It is estimated that the 60% of the world population will live in cities by 2030 (United Nations, 2016). This assumption with the dawn of environmentalism and concerns regarding resource depletion, the oil crisis in the 1970s and global climate change brought the beginnings of a discourse on building form and energy consumption. Consequently, an increasing attention for energy conservation in buildings has been given in recent years. As a reaction to climate change, nowadays improving the energy performances of cities has become an important topic in the agenda of governments and decisions makers. In order to build sustainable cities, considerations in urban planning have to be made.

Therefore, this dissertation born with the objective of evaluating the buildings heating energy consumption at multi-scale through the consideration of variables at building and urban scale. The purpose of this study is to perform a multiple linear regression model in order to evaluate the heating energy consumption of a large part of buildings of Turin and apply a cluster analysis to find buildings with similar energy consumption and identify the characteristics that make to each group consumed the specific amount of energy.

The analysis was developed using a GIS and statistical software. ArcMap was the GIS software used for the association of different sources and their georeferenced while the statistical software allows the implementation of different statistical techniques as principal components, multiple linear regressions, and clustering algorithms in order to achieve the purpose of this study.

The dissertation is presented into ten chapters. Chapter 1 introduce the problem and stabilize the aim and objective that the study would cover. On the other hand, chapter 2 provides a background to the issue of sustainable development concept, tracing its origins and evolution and how policymakers begun to address the issue of energy efficiency. The chapter 3 describes the methodology applied, the different datasets available and explain the indicators that measure the built environment. Chapter 4 will focus on the case study, Turin giving a brief evolution of the city and describing the climatic conditions and the characteristics of the building stock. Besides, the chapter 5 will analyze the data input while the chapter 6 will focus on the statistical approaches and the chapter 7 will be dedicated to the model applied in order to understand the most influent variables at building and urban scale in energy consumption. Chapter 8 is dedicated to the cluster analysis for identify groups of building while chapter 9 will discuss the results and the research implications of the work and finally the conclusions are presented with general remarks and proposing further improvements to the methodology.