An access path is an index, view, key or logical file. For performance reasons, most access paths should be maintained *IMMED. Access paths should exist in permanent libraries on the system so that programs and SQL can use them to efficiently retrieve data. Reading a small set of data via an index is much quicker than "table scanning" the file, reading and checking every record.

An access plan is completely different. It applies only to SQL and queries. Access plans are built by the operating system or database engine. Image running the DSPDBR command for a file. Then run the DSPFD command for each of the logical files. Look through the key fields on all of the indexes to see which one you should use in your code. This is basically what the operating system is doing when it builds an access plan. This takes a lot of time and resources. It is important for this information to then be stored in a permanent location. If your SQL is rebuilding access plans 50,000 an hour - performance is suffering.