Multi-stakeholder decision-making in low-energy building design optimization

The climate emergency means that policy makers are now seeking to reduce our carbon dioxide emissions. The construction and operation of buildings is a major source of emissions. Adopting low-energy building design standards has a role in addressing the problem by reducing operational energy demand. However, the decision-making process in building design is complex and involves a wide range of stakeholders with diverse priorities. Decisions made in the early stage of design impact on energy performance and are often made based on stakeholder preferences, without recourse to building simulation. Furthermore, the complex relationship between building geometry, design parameters and building performance can make emissions reduction more challenging, or even impossible, to achieve depending on the decisions made at this stage. This thesis investigates incorporating desirability into computational optimization, as method of condensing multiple criteria into the objective of satisfying the preferences for each of multiple stakeholders. Hence, it is informed by the techniques of MCDM and optimization. It necessitates an ‘a priori’ approach because stakeholders’ preferences are fully integrated into the optimization process. The outcome illustrates the decision-making process as a trade-off between stakeholders, which can be considered as a negotiation process. The research is informed by multi-criterion methods for multi-stakeholder decision-making and their application to achieve a compromise between building aesthetics and energy performance. The limitations of existing techniques are then addressed by incorporating preference profiles into the optimization process to illustrate the results as a negotiation process between decision-makers. It concludes that there are fundamental differences in the design process needed for low-energy building. Energy calculations need to be undertaken early and updated as the design evolves. This aspect necessitates the inclusion of an additional stakeholder: the energy modeller or building physicist. Hence, there are transaction costs that are borne by the client. Existing methods can be adapted to include stakeholder preferences; however, they are not fully integrated into the process and lack the ability to illustrate the trade-offs between decision-makers. Incorporating desirability functions into optimization using an additive method is proposed. The approach illustrates conflict between stakeholders in the design process and identifies contradictions in the preferences of a single decision-maker.