The smart city model
Standardization and aggregation
To compare the different indicators it is necessary to standardize the values. One method to standardize is by z-transformation (see formular). This method transforms all indicator values into standardized values with an average 0 and a standard deviation 1. This method has the advantages to consider the heterogeneity within groups and maintain its metric information. Furthermore a high sensitivity towards changes is achieved.
To receive results on the level of factors, characteristics and the final result for each city it is necessary to aggregate the values on the indicator level. For the aggregation of a respective group of indicators to domains we consider also the coverage rate of each indicator. A certain result from an indicator of an indicator covering all 77 cities weights therefore a little more than from an indicator covering only for instance 60 cities. Besides this small correction the results were aggregated on all levels without any weighting. The aggregation was done additive but divided through the number of values added. That allows us to include also cities which do not cover all indicators. Their results are calculated with the values available. Still, it is necessary to provide a good coverage over all cities to receive reasonable results.