I'd like to find out what approaches the community have used to achieve similar results using either:

- Data Access Module or;
- NumPy

The following table (**stats_table1**) is my starting point:

I'd like to populate a new table (**stats_table2)** based on the following structure:

for **each row** in the first table:

**stats_table2**[SETTLEMENTNAME] =**stats_table1**[SETTLEMENTNAME]**stats_table2**[SOCIAL_FACILITY] =**stats_table1**[NAME]**stats_table2**[TIME0_15MIN] =**stats_table1**(((TIME5 + TIME10 + TIME15)) / TOTALBUILD)*100)**stats_table2**[TIME15 _30MIN] =**stats_table1**(((TIME20 + TIME25 + TIME30)) / TOTALBUILD)*100)**stats_table2**[TIME30 _60MIN] =**stats_table1**((TIME60 / TOTALBUILD)*100)**stats_table2**[TIME60_PLUS] =**stats_table1**((TIME60P / TOTALBUILD)*100)

Final Results:

i.e.

Data Access Module:

- Would you use a Search Cursor to loop through
**stats_table1,**perform the following calculations and write the results to a python dictionary, then use a Update Cursor to populate**stats_table2**

NumPy:

- Would you convert the
**stats_table1**to a NumPy array, perform the following calculations and write the results into a temporary array and and back to a table to be appended to**stats_table2**

Any sample code or references will be appreciated as I originally was looking at nesting a Search Cursor with a Update Cursor, then realised it was a bad idea.

Richard Fairhurst has a nice blog post titled Turbo Charging Data Manipulation with Python Cursors and Dictionaries that might help with the cursor side of your question.