A major challenge to urban sustainability research today is to understand the urban metabolic profile of cities and predict how cities with different socio-economic, demographic, and geographic characteristics interact with the natural environment in which they exist. Due to the increasing growth in cities and high dense concentrations it is necessary to understand the complexity of the urban socio-economic phenomena, different forms of resource consumption and energy intensities in cities. Data was collected from public and private databases; scales of data included local, regional, and national level. The study applied mathematical modeling and energy flow simulation model approaches: 1. convection-diffusion model [1]; 2. quantitative network model and 3. neural network model, to capture the metabolic flow profile of a system between cities in the Metropolitan City of Milan, Italy. The data analysis combined a set of statistical socio economic, material and energy flow data; and multi-parameter clustering analysis to demonstrate energy and demographic behavioral flow interconnections, similarities and differences of material and energy consumption flows within build clusters. Applied, Group Method of Data Handling approach to forecast material and energy consumption, economic growth and the relationship of the data to understand the territorial metabolism [2] of a city through networks, economies and infrastructure.

AMMODIT and final EUMLS Workshop Mathematics for Life Sciences (http://www.math.uni-luebeck.de/EUMLS-AMMODIT2016/)