Tel.+49 30 838 62550 Fax.+49 30 838 46029
There is a great need for quality assurance systems that can be used to check the microbiological status of foods. Test strategies of this kind should not be intuitively created however, but should be placed on a statistical basis. In the course of a critical study of the literature the authors show how the FOSTER plan for monitoring Salmonella contamination developed from the non-specific MIL-STD 105D and how it in turn resulted in the "ICMSF Sampling Book". The ICMSF sampling procedures envisage two-class and three-class attributes plans, the latter being largely for the evaluation of bacterial counts. Without calculating the relevant operating characteristic curve and thus making the assessment situation transparent, these three-class plans have been refunctioned variously into three-decision plans. Variables plans, in which the original bacterial counts and their variance are taken into account, are not as commonly used as attributive plans. Sequential methods or control charts have also had little success up to now. In spite of the great efforts that have been made the development of quality assurance systems cannot yet be regarded as being competed particularly as the expenditure involved in microbiological sampling methods too often exceeds the effect achieved.