have a hypothesis testing question which I'm having trouble with, any informative help would be welcome.

There is a hypothetical brewery and it has 5 taps for filling cans (which tap fills a can is entirely random). The can must be filled with 1000units, but above this has to be kept to a minimum to avoid overfilling. Every now and then a red can is sent through the system of blue cans, and the tap used to fill it and how much it is filled with is noted. Therefore I have a set of data with 30 fill amounts for each of the 5 taps.

By analysing the data decide if any tap appears to be filling the cans with more beer than other taps?

For this I have done hypothesis of; H0: u1-u2>=0 and HA: u1-u2<0. Is this correct?

If so, do I use the left tail of a one tailed distribtuion diagram?