Best Fit:

Engineering data normally contains significant amount of error. Hence the interpolating curve passing through all data points is neither essential nor practical. A curve that provides the best fit to the given data is what needed in many cases. The best fit curve represents the general trend of the data (without passing through every point). Several criteria may be used to derive the best fit curve. This is very useful in characterizing the data or in deriving the correlating equation. For instance, correlating equations representing the behavior of I.C. engines under various air-fuel ratios are useful in the design and analysis of automobiles.