The research has been focussing on three integrative aspects for the development of such tool:
1. An intuitive feedback on environmental performance in early design stages.
2. The assessment of flexibility in regard to the impact of alterations on the service life-time of buildings.
3. The integration of available Life Cycle Inventory (LCI) data into predictive modelling to generate a range of probable outcomes indicated as Global Warming Potential (GWP, usually measured in kg of CO2 eq per kg of material).
The research has been constrained to residential typologies and specifically to hybrid concrete-timber construction systems in order to establish an applicable methodology and the respective mock-up tools.
Firstly, the mock-up tool for simplified Life Cycle Assessment (LCA, measured in Global Warming Potential GWP) demonstrates how designers can get an intuitively legible and visual feedback to systematically compare the environmental performance of alternative design iterations at initial design stages. To enable such a systematic comparison of design variants the workflow follows an 'Open Building' approach and segments designs into permanent support and adaptable infill systems. Here, it goes further and differentiates into various degrees of flexibility in each of the main systems: concrete construction systems would normally be used as permanent support structures such as cores (containing circulation, infrastructure, and service functions), whereas partly load bearing components made from timber could be adapted and changed over long periods of time, and hence they would have a basic degree of flexibility. The infill systems are also further distinguished into conventional partitioning systems (such as drywalls that already have a higher degree of flexibility, but to the cost of their destruction) and modular systems that can be altered in various combinations. A Shoebox approach was adopted for the visual representation of a building concept in a simplified and intuitively legible interface. In the Shoebox representation, a series of dynamically alterable modules represent alternative load-bearing systems and variable material fractions. These are linked to a simplified parametric building model to extract data for the comparison of GWP results.
Secondly, it focussed on the flexibility of buildings so that they can respond to demands for functional changes and make use of a systematic differentiation between permanent support- and adaptable infill components. As mentioned, both systems are further categorised into kits of parts with variant degrees of flexibility. Predictive mathematical models are used to translate cycles of changing demands during service life-time into varying configurations within the infill system. The degree of flexibility is used to specify the potential service lifetime extension of buildings, helping decision-makers to decrease the overall environmental impact of a building.
Thirdly, the research outlined the concept for a predictive mathematical model to overcome current challenges in the available LCI data: 1.This data is mostly specific to particular markets and regions only; 2.Varying Life Cycle Assessment (LCA) methods are applied, which makes it difficult or even impossible to compare; 3.There are huge gaps in the data in those global regions that undergo the fastest growth in urbanisation. The methodology of this workflow suggests the collection of data from existing buildings with simplified input data (reduction to the essential parameters), integration of the information into Bayesian Neural Network models (capable of machine learning by updating its predictions with the availability of more current data) and predicting a range of possible outcomes to help a user make more informed decisions.
