Urban systems exploration: A generic process for multi-objective urban planning to support decision making in early design phases

The competition for limited urban spaces due to the growing demand for both living space and green areas causes conflicts in urban development. This paper aims to support urban planners with a systematic multi-criteria decision making process and a supporting tool-chain. The process uses multi-objective optimization (MOO) to quantify trade-offs, allowing planners to gain a broad perspective on optimal solutions. The generic process builds an urban simulation model, applies MOO, explores the results in terms of multi-criteria trade-offs, and guides the decision making of urban planners. To enable this process, we present a tool-chain for applying a Gaussian Process Regression based MOO algorithm to a computationally expensive urban simulation model. The tool-chain allows to identify Pareto-optimal solutions and their properties with reasonable computational effort. A case study model is set up in the Grasshopper environment and couples simulation components for outdoor thermal comfort and Life Cycle Assessment. It allows to identify multi-objective trade-offs for a high-dimensional space of urban configuration degrees of freedom such as outdoor vegetation, photovoltaics, and building characteristics. We compare the workflow results to other MOO algorithms and show how it can support decision making in urban planning at early design phases. In our case study, the tool chain was able to investigate the multi dimensional space of urban configurations. It systematically identified Pareto-optimal solutions therein and reduced the number of model evaluations significantly. The case study results showcase the trade-off between lifecycle-based global warming potential (GWP) and outdoor thermal comfort. We identified the number of trees and the coverage of the east and west façades with photovoltaics as the most sensitive parameters. The proposed process proves to be a powerful multi-criteria decision support tool for urban planners. It allows to identify and quantify the Pareto Front of competing urban target trade-offs at early design phases. Additionally, it visualizes them according to the boundary conditions of urban development. The input configurations of the obtained Pareto-solutions serve as a base of urban planning recommendations. In our case study, trees and photovoltaics prove to be good leverage points in the area of GWP optimal solutions. However, urban planners need to carefully coordinate inputs when aiming for a specific trade-off balance. The tool-chain and the simulation model offer further potential for investigating neighborhood typologies. Thereby, applicants can derive scalable guidance to support the sustainable transformation of the urban environment.

Building and Environment