It appears to me that your teacher is asking you to determine something that is non-deterministic.
1. The columns are not separated by TAB or COMMA characters (which is very odd for this kind of file).
2. The columns are not aligned ("hours" does not line-up with its column)
3. The only immediate thing left is to count the number of spaces between words, and group those with only one interleaving space character. This is a dangerous solution, simply because someone could very easily (and unknowningly!) reduce the number of spaces between columns to one space when they edit the file...
4. The only other solution is to have an a priori knowledge of the available departments. (That is, your program would have a list of department names which it can match against each line in order to split it up.)
You need to ask your professor for clarification:
1. Can't do.
2. Read column headers and get indices of begin of each column. Each column's data lies between the indices beginning the current and next column (or EOL).
3. Is counting spaces reasonable? (Is it what he wants? [It isn't safe.])
4. Does he want you to have a built-in list of department names and the like?