The most popular and comprehensive Open Source ECM platform
Let us know more
Technology: Can Data from Patent Registrations Predict the Rate of Future Innovation?
For more than a decade a group of researchers at MIT have tried to develop a method to predict the possibility of up-coming innovations and how long it may take before the capability can be realized.
They began with an observation, or actually an extension of Moore’s law applied to semiconductors which has been followed extensively over the last fifty years. The idea is that Moore’s law is very applicable to almost any type of technology, not just semiconductors. What does differ is the time period over which incremental improvements are made.
Moore’s law for semiconductors says that every 18-24 months that the number of transistors packed into a computer processor chip will double. The group at MIT led by Chris Magee said the same kind of trend is seen with LEDs which improves by a factor of ten every decade and batteries which improve performance by a factor of ten every 40 years.
The idea is that we can’t predict how we’ll get there, just that somehow we’ll find a way and the cycle of improvement is typically very consistent. But, how do we define a technology and determine what these cycles are?
A paper in late 2021 by the group at MIT suggests that looking at activity in patent registration could provide a clue. After a review of patents, the group identified 1757 different possible technology domains. But the problem of prediction and trend identification is limited by the number of patents that span some of these fields.
Anuraag Singh, lead author of the paper, said that “the rate of improvement can only be empirically estimated when substantial performance measurements are made over long time periods. In some large technological fields, including software and clinical medicine, such measures have rarely, if ever, been made. Our method provides predictions of performance improvement rates for nearly all definable technologies for the first time. Fast-improving domains are concentrated in a few technological areas. The domains that show improvement rates greater than the predicted rate for integrated chips — 42 percent, from Moore’s law — are predominantly based upon software and algorithms. Technologies that improve faster win the market”
