Smart Cities

Chapter 16Clustering and Fuzzy Reasoning as Data Mining Methods for the Development of Retrofit Strategies for Building Stocks

Philipp Geyer¹ and Arno Schlueter²

¹Department of Architecture, Faculty of Engineering Science, Katholieke Universiteit Leuven, Belgium

²Institute of Technology in Architecture, ETH Zurich, Switzerland

Chapter Menu

Introduction

Method

Application Case

Data Sources and Preprocessing

Clustering

Fuzzy Reasoning

Mixed Fuzzy Reasoning and Clustering

Postprocessing: Interpretation and Strategy Identification

Comparison and Discussion of Methods

Conclusion

Objectives

• • To familiarize the reader with data mining methods for the development of effective retrofit strategies for a building stock, including energy efficiency measures (EEMs) and automated network identification (ANI) for smart energy networks
• • To identify the benefits and methodological differences between sparse information approaches, that is, the type–age classification, and novel approaches based on information available from building catalogs and databases, measurements, as well as data mining methods in smart city contexts
• • To introduce readers to hierarchical agglomerative clustering and fuzzy reasoning as intuitive data mining methods for strategy development for the building stock
• • To provide a guide for the application of such methods to the readers' own situation

16.1 Introduction

16.1.1 Problem Description

A successful energy transition to a low-carbon built environment requires ...