Accessing Hidden Material Flows in the Swiss Built Environment: A Parametric Predictive Model for Building Systems’ Components Reuse Estimation

Abstract

Abstract The global construction industry plays a pivotal role in resource consumption and greenhouse gas emissions, emphasizing the urgent need for sustainable development practices. Despite this urgency, a major challenge lies in the lack of effective resource utilization models that are consistent with the principles of a circular economy. This gap hampers progress towards sustainable resource management, especially given the trend towards urbanization and growing material needs. To address this challenge, our study introduces a Parametric Prediction Model (PPM) designed to improve resource utilization efficiency with a special focus on building systems materials, which are often neglected in previous studies. The model employs a building-by-building approach to accurately assess material inventories within the building system using local databases, thereby increasing the granularity of system composition data. By utilizing state-of-the-art machine learning algorithms, i.e. linear regression and neural network, the model can handle both categorical and non-categorical data. To illustrate the effectiveness of the model, we show a sample material flow analysis schematic using a typical radiator assessment as an example. The schematic provides predictive and concise information about material flow at aggregated level as well as individual details such as location and timing at local level. The resulting output - a refined and comprehensive database - contributes to more informed decision making in sustainable resource recovery and allocation. In addition, this contribution is aligned with broader goals, including waste minimization and resource efficiency in the built environment.