I think we'd all like to see a bit of csv data for a start, so we can see if it's rows or columns or just pure csv. It seems like the approach would be to repeatedly add normalized data to an Access table, and have a table that lists (not in columns) the relevant stop types so that only the ones you want will appear in the report. This would negate the need for manually flagging the records with relevant stop values. That table would also provide only the relevant the stop type values for data input.
Maybe explain the project in greater detail at some point (perhaps after people here have had a look at the data). F'rinstance, why would the final output need to be csv formatted? Maybe it would be simpler if Access could modify your initial csv data instead of grabbing/converting/storing/editing/converting/outputting.
The more we hear silence, the more we begin to think about our value in this universe.
Paraphrase of Professor Brian Cox.