Databases Merging

#Purpose

The purpose of the herein presented activity is to merge the Web Database with the latest publicly available EEA database, in order to use and assign registration numbers available at the latter, to each single record of the former.

#Methodology

Two different methodologies have been used:

1. The first, is based on merging the two databases on groups of data with the same range of vehicle mass and CO2 emissions. This methodology has been used and is analytically described in the following two papers:

   • The change in the average European CO2 emissions from passenger cars due to the introduction of the WLTP. A Monte Carlo analysis based on CO2MPAS.

   • A simulation based approach for supporting CO2 reduction policies for light duty vehicles. A case study on WLTP introduction.; under review for publication in the Transportation Research Procedia, 2017.

2. The second, is based on a more "sophisticated" approach, using string matching and clustering algorithms, and tries to group together all records of both databases which refer to each individual vehicle on the market. The on-going work is provided and described on the relevant ipython notebook here, while the raw files of the on-going work are available on dropbox here.

#Results

To be filled.