A number of methods for predicting the rate of technological development have been developed. The most famous of these is Moore’s Law, named after Intel’s co-founder Gordon E. Moore who in 1965, wrote that the number of components in integrated circuits had doubled every year and predicted that the trend would continue. The prediction has proved accurate ever since.
Moore’s Law has been applied to other forms of digital technology, which have been found to develop exponentially but at different rates. For example, hard disk storage has been found to half in cost every 1.1 years whereas RAM halves in cost ecery 1.6 years.
Now, a group of MIT researchers have tried applying the technique to a range of technologies, including in the chemical and energy industries, as well as digital technology.
They found that they were able to accurately “hindcast” the rate of development that has occurred in these industries using Moore’s Law with different doubling periods.
For example, the study found that the cost of aluminium production halves every 67 years, acrylic fibre every 6.8 years, beer every 20 years and sugar every 150 years.
In the energy industry, the cost of crude oil halves every 68 years, combined cycle gas-fired electricity every 34 years, geothermal electricity every 13 years and wind electricity every 7.5 years.
Using their estimate that the cost of residential PV halves every 6.7 years, the researchers estimate that it will cost about 6 cents per kilowatt-hour in 2020 and just 2 cents per kilowatt hour by 2030. This implies that industrial-scale PV will be about 4 cents per kilowatt hour in 2020, compared with about 5 cents per kilowatt hour for wholesale coal-fired electricity at the plant (in the United States).
The graph shows their projected residential PV electricity costs (The dotted line indicated the projected error range.)