Enabling Design for Circularity with Computational Tools

Abstract

Circular construction is a design task that requires new datasets and computational tools for matching supply and demand within an urban circular system. Material passports (MPs) contain detailed inventories of materials and products, as well as their specifications, location, and connection details. Circularity indicators (CIs) allow an assessment of a design’s environmental impacts with respect to circularity: the degree to which solutions minimise extraction and waste in favour of reusable, renewable, or recyclable resources both in construction and at end-of-use. Often implemented as an extension to detailed BIM models, MPs and CIs are presently applied in the permit and documentation phases. However, these metrics also establish parameters in early design phases, where circular design thinking and evaluation are most impactful. Circular construction consequently calls for a new suite of design tools that can be integrated into existing workflows, are applicable within the uncertain context of the early design phase, and ideally offer immediate feedback related to formal deliberations, structural considerations, material selection, and detailing. This chapter describes the importance of CIs as design parameters across phases with a special focus on recent early design developments such as the software application RhinoCircular.

1 Computational Tools Enabling Design for Circularity

A crucial step in building the capacity of a circular economy is assessing the quantities and impacts of material inputs, assets, and outputs in the built environment to ensure that building materials remain at their highest utility and value through use cycles. This serves two main purposes: they help understand the carbon impact of construction through the quantification and measurement of a building’s component materials, and they help create a material inventory which allows for their future activation in the form of a material depot. Both tasks require computational tools which can generate data on material quantities and qualities, connections, environmental impacts, recyclability, and reusability (Heisel et al. 2022a).

A solution can be found in the creation of material passports (MPs), digital inventories of material assemblies, and their associated metrics (Heisel and Rau-Oberhuber 2020). MPs document material quantities as well as weights, volumes, dimensions, specifications, and locations of materials within a structure (van Capelleveen et al. 2023). Efforts from companies and organisations such as Madaster, BAMB, DGNB, and the European Union offer platforms for creating, storing, and sharing MPs, often integrated with building information models (BIM) to determine the quantities of components automatically (BAMB 2020; DGNB 2023; Madaster 2023; WBCSD 2023).

Such passports may also report the overall circularity of a construction represented by a circularity indicator (CI), a metric which evaluates both reuse and recyclability of the input material stock and their end-of-use pathways (Heisel and Nelson 2020). The metric itself is a number between 0% and 100%, determined via a set of equations that are dependent on parameters such as life span, efficiency of recycling, and the fraction of feedstock taken from renewable, recycled, or reused sources. Typically, equal weight is given to reused, recycled, or biogenic/compostable material pathways in both the production and end-of-use phase. An initial CI version was developed by the Ellen MacArthur Foundation in 2016 and then further advanced for use in the built environment by Dutch company Madaster in 2018 (Ellen MacArthur Foundation & Granta Design 2019). Building on this work, a variety of definitions and equations on material circularity have been published since. Recent examples are the Urban Mining Index developed at the University of Wuppertal, which takes deconstruction times and difficulties into consideration (Rosen 2022), or the Gebäuderessourcenpass (building resource passport), which is now officially integrated into the certification system of the German Sustainable Building Council (DGNB) (Concular 2023).

CIs are an important reference not only in the design phases of buildings. Since these metrics can inform stakeholders of possible or anticipated pathways for various materials and components, i.e. whether they are structurally sound and/or designed for disassembly, they are also relevant at a building’s end-of-use; in cases of adaptive reuse; when tracking renovations, retrofits, and repairs; or in evaluating recycling or reuse scenarios. Both CIs and MPs are applicable across scales, ranging from construction details to the scale of large urban areas. The tools function in similar ways across these scales; however, the feedback they provide informs different actions and stages of the design process. This makes CIs useful across the entire life span of a building and applicable from early design-to-construction documentation.

Despite the versatility of circularity metrics, present software tools are typically applied in the later stages of building development, when most building parameters are known, leaving little flexibility for informed design. The reason for this lies in the availability of data: at this stage, designs are highly resolved, and information which is needed for calculations, such as detailed product specifications and material quantities, can be computed and stored in MPs. Although they can help evaluate the circularity of a building, CIs generated at this stage rarely actuate significant changes in design or specifications because many critical design decisions have already been determined by the time they are used. Instead, they are often used for compliance to meet technical standards or legislative requirements.

This approach to circularity is effective at creating detailed accounts of a building’s material content for future reference but does not give designers flexibility within the early design phases in making major changes based on design evaluation and feedback. This creates an opportunity for new design tools which can address this phase. We argue that greater impact can be achieved by implementing circularity analysis in the early, pre-BIM design phase. Design is an iterative process, and a building’s design is more prone to change in early phases. More importantly, major changes in early design phases are easier and cheaper to implement (Lawson 2005). As a design develops, decisions are more finely tuned as architects synthesise feedback from clients, engineers, contractors, stakeholders, and various consultants until the design is finalised. Because of this, the earlier circularity analysis occurs within the design process, the greater impact it will have in later design phases. Additionally, it can help reduce the uncertainty of decisions in the early design phase by providing information on the impact of individual decisions through an (ideally rapid) information feedback loop.

Before CIs can be incorporated in early design phases, the strengths and weaknesses of existing computational tools in addressing varying scales should be assessed to better understand how they can achieve the greatest impact in designing for circularity. These tools need to be adaptable and be able to reduce the uncertainty of decisions regarding design and material salvage. This chapter will discuss the scales, stages, and metrics needed to apply these tools. It will evaluate some of the existing technologies and establish guideposts for future tools. By improving flexibility in their inputs, and introducing user-adjustable parameters and immediate feedback within the CAD environment, new design tools will hopefully steer the hand of designers towards increased circularity.

2 Computational Tools Across Scales of the Built Environment

Computational tools play a key role in today’s practice of architecture. Since their introduction, they have greatly influenced a wide range of areas relevant to the design and construction of buildings, ranging from the design and modelling of buildings to the simulation and analysis of their performance (or the performance of their component parts), their visualisation and presentation of their design strategies to the construction, and off-site fabrication of parts or on-site fabrication of full buildings. In the future, the influence of computational tools is bound to increase even further, whereby a building’s whole-life perspective in both evaluation and implementation should be a focus.

Computational tools that assess circularity apply to a wide range of scales. These tools could be employed to assess the circularity of items as small in scale as a product and construction detail or as massive as extended urban areas and material mines (Heisel and Hebel 2021). This section will discuss some of the applications CIs and computation tools for circularity can have at these varying scales.

Computational tools for circularity are extremely relevant at the scales of materials, products, and construction details. Information on the specific pathways of selected materials from production to end of life provides relevant inputs for making design decisions. Similarly, product information in the form of environmental product declarations (EPDs) or other verified datasets can provide important design parameters in the selection and combination of building assemblies. The detail is one of the most critical points of a circular design, as details are where building products with different physical and chemical properties are joined and fastened together. Choices in these products and fasteners directly affect the salvageability or reusability of the component materials at the end-of-use. As a result, assessing the circularity of details is a critical step in designing for circularity. In design workflows, details are typically specified at a stage when architects and engineers have a clearer understanding of the loads, products, and fasteners that are required for a specific assembly. However, specific design intentions for developing details and constructions may well be design parameters that influence material choices and structural systems and thus play an important role in early design phases as well.

At the scale of the building, design tools for circularity can potentially have an extended impact. In addition to accounting for the circularity of materials and connections in total, such tools can also be used to track the circularity of a building over time. Materials and building passports should be updated when materials or components in a building are damaged, removed, repaired, exchanged, or renovated. Whether from manual or sensor input, circularity design tools can recompute a building’s circularity score throughout a building’s operation and store the information in the form of digital twins and MPs.

Surpassing the scale of the individual building, tools which enable design for circularity can have a major influence at the urban scale in closing the gap between demand and supply and in supporting green policy. As cities and regions move to decarbonise their building stocks (Root 2021), these tools can help make important decisions regarding which buildings to retrofit and in what order and how to have the greatest impact.

Exchanging fuel sources and heating equipment are often priorities in green policy. Many buildings also require envelope retrofits to compound the benefits of technical upgrades. This mandates an exchange of material, and with it a consideration of material and product impacts related to embodied carbon and the circularity of these new applications Heisel et al. (2022b).

Equally important at the urban scale is the analysis of the waste streams that result from comprehensive policy changes. The large-scale removal of building elements without a clear next-use scenario will inevitably result in an influx of construction and demolition debris sent to landfills or recycling facilities, while they could constitute a material resource. By generating this information, organisations and companies which have the physical and digital infrastructure for the salvage, reuse, and resale of building materials can be made aware of stock changes in advance. In general, computational tools for a circular economy can play a more significant role in the matching of demand and supply by aggregating available and sought-after materials and products from individual buildings within urban-scale databases.

3 Computational Tools for a Circular Economy

Within the narrow, slow, close, and regenerate framework, computational tools for circularity fall into two categories. These tools ‘narrow’ the amount of material used in buildings by providing a greater understanding of the construction specific impacts of material choices to architects and engineers. While two materials maybe seem identical in mass and volume within a design, CIs offer a wider perspective by accounting, for example, for material loss in production processes or recycling and reuse potentials at end-of-use. In creating CI metrics, computation tools for a circular economy provide the transparency in the hidden waste behind these materials to designers, enabling them to narrow their material use. Computational tools for circularity also serve to ‘close’ material loops. They do so in creating MPs which track the location and quantities of materials within buildings as well as their design for disassembly instructions. These are critical documents in closing material loops, as they facilitate the execution of deconstruction, reuse, and other end-of-use activities.

3.1 Required Flexible Inputs for CI Generation

Within a common design process, the level of detail increases with time. In early design applications, input parameters may be restricted by missing design specificity and thus need to be easily adaptable to keep pace with the iterative design process. To generate a circularity assessment, two inputs are first required: a geometry and a limited set of metadata. The geometry is user generated and can be either drawn/modelled or generated parametrically. The software does not differentiate between a new construction (e.g. massing study) and the assessment of an existing structure (e.g. a survey). Based on this initial geometry of any level of detail, volumetric and surface area information can be calculated.

A second step in generating a circularity assessment requires the pairing of this geometry with relevant metadata, which can be broken down into four subcategories:

• Materials (with associated circularity metrics)

• Constructions/ details (with associated circularity metrics)

• Shearing layers

• (Anticipated) pathways in production and end-of-use stages

While specificity is preferable during later design phases, emphasising generality within these subcategories in early design phases offers designers greater freedom to compare alternatives and make changes to increase circularity. For example, materials can be specified instead of products. Product specifications can change based on independent manufacturing aspects such as supply and demand, the geographic location, or the utilised energy mix. Materials at this stage are more general and can encompass multiple potential products.

Likewise, construction typologies are more general and representative of industry standards and regulatory frameworks than unique assemblies of building products. This allows for the use and assignment of general industry values for production and end-of-use circularity metric calculations, which supports the goal of using CIs as a design parameter in the immediate comparison of alternatives in early design additional to the specification of construction details in later stages.

Numerical user-specified parameters, such as the amount of reused materials in a design or targeted design for disassembly values, are similarly flexible and can be adjusted and refined throughout the design process, thus allowing circularity to be recalculated as the design progresses.

Another recommended input value is the assignment of geometries to shearing layers (Brand 1995). Shearing layers, or ‘layers of change’, provide a filter for materials, products, and components based on their anticipated performance and durability within a building and are defined as site, structure, skin, services, spaceplan, and stuff. Layers such as structure and skin have been observed to have longer use cycles and are more permanent in their arrangement within the building, whereas less determined layers such as spaceplan and stuff are subject to change more rapidly or often. Generating CIs for a breakdown of shearing layers may help understand the implications of specific systems within the overall design, may they be related to material choices, structural requirements, aesthetic preferences, or length of the use cycle.

It is also critical to understand the production and end-of-use pathways for various materials as an input. In computational tools for a circular economy, these material flows are often baked into material databases and not primarily user-inputted. These suggested values provide users with a baseline for different material types and families. If users select a product or material with a unique pathway, they should then be able to update or overwrite relevant metrics. This flexibility is important in the design of these tools, as it guides but does not constrain users in their calculations.

3.2 Assessing Outputs in Early-Stage Design Changes

In contrast to the adaptability of input parameters, output parameters need to be clear and give succinct guidance as to how and in what areas designers can increase circularity. This is especially true in the early design phase, when the design is still relatively conceptual and changes are easier to implement than in later stages.

Consequently, tools for circularity need to communicate a breakdown of CIs for each stage of the design, such as production and end-of-use pathways, and as described above for each shearing layer, assembly and subassembly. This can inform users on where to change parameters or redesign details in the effort to achieve higher circularity values. For instance, an early design tool could identify a subassembly with a particularly low CI as a good opportunity to incorporate design for disassembly strategies or material substitution, thereby raising the CI score of both the subassembly and the entire design.

Similarly, the outputs of such a design tool need to be visualised directly and in ways that are understood intuitively. Given the indeterminacy of these early design stages, CIs need to be displayed rapidly to provide immediate feedback to even slight changes to input parameters. This requires the ability to update relevant metrics based on changes in the formerly discussed parameters. A live feed of effects on the model’s circularity score, for example, would inform the user on whether material choices or design decisions favour regenerative, reusable, or recyclable pathways, reducing the uncertainty while not burdening the designer with an overly prescriptive workflow. In doing this, the act of drawing or modelling would be elevated to an act of constant simulation, making readily apparent impacts associated with the user’s decisions (May and Latour 2019).

These points can be generalised to any computational design tool but take on an increased relevance when designing for circularity. The impact of designing for circularity extends beyond the use of the building and into the entire use and life span of the building’s components. Models need to be updated as changes are made to the building over its use time so that they may inform the continuous feedback loop.

3.3 Circularity Indicators at the End-of-Use Phase

Not only critical as feedback for the early design phase, CIs also take on important relevance at the end-of-use phase. Generating MPs and CIs for existing buildings allows for an assessment of the circularity of the present building stock and can inform salvage and deconstruction efforts for buildings which were not designed with disassembly in mind. A high CI can indicate that a material is easily reusable or salvageable and that using it might help save costs and carbon. This helps direct deconstruction and salvage efforts, as materials that are easily removable and reusable can then be prioritised for recovery, thus maximising the material value and utility of salvage and reuse from existing buildings – especially important in deconstructions with a narrow scope and tight timelines.

3.4 Circularity Databases

Creating a database which enables the calculation of CIs is one of the largest challenges in developing computation tools for circularity. Libraries of EPDs are expanding, but these documents are primarily concerned with carbon and production impacts and not (yet) the end-of-use of materials. Both production and end-of-use material streams vary dramatically based on location due to market factors or local legislation. For example, one municipality which requires the separation and sorting of all construction and demolition waste for recycling might have completely different circularity results than a municipality in which contractors haul all end-of-use waste to landfills because such requirements do not exist.

As a result, datasets on material circularity must be generated or adjusted at least on regional scales. Larger efforts to accomplish this are so far challenging, especially in the United States, where waste streams are often still recorded on handwritten reporting sheets (and are therefore hardly machine-readable), if such information is documented at all. Because of this, most datasets regarding the circularity of materials in the US context must make broad assumptions based on industry standards. Data availability is better in other contexts such as Europe or Asia, but it is still limited. EPEA (in collaboration with Madaster) recently published a dataset of 187 general materials and their associated circularity values, which is now accessible for users with a Madaster account (EPEA and Madaster 2023). Similarly, several other commercial providers of MPs (such as Concular) or life cycle assessment tools (such as One Click LCA) are developing their own internal databases (Campanella 2022).

As computational tools for circularity grow in popularity in the architecture, engineering, and construction industry, there will be a greater demand for certified and regionally adjusted datasets. At the same time, the use of these computational tools creates an opportunity for this data generation. As companies and organisations continue the application of CIs, they will create their own libraries of materials and assemblies and relative circularity metrics. In January 2023, the Circular Construction Lab at Cornell University published a first freely available dataset including associated sources in the hope to launch a collaborative open-source effort on the generation and collection of such circularity datasets (Heisel et al. 2023).

4 Examples of Computational Tools Enabling Design for Circularity

Several tools are presently available for supporting architects, engineers, and other building professionals in making decisions regarding circular design. Most of these tools generate MPs which allow for the documentation and assessment of new constructions. However, significant differences emerge when considering the requirements and considerations described above, specifically the applicability within different design phases, the required resolution of their input parameters, and the intended scale of the built environment.

4.1 Madaster

The Madaster platform originally developed in the Netherlands is described as an ‘online registry for materials and products’ (see Chap. 5 on material passports by Honic et al.). The platform is a browser-based framework for the analysis and storage of buildings and infrastructure projects utilising BIM models (.ifc files) or bill of quantities (Excel) as data input. Volume, area, or quantity information is linked to metadata providing the necessary foundation for calculations of circularity, embodied carbon, and financial material values over time. Madaster has developed into one of Europe’s leading MP systems. The platform provider is involved in the creation of policy frameworks and technical tools across the European Union, has launched the platform in multiple languages and jurisdictions, and is forming partnerships with industry partners and manufacturers to create and collect a robust, up-to-date, and location-specific dataset for its calculations.

Once geometry and metadata are paired, the platform can create highly detailed building passports that allow for the tracking of materials within buildings throughout use cycles, until their eventual end-of-use scenario. Madaster’s ability to track and estimate materials’ (residual) monetary values is a critical feature for stakeholders interested in the deconstruction of a building at its end-of-use, as the resale value of the building’s materials and components – and their ability to offset contractor costs – is a key factor in determining the profitability of a deconstruction project.

Madaster’s primary use case is the documentation of buildings once their design has been finalised and a detailed BIM model can be saved and exported. Madaster is a highly evolved documentation and assessment tool and has recently set eyes on the development of design tool functionalities as described within this chapter.

4.2 One Click LCA

One Click LCA is a building life cycle assessment tool, allowing users to analyse the environmental impact of their buildings based on metrics such as embodied and operational carbon, to meet sustainability certifications such as LEED (Leadership in Energy and Environmental Design), BREEAM (Building Research Establishment Environmental Assessment Method), DGNB (Deutsche Gesellschaft für Nachhaltiges Bauen e.V.), or the Living Building Challenge. Its primary purpose is the assessment of carbon in buildings, but its functionalities in generating MPs are quickly expanding.

Like Madaster, One Click LCA operates as a stand-alone online platform, with its primary input being a detailed browser-based bill of quantities. It is supported by a well-organised and robust internal dataset, which uses an increasing library of EPDs as its primary resource, focusing primarily on the United States and Europe. One Click LCA also offers software plugins for CAD and BIM environments, such as Rhinoceros3D or Revit, allowing users to assess embodied carbon and circularity metrics as they design their buildings. The plugin is specifically advertised as a means to integrate the software’s capabilities into early design phases and is accessible to owners of a commercial One Click LCA licence (One Click LCA 2021a; One Click LCA 2021b). These plugins are understood as extensions to the online platform to ease the linking of geometry with material and metadata. As such, the plugins display a partial amount of assessment results in the CAD environment, while the online platform computes high-resolution MPs even when starting at a low resolution of design parameters. These differences in resolution and output formats are meant to address the changing requirements and scales of the design and construction process (One Click LCA 2023).

4.3 RhinoCircular

RhinoCircular is an application for Rhinoceros3D and Grasshopper developed within the Circular Construction Lab at Cornell University to specifically evaluate material circularity in the early design phases, and the goals of the application match the framework outlined in Sect. 6.3 (Circular Construction Lab 2023). RhinoCircular’s key focus is presently the assessment and visualisation of a design’s environmental impact with respect to circularity: the degree to which design solutions minimise extraction and waste in favour of reusable, recyclable, and renewable material resources.

RhinoCircular allows direct and immediate feedback on design decisions regarding formal deliberations, structural considerations, material selection, and detailing based on MP and CI assessments. It can be integrated in existing and complex workflows and is compatible with industry-standard databases while providing its own starter dataset.

Figure 6.1 shows some of the potential of this tool to provide rapid and targeted feedback within the Rhinoceros3D environment. In this specific example, a detail model is assessed for circularity. CIs are generated for each element within the detail. Once these metrics are generated, they are remapped to the model geometry, demonstrating to users which elements are highly circular and which are less so.

Fig. 6.1

Evaluating a detail with RhinoCircular

Built as a native Grasshopper application, the tool consists of several components that can be combined or connected to suit the specific needs of a proposed project in any design phase or on any level of detail. Designed to be compatible with other applications in the Grasshopper ecosystem, RhinoCircular’s circularity evaluations can be combined with structural simulation tools like Kangaroo3D or environmental systems simulation tools such as ClimateStudio. While the learning curve for the tool is relatively gentle, it assumes users have basic skills and an understanding of visual scripting in the Grasshopper ecosystem (Fig. 6.2).

Fig. 6.2

Evaluating a structure with RhinoCircular

Relative to the prior discussed tools, RhinoCircular generates lower-resolution outputs that represent a close approximation of a building’s CI. This is because the goal of the tool is not a comprehensive MP or LCA but instead to inform designers quickly and immediately in an early-stage design where data resolution and product specification is equally lower. To support this mission, computation results are displayed directly in the modelling space and can be mapped onto the geometry, offering visual and targeted feedback to designers in the effort of informing the decision-making process.

5 Discussion

Assessing the relative strengths of the above MP and CI tools provides insight into where architects and other professionals can most effectively use them. When assessing the utility of each tool in various scenarios, those geared towards the earlier design phase better help to ‘narrow’ our present material consumption. Those which require more detail while also narrowing material consumption and emissions are more effective than earlier stage tools at closing the loop and enabling material salvage.

The software tool RhinoCircular is most relevant within early design phases when the design has yet to be finalised and rapid feedback is needed. After those phases, tools such as Madaster and One Click LCA generate detailed information on circularity and embodied carbon that can be delivered to stakeholders and compliance agencies. MPs that are produced by One Click LCA and Madaster are also relevant through the building’s use and end-of-use, allowing for the recovery and reuse of building elements in the future. Madaster’s platform is particularly useful when pricing materials for resale through its residual value metrics based on global material markets, which gives the user a sense of the resale value that can be realised when a building is deconstructed. Ideally, all these tools can be combined in a workflow that can leverage the benefits of CIs as a metric across scales.

Independent of the tool, circularity and CI evaluation must become a key design parameter across scales. The earlier architects, engineers, and designers can advocate and implement circularity into buildings, the greater the future impact, both in material sourcing and with respect to their end-of-use pathways. CI tools encourage circular behaviour, but it is the role of practitioners to apply and implement the feedback and optimise buildings for circularity.

As a result, greater collaboration is needed between practitioners in both sharing datasets which enable circularity assessments and their compatibility and in strengthening the accuracy and reach of computational tools for circularity. The opaqueness of supply chains and manufacturing processes inhibits professionals from having the data needed to confidently assess circularity in the built environment. These tools are only as accurate as the data which feeds them, and therefore data on materials needs to be collected and shared across all four phases (slow, close, narrow, regenerate) in order to give practitioners greater confidence in computational tools for a circular economy and their outputs.

6 Key Takeaways

• Tools which calculate circularity indicators have the greatest impact in the early design phase.

• Tools which calculate circularity indicators are most accurate in the construction documentation phase.

• Material passports have the greatest relevance during a building’s use and end-of-use.

• More robust databases and data transparency are needed.