Dear Editor:

We are surprised to see that Prof. Peleg feels that our manuscript has anything to do with opposing his ideas in microbiological modeling. We are not microbiological modelers. We have no preconceived opinion and we have never taken a stand in the manuscript by choosing one microbiological model over another. We simply report the uncertainties in process calculations for a simpler and a more complex model. The uncertainty in prediction of the log reduction is only one of many important considerations in choosing a model and this choice is left to the reader. As a matter of fact, we believe a model with more parameters (such as Weibull) will fit any set of data better than a model with lesser parameters (such as first-order). It is interesting that Prof. Peleg actually mentions in his second comment (see below) that our work “can be useful to assess their [survival parameter's] safety implications.” That is all what this paper is about; that is, our work is not about proving superiority of one model compared to another.

For specifics, we could find only 3 places in Dr. Peleg's letter that our work (Halder and others 2007) is directly quoted and those are the only instances we can reply to. The rest of the comments in Prof. Peleg's letter are irrelevant to our paper and misdirected. We refuse to be dragged into a controversy that is not the focus of our paper.

  • 1
    First comment (referring to the last line of first paragraph that variability should not be a consideration in comparing models). To quote a very recent work (Nauta 2006) of a microbiological modeler, “…predictive models cannot provide exact predictions of what will happen in food microbiology. This is not surprising because, by nature, uncertainty and variability play an important role in both foods and microbiology…” Thus, inclusion of variability in a process model that includes microbiological parameters should be a major consideration. To discuss model without including the variability, when the latter is omnipresent in microbiology, is like telling half truth. The entire field of stochastic calculus (Itoh 1951) has developed to answer questions of variability and uncertainty. Only a few studies in food have included such variability (for example, Lenz and Lund 1977; work by Nicolai and others, such as Nicolai and others 2007) in the past simply due to the lack of data on the variability of the input parameters and the less sophisticated nature of analysis in general– more needs to be done. This is also implied in Nauta (2006) and is one of the specific outcome of the brainstorming during the IFT Research Summit (Heldman and Newsome 2003) that concluded, among other things, “…Statistical methods to account for variability in measurements and analysis should be incorporated into process calculation procedures and validated.”
  • 2
    In the second comment “The same can be said about uncertainties and variations in the temperature histories of individual cans.” Some of the possible sources of uncertainties in temperature history are thermal properties and heat transfer coefficient. These were in fact included in our work in addition to the uncertainty in microbiological parameters!
  • 3
    The third comment starting with “Halder and others' suggestion that [some of the] nonlinear semi-logarithmic survival curves might be a result of..” We are simply quoting published work (Stringer and others 2000) from a respected journal in a context other than comparing 2 models. Prof.Peleg is free to disagree with published work but we only cited it as part of literature review.


  1. Top of page
  2. References
  • Halder A, Datta AK, Geedipalli SSR. 2007. Uncertainty in thermal process calculations due to variability in first-order and Weibull parameters. J Food Sci 72(4):E155E167.
  • Heldman DR, Newsome RL. 2003. Kinetic models for microbial survival during processing. Food Technol 57(8):406,100.
  • Ito K. 1951. On Stochastic Differential Equations. American Mathematical Society, New York .
  • Lenz MK, Lund DB. 1977. The lethality-Fourier number method: experimental verification of a model for calculating temperature profiles and lethality in conduction heating canned foods. J Food Sci 42:9891005.
  • Nauta MJ. 2006. Uncertainty and variability in predictive models of microorganisms in food. In Modelling Microorganisms in Food. Edited by SBrul, SVGerwen and MZwietering, CRC Press, Boca Raton , Fla .
  • Nicolai BM, Scheerlinck N, Hertog MLATM. 2007. Probabilistic modeling. In: SablaniSS, DattaAK, RahmanMS, and MujumdarAS, editors. Handbook of Food and Bioprocess Modeling Techniques. CRC Press, Boca Raton , Fla.
  • Stringer SC, George SM, Peck MW. 2000. Thermal inactivation of Escherichia coli O157:H7. J Appl Microbiol 88:79S89S (Suppl. S2000)