACCP process to ensure it can be realistically and consistently maintained within these defined limits. As noted by Evans, the process must first be in a state of statistical control before performing a process capability study . The major function of a capability analysis is to determine by measurement how well the control measure used at that CCP is functioning when compared to the specifications set at the CL . Establishing CLs that may be outside the capability of the process will ultimately jeopardize the integrity of the entire HACCP plan .
Several texts reference how to conduct a process capability study [2,43]. Assessment of process capability is required to determine the relationship between the natural process variation and specified tolerances. Thus, individual temperature readings for producing safe juices should always operate within the CLs for safe pasteurization temperature. Evans has expressed process capability as the ratio of the tolerance width to the natural process variation. In the context of HACCP, this would be defined as shown in Figure 15.4 . As noted by the CL definition, only one-sided limits are necessary for most HACCP capability studies. When this is the case, the formulae in Figure 15.5 apply. Examples of when this might be appropriate include the maximum pH allowed for an acidified vegetable product where no heat treatment is applied during the process and the product is stored in ambient temperature, or the minimum scheduled process temperature allowed for a low-acid canned vegetable to ensure safety . When a CL is violated, it signals that an unsafe product may have been manufactured at this CCP. Immediate action must be taken to bring the CCP back into its CL range. Also, any product manufactured at the time the CL was violated must be held for evaluation and/or reprocessing.
When CLs have been set for all CCPs, the task is to keep the parameter being measured in control within the established tolerances. This may or may not be an easy job depending on the kind of variation in the process.
_ tolerance width of CL _ CLy - CLl p natural variation 6a
CP = quotient of tolerated variation CLj = upper critical limit of CCP CLl = lower critical limit of CCP
6a = actual process variation, assuming a normal distribution FIGURE 15.4 Calculating upper and lower critical limits at a CCP.
CPu = process capability in relationship to the upper Critical Limit
CLy = upper Critical Limit x = process average 3a = only right side of normal distribution
CpL = process capability in relationship to the lower Critical Limit
CLl = lower Critical Limit x = process average 3a = only left side of normal distribution
FIGURE 15.5 Calculation formulae for a process requiring only one critical limit.
Establishing operating limits is a practical means to help prevent routine violation of the CLs . Operating limits are criteria that are more stringent than CLs and are established at a level that would be reached before the critical limit is violated . Process adjustment should be taken when the operating limit is exceeded to avoid loss of control and the need to take corrective action at the CL.
Was this article helpful?