Letter to the Editor—Optimum Preventative Sampling Via Dynamic Programming

In an interesting paper entitled “Optimum Preventative Sampling,” Whittle (Whittle, P. 1954. Optimum preventative sampling. J. Opns Res. Soc. Am.2 197–203.) has discussed the general problem of carrying out a sampling operation—not for the customary purpose of estimating a characteristic of a population but for the purpose of discouraging the arising of defectives. In general the more intense the sampling is, the greater is the deterrent effect on a potential transgressor. We wish to show how the optimization problem to which Whittle's model leads, and which is typical of a large class of related problems, may be treated advantageously by the functional equation technique of dynamic programming (Bellman, R. 1957. Dynamic Programming. Princeton University Press, Princeton, New Jersey.).