He seems to be dividing 300 (people) by 15 (min) = 20...

He seems to be dividing 300 (people) by 15 (min) = 20 (people/hour) which is incorrect maths on many levels.

Think he was reaching for a snappy quote to show how 'slow' 15min rapid test could be (against lab tests that take a 1 - 1.5days to get results back..LOL)....and forgot that RDT take 15min to 'read' the result but to administer the test takes only <1min (usually).

No one needs to 'wait' 15min to get next test underway...so dividing above two figures as a linear equation is nonsense. Depending on how many people are available to administer a RDT (or even better if its a self-test that the traveller could use immediately & display result after 15min) + another person to collate results in parallel...any number of people per hour could be tested (88%-96% accuracy based on type).

Yes, requires a well managed process + logistics + personnel chain, especially at busy airports, but it could be done - if people wanted it done. Airports are used to figuring out this type of rapid time based response for mass travel. Other EU/US countries, as we know, are already trialling/doing stuff like this - using variety of RT-PCR, Ag/Ab RDTs + PCR lab test layers.