The answer depends on many issues.
 
1.       Do you feel the need to do any piloting? Or is it well enough known to just skip the entire step(s)?
2.       How many steps are you going to go through? Lab, scale-small, pilot-large, pilot-demo, unit-production unit? Or lab scale-production unit? Or some variation?
3.       How much development is required in each stage?
4.       Can you sequence some stages simultaneously with a higher element of risk/cost? Can you start the pilot design before the lab tests are done? Can you start the demo before the pilot is producing data?
5.       When are you willing to invest in the production design? Based on the lab scale and pay extra if a later stage requires changes? After the pilot plant scale? Only after the demo?
6.       How well known is the technology? If it is novel, you may not be able to do much of anything until all previous stages are complete, analyzed and possibly done again after modifications.
 
So the answer is there is not a general answer. You would need to do a lot of work and thought to see how long it might take based on the assumptions you make. People have reported times from a year to a decade (seriously).