Abstract
Statistical Analysis The foregoing examples are obviously very important ones, but they are rather too complex to illustrate methods of operational research. For this purpose a different approach is necessary. In most problems, there are a number of variables to be considered. Our first object is to find the correlation between these variables, our second to test the stability of this correlation by finding the causal link lying behind it, and our final object is to estimate the likely consequences of any particular change that may be imposed upon the system. We thus provide ourselves with a quantitative basis for any decision to be made regarding the change. For example, how does the number of ships sunk in a convoy depend upon the speed of the convoy? Is the speed the direct cause of the effect? How many ships are likely to be saved if the speed is increased? What will be the cost resulting from the elimination or re-routeing of slow ships, etc.?
Article PDF
Rights and permissions
About this article
Cite this article
Methods of Operational Research. Nature 161, 379–381 (1948). https://doi.org/10.1038/161379a0
Issue Date:
DOI: https://doi.org/10.1038/161379a0