A new risk analysis of Clean-In-Place milk processing

A new risk analysis of Clean-In-Place (CIP) removal of proteinaceous milk deposits is presented and compared with traditional approaches. The aim was to gain insight into random (stochastic) errors that may lead to unexpected failure of an otherwise well-operated CIP plant. CIP failure is defined as failure to remove proteinaceous deposits on wet surfaces in an auto-set cleaning time. The method is based on the two-stage, unit-operations model and experimental data of Bird (1992) and Bird and Fryer (1991a,b). A risk factor (p) together with a practical tolerance is defined in terms of the auto-set CIP time to remove a specified deposit and the actual cleaning time as affected by stochastic changes (tT′). This is computationally convenient as it can be articulated so that all values p > 0 highlight an unwanted outcome i.e. a CIP failure. Process behaviour is simulated using a refined Monte Carlo sampling of the temperature distribution of the alkali cleaning fluid (1% w/v NaOH) on a (0.00015 m) proteinaceous deposit together with a specified range of tolerances (5%) on the auto-set clean time. Results reveal that with a cleaning fluid at a nominal mid-range temperature of 50 °C CIP will fail in 4.2% of all operations. On average, this equates to 15 failures per annum. This information cannot be obtained from traditional single point simulation (with or without sensitivity analyses). This new (Friday 13th) risk analysis represents a significant advance because all practical scenarios that could exist are computed. Quantitative evidence reveals that CIP is actually a mix of successful cleaning operations together with failed ones. The insight provided can be used quantitatively to reduce cleaning failures and improve safety through changes to the plant including improved temperature control.

► New stochastic model highlights CIP failure to remove proteinaceous milk deposits. ► Quantitative evidence shows CIP mix of successful cleaning together with failures. ► Represents significant advance over traditional single point analysis. ► Demonstrates benefits of emerging Friday 13th (Fr 13) risk modelling.