BASIC IDEA: One adaptation of the S-curve is known as the envelope S-curve, which takes into consideration successive generations of technologies that provide the same benefits. The term "envelope" refers to the curve that connects the tangents of the successive individual S-shaped curves.
PROCEDURE: Start by plotting the growth curves for successive generations of a technology. Try connecting the tangents of these curves to form an "envelope" and base the forecast on the extrapolation of the envelope curve.
EXAMPLE: In the graph below, the
sales data for 4k and 16k microprocessors is shown. The dotted line represents
the envelope for these two S-curves which can be used to forecast future generations
There are several other functional forms that can be used for curve fitting, including exponential, double logarithmic, etc. A difficult part of this application is finding the curve that will best fit the available data.
Very subjective and there is the probability of large errors
Major problems, including:
What is appropriate shape of the curve
Which stage is variable in at the present
What are lengths of the curve(s)
Theoretical or practical (including social, economic, etc.) limits to the curve