A Future-Proof Built Environment through Regenerative and Circular Lenses—Delphi Approach for Criteria Selection

Abstract

Despite the increasing use of neighbourhood sustainability assessment tools (NSAT), their linear approach may be insufficient to tackle the global and local social and ecological challenges. The circular economy (CE) has recently emerged as a new pathway, adopted by corporations and public organisations. Understanding how to apply CE to existing communities, while addressing some of its shortcomings, particularly the strong focus on resource management, is the main goal of this paper. Building upon a Regenerative Circularity for the Built Environment (RC4BE) conceptual model that merges circular economy and regenerative design concepts, a framework with criteria for its implementation in the transition of existing urban areas is proposed. A preliminary framework structure with criteria mapped from literature is proposed and validated through a 2-round Delphi consultation with 31 international experts. The final framework, with 136 criteria, addresses some of the identified gaps and different urban cycles related to physical resources, ecosystems, liveability, infrastructure, governance, participation, local economy, and other socioeconomic aspects of urban communities. This expanded take on CE should be useful for built environment professionals and other urban stakeholders interested in regenerating their communities and precincts by going beyond current green approaches and existing tools to effectively generate positive impact for people and the planet.

1. Introduction

A call for action on environmental, climate, and socioeconomic issues has been alive for many decades, but with slow and inadequate progress. The latest IPCC report [1] highlights that in most global regions, there has been an increase in greenhouse gases (GHG) emissions between 2000–2015. Urban areas account for 67–72% of global carbon emissions, 18% alone from the 100 highest emitting cities. This shows the inequality in emissions, as cities in high-income countries have a per capita emission rate 7 times higher than those in the lowest emitting areas. To avoid the increased emissions from new and updated buildings and infrastructure, the IPCC also adds that a deep decarbonisation approach combining reduced energy consumption, compact and efficient urban form, net zero energy sources, and increased carbon uptake and storage is needed. However, climate is not the only planetary system beyond the safe boundaries, as biodiversity extinction [2], chemical pollution [3], and nutrient flows have already exceeded the safe boundaries that could lead into cascading ecological failures [4]. Similarly, the social foundations of many locations, as access to energy, water, and housing, levels of income and jobs, among others, are below the thresholds that would ensure thriving lives and a just space for all of humanity [5].

Following a global trend, the commitment to green practices in the built environment (BE) sector has been an increasing, but slow, reality for over two decades. Many of these initiatives have been upscaled by the Green Building movement, exemplified by certification schemes or rating systems such as the British BREEAM (Building Research Establishment Environmental Assessment Method), the American LEED (Leadership in Energy and Environmental Design), the French-Brazilian AQUA-HQE (High Environmental Quality), and the Australian Green Star, among others. Over time, these building schemes evolved from a building focus to include neighbourhood sustainability assessment tools (NSAT). Several studies [6,7,8,9] have identified several flaws and limitations, including:

• Non-transparent top-down approaches with limited or non-existent public participation,
• Reduced and unbalanced consideration of sustainability pillars, with no measures to enforce basic sustainability aspects and address intergenerational aspects,
• Complexity, rigidity, and excessive prescriptiveness,
• Limited flexibility to respond to boundary and local context issues
• Reduced interlinkages between indicators,
• Limited alignment between methodologies of different tools,
• Poor post-occupancy monitoring of performance levels.

And despite the lack of a systemic approach in NSAT, LEED for Neighbourhood Development shows some alignment with low carbon policies in Sao Paulo, and the GHG Protocol for communities, which still could be improved [10]. Despite that, most of these tools adopt a green approach, i.e., a focus on efficiency improvement that allows for impacts [11]. New approaches have been emerging, and different sectors have started to embed the circular economy (CE) concept into their practices. Among many perspectives, CE has been defined as an “economic system that uses a systemic approach to maintain a circular flow of resources, by recovering, retaining or adding to their value, while contributing to sustainable development” [12] (p. 1). This reflects how most of the current CE initiatives are driven by a focus on resource management.

Considering that from current global GHG emissions, 55% relate to energy, and 45% to product making, the Ellen MacArthur Foundation [13] contends that a CE approach can reduce 45% of the latter category, particularly by working with key industries as cement, steel, plastic, aluminium, and food systems. Another study in five European countries indicated that a mixed approach of energy efficiency, renewable energies, and an efficient manufacturing has potential to reduce two thirds or more of the carbon emissions, deliver huge cuts in unemployment rates, and improve the trade balance by at least 1.5% [14]. The IPCC, on the other hand, states that to date there are still limited contributions from CE to climate mitigation, probably due to a limited CE implementation. However, there is future potential, and CE is acknowledged for its contribution to manage waste, resource cycles, transport, and reduce primary production. Attention to potential rebound effect (increase of demand) of freight and ridesharing vehicle-km travelled is needed though. Some of its social benefits include the improvement of wellbeing, the creation of new jobs related to circular business models, and the empowerment of social actors for climate mitigation initiatives and the needed collaboration between stakeholders. [15]. A case-by-case analysis is needed to calculate the actual contribution of each solution to mitigation [16]. Also, ensuring these benefits require targeted policies, regulations, tax systems, and technological development [14,15].

Other issues pointed out by several authors about CE regard an inadequate consideration of biodiversity conservation [17], its strong technical approach that ignores needed changes in the society [18], a potential shortage of secondary resources as replacement for primary resources [19], and an incomplete social focus mainly on employment, health and safety, and participation [20].

1.1. Circular Economy Frameworks for the Built Environment

Several governments, despite some limitations, have started to consider CE in their roadmaps at the country [21] and city level [22,23,24], and for specific sectors such as construction waste [25]. The concept is also emerging as a reference for new or retrofitted neighbourhoods. Buiksloterham, in Amsterdam, Netherlands [26], Yarrabilba in Queensland, Australia [27], and Kolding Marina City, in Denmark [28] are a few examples. Each of these projects, however, have defined their own guidelines and strategies without the support of an existing external framework. There are still few circularity frameworks available for the built environment. While existing CE frameworks present many merits, sometimes there are limitations due to context or other constraints. A quick analysis (

Table 1. Assessment of circularity frameworks for the built environment.

Framework 1	Ref.	External Validation 2	Project Specific	Typology Specific	Beyond Resource Focus	Metrics
Circular Cities Assessment Framework	[29]					X
Circular assessment framework for Spaarndammer tunnel, Amsterdam	[30]		X	X	X
London circularity indicators	[31]	X				X
Circular City Actions Framework	[32]				X
Circle City Scan Tool	[33]
Circular Economy in Cities Project Guide	[34]				X

¹ This is not an exhaustive or systematic selection of existing frameworks. ² It was not possible to identify if the other frameworks had any consultation or validation process with external stakeholders.

) reveals that some of these existing CE frameworks proposed for urban areas fall into at least one of the following characteristics:

• No consultation with experts or stakeholders.
• Specific for one project or typology.
• Focus exclusively or mainly on resources and flows management, with an increasing but yet reduced consideration of other sustainability-related scopes.
• No specific metrics included.

1.2. A Framework for a Regenerative Circularity for the Built Environment (RC4BE)

Moving the BE sector away from current green approaches requires new frameworks enabling regenerative and circular thinking and practice [35]. However, adequately designing a framework for a systemic transition of urban areas may be a complex process. It is not about having one single framework for all situations, particularly when ‘context’ should be a core concept, but rather about exploring different perspectives and pathways to support the needed change in the predominant mindset. Bell and Morse very adequately point out that “sustainability changes as an idea (or as a system) in terms of the perception of the onlookers, and they will also change with time [thus] the view of sustainability must be developed so that it takes onboard the legitimacy of different views of sustainability” [36] (p. 127).

Previous research established the underpinning ideas of a ‘Regenerative Circularity for the Built Environment’ (RC4BE) conceptual model described below (source blinded). It intends to address some of the shortcomings of the CE concept by merging it with the regenerative design (RD) approach, with which it shares its origins [37]. This becomes clear in Lyle’s [38] (p. 10) definition in which regenerative design “replaces the present linear system of throughput flows with cyclical flows at sources, consumption centers, and sinks”. RD advocates for positive impact, in contrast to less-harmful or zero impact solutions, and acknowledges humans as an integral part of nature, rather than alienated from it [39]. It also addresses the regeneration of community relationships and engagement [40].

The RC4BE conceptual model (Figure 1) adopts a consideration for the different urban flows, and is defined as one that seeks:

• “a circular metabolism of urban flows and stocks based on the fair share and regeneration of resources.
• adaptive-resilient and high-quality urban systems that adapt to and recover from future conditions, and provide liveable, accessible, and safe urban spaces to reconnect citizens and promote sustainable lifestyles.
• healthy and bioconnected urban ecosystems that promote the reconnection between humans and nature, the stewardship and regeneration of urban and non-urban ecological systems, and the support of nature-inspired solutions, thus enabling the provision of ecosystem services sustainably into the future.
• good governance and thriving communities that stimulate a just management of natural, social, and economic capital, with inclusive, equitable, and collaborative community participation, access to knowledge, and regenerative local economies and livelihoods.
• all based on systemic approach and positive impact that, through integrated planning, embrace life cycle thinking, respect the local and global social and ecological boundaries, and create value for all” [41] (p. 11).

This paper aims to understand in what ways urban stakeholders can implement a regenerative and circular transition of urban areas. It seeks to address some of the gaps identified in existing in NSAT and CE frameworks for the urban scale, as discussed above. Another aspect to be addressed regards the adoption of a cross-pollinated perspective to CE with regenerative design and positive impact concepts [41]. It is the fourth stage (Figure 2) in the development of a framework based on the RC4BE conceptual model—each stage building upon the previous one. Stage 04 seeks to define the core ideas and pathways that need to be addressed by examining the literature and validating them with a panel of experts. This overarching structure, made of a set of themes, categories, and criteria, is a steppingstone to later support the definition of an action framework that takes into consideration the possible synergies among different aspects, the most adequate metrics, the different levels of the transition process, and the methodology for its implementation.

Here, the systemic processes and synergies among actions and solutions are not addressed as they will be defined in a future stage.

We expect to expand the current notions of a circular economy for cities by merging it with regenerative design concepts. From focus mainly on the cycles of resources to one that considers the different flows of a city, such as those related to physical resources, ecosystems, liveability, infrastructure, livelihoods, and other socioeconomic aspects of communities. It can be a useful tool for urban stakeholders of all kinds (residents, policymakers, developers, practitioners, community managers, among others) to effectively generate positive impact for people and the planet.

2. Materials and Methods

This study was conducted in 2 main phases (Figure 3): (a) the mapping, selection, and definition of relevant criteria from existing frameworks, and (b) their refinement and validation with field experts using a 2-round consultation process.

2.1. Consultation Process (Delphi Technique)

Defining the most adequate methodology for the design of a systemic framework suitable for a regenerative circularity approach is a complex process. This study focuses on defining the set of relevant criteria within the proposed structure. Implementation methods, possible synergies under a systemic approach, and if and how weights and scores should be attributed to the items, are tasks to be conducted in phase 05 of this research.

Initially, the Analytical Hierarchy Process (AHP) was considered. It is more suitable for the pairwise comparison of smaller sets of items [42]. In this research, the large number of items for evaluation (132) would have rendered it long and impractical for the respondents, which could reduce even more the rate of responses. Additionally, this stage did not intend to define a scoring and weighting system, for which AHP is usually used [42]. Therefore, the chosen methodology was the Delphi technique, which is “a process for gaining consensus through controlled feedback from a panel—a group made up of experts or individuals knowledgeable on the subject” [43] (p. 12). The method is widely used in the development of frameworks used for the selection of indicators by trusting on the knowledge of experts in the field [44,45,46] that are not directly confronted with each other [47].

There are at least ten possible design types of the Delphi methodology, and no matter which is employed, one should have in mind that Delphi results are not an absolute truth, but rather, a snapshot of the selected panel’s perception based on their knowledge, experience, context, in that specific moment in time [48]. This study applies a combination of modified Delphi and eDelphi (Delphi conducted online [49]).

In traditional Delphi, round 1 works with open-ended questions to promote a brainstorming by experts that will identify the relevant aspects to the specific item under debate, and then generate a list of items to be judged in the following rounds [47]. For sustainability assessment this step is usually used to create a preliminary list of categories and criteria [44]. While this approach reduces the biases in the process of selecting the variables to be judged [43], it can be even more time-consuming and reduce the rate of participation in the process. Replacing this process with a literature review is an acceptable choice that has been widely used [43,50,51] and is the pathway adopted here. For the purpose of describing the methodology here, we consider round 1 and round 2 as the consultation steps with the experts in Phase 2 (see Figure 3 and Section 2.3). The preliminary step substituted by the literature review is part of Phase 1, described in the following subsection).

2.2. Mapping and Pre-Selection of Criteria

Documents were identified through a non-systematic literature review. Searches in Scopus and Google Scholar used different combinations of the keywords “circular economy”, “circularity”, “regenerative design”, “regenerative development”, “sustainability”, “sustainability assessment tool”, “rating system”, “framework”, “built environment”, “urban planning”, “urban development”, “neighbourhood”, “district”, and “infrastructure”. Snowballing was used to identify other relevant documents. Grey literature documents were identified in platforms, databases, and NGOs linked to circular economy, regenerative design, and green building practices. Documents were then scoped based on their relevance for the study.

This study selected 45 frameworks (See Appendix A,

Table A1. List of frameworks and tools used for criteria mapping.

Ref.	Tool or Framework
[160]	DGNB-UD Urban Districts
[163,164]	LFC LEED for Cities and Communities 4.1|existing|plan and design
[165]	LEED-ND Neighborhood Development
[166]	CASBEE-UD Urban Development 2014
[167]	CASBEE-City 2012
[168]	Green Star Communities v1.1
[169]	EcoDistricts Protocol v1.3
[170]	Well Communities v2
[171]	Fitwel Community 2.1
[172]	OPC—One Planet Communities
[173]	LCC—Living Community Challenge 1.2
[174]	LENSES Framework
[175]	SITES v2
[176]	AQUA-HQE for Urban Planning
[177]	Envision ISI v3
[178]	IS Infrastructure Sustainability Australia v2.0
[179]	CEEQUAL v6 International
[180]	Regenerate App
[181]	Doughnut Economics: Thriving City Portrait (TCP)
[182]	EU Monitoring framework for CE
[183]	Circular Economy Action Plan—Construction and buildings
[184]	CIRCuIT—Circular Construction in Regenerative Cities
[185]	Measuring Scotland’s progress towards a circular economy to help combat the climate emergency v1
[186]	Circular City Netherlands
[187]	C2C for BE principles
[30]	Metabolic CE indicators for infrastructure (NL)
[188]	Framework for Circular Building (for BREEAM NL)
[189]	Metabolic—measurement framework to track circular progress—Metro Amsterdam
[190]	Level(s)
[24]	Roadmap for Helsinki’s circular and sharing economy
[29]	Port Cities—criteria and indicators of circularity in the built environment sector
[31]	London circularity indicators (Measuring London’s progress towards becoming a more circular city)
[191]	Urban Agenda Indicators for circular economy (CE) transition in cities v4
[192]	Amsterdam Circular Monitor
[193]	GCE Green and circular economy Barcelona
[194]	ROCK Project: Circular city—A methodological approach for sustainable districts and communities
[195]	CCAF Circular City Analysis Framework Porto
[196]	Opportunity-spaces for self-regenerative processes
[197]	Creating circular neighbourhoods
[198]	Circular City Index Italy
[32]	Circular City Actions Framework
[33]	Circle City Scan Tool
[34]	EMF/ARUP Circular Economy in Cities Project Guide
[199]	Urban Metabolism as Framework for Circular Economy Design for Cities

) for buildings, precincts, infrastructure, and cities, from which 18 were sustainability assessment tools commonly used in the BE sector—five of those had a circular or regenerative approach. The other 27 frameworks included seven frameworks by NGOs or consultancies, nine from academia, and 12 institutional documents by cities, countries, or supranational organisations. We also considered the inputs obtained from sector experts in [52].

The pre-selection of criteria took several iterations (Figure 3). The indicators, criteria or strategies of these documents were mapped, analysed, and grouped under identified topics. These topics were then linked and harmonised with the RC4BE five pillars (source blinded). The items were then merged, deleted if not relevant, and rewritten for the context of urban transitions under a regenerative circularity approach.

For this stage, we adopt criterion over indicator, considering that a criterion is a “principle or standard that a thing is judged by, [adding] meaning and operationality to a principle without itself being a direct measure of performance” [53] (p. 87). And indicators “provide the operational measures (quantitative or qualitative) for each proposed criterion. “Indicators are (…) measures that convey ‘value added messages’ in a simplified and useful manner to stakeholders” [54] (p. 455).

2.3. Experts’ Panel, Questionnaire, and Analysis Conditions

Delphi panels do not usually require a large number of participants to achieve a statistical representation [55]. Instead, there should be a careful curation of experts in the topic to be consulted, trying to achieve heterogeneity in the selection [43]. The literature indicates Delphi panels usually have less than 50 subjects, with mostly varying between 15 and 20 people [56].

Here, experts were selected and invited by email through purposive (or judgemental) sampling [57]. Professionals, selected from academia, industry, and policy/certification sectors, should have at least five years of experience (in any sector), and a good level of knowledge and experience with sustainability, circular economy, and/or regenerative design in the built environment. Some had previously participated in the survey and/or interview of stage 3 of this research.

To compensate for possible lower response rates, the questionnaire for round 1 was prepared and sent with an individualised link by email through the Qualtrics platform to 122 participants and made available for 18 days, with a reminder sent. The first round had 31 (25%) valid responses. Before the questions, it provided a brief illustrated introduction to the RC4BE conceptual model. Structure and analysis conditions are summarised in

Table 2. Structure of questionnaire and consensus conditions.

Item	Round 1	Round 2
Demographics	8 questions	N/A
Reliability check	Cronbach’s α ≥ 0.70
Structure and methodology questions	5-point Likert-scale questions (1 strongly disagree, 2 somewhat disagree, 3 neither agree nor disagree, 4 somewhat agree, 5 strongly agree) and open-ended questions for comments.
Structure and methodology consensus conditions	Median equal or greater than 4. 60% or more of experts selected the two upper bands. Standard deviation equal or smaller than 1.5.
Criteria questions	6-point Likert-scale questions (1 remove, 2 not at all important, 3 slightly important, 4 moderately important, 5 very important, 6 extremely important) and open-ended questions for comment, divided into the 5 themes.	6-point Likert-scale questions (1 not at all important, 2 slightly important, 3 moderately important, 4 very important, 5 extremely important) and open-ended questions for comment, divided into the 5 themes.
Criteria quantitative consensus conditions	Median equal or greater than 5. Score ≥ 60% (sum of two upper bands %). Standard deviation ≤ 1.5. No more than one ‘remove’ response.	Median equal or greater than 4. Score ≥ 60% (sum of two upper bands %). Standard deviation ≤ 1.5.
Criteria qualitative consensus conditions	No relevant comment or suggestion for alteration
Ranking of criteria within categories	N/A	From highest to lowest score within each category
Ranking of themes and categories	N/A	From highest to lowest score calculated from criteria’s score average.

.

At the end of round 1, a report with the results and comments from experts, and feedback on the relevant issues raised was prepared and sent to participants alongside the invitation for round 2. Only the participants of round 1 were invited for round 2 [43], initially open for 16 days and extended to 20 days after sending reminders for completion, at the end of which it had 16 (52%) contributions. Time restrictions and the number of items in round 1 may explain the reduced participation in round 2, which is an issue found in other Delphi studies; yet, the number still falls within the typical ranges found in other studies [56]. It is not about numbers, but rather about the quality of the selected experts for the panel [55]. Sharing the feedback is an intrinsic feature of Delphi compared to other techniques, as it allows participants to reflect and revise their responses based on the collective results [58].

The first analysis consisted of checking the reliability of the set of responses by calculating Cronbach’s alpha, which should fall between 0 and 1; the closer to 1, the more reliable. Values above 0.80 are adequate for group research [59], but even values above 0.70 demonstrate a strong association between ratings [46].

There is no unanimity on how to define consensus for Delphi studies [43]. Here, a combined quantitative and qualitative approach [56] to define consensus was adopted. Quantitative consensus was based on three or four conditions depending on the type of question or round (see Table 2). The verification of dispersion measures defined the standard deviation should not be greater than 1.5 [60]. For central tendency measures, median and a major preference for the two upper bands of agreement or importance (which we will call ‘score’) were the following two conditions [61]. Median was chosen over mean as a more adequate criterion for Likert-type scales [56], whereas for the score, a threshold of 60% or more, rather than 80%, aimed at the inclusion of an optimum quantity of criteria [61]. A fourth condition was used only for criteria in round 1. The first level (1) of the six-point Likert scale referred to the option ‘remove’. The condition of no more than one ‘remove’ response was defined to account for the biases of outlier respondents. While the average number of ‘remove’ responses per participant was 1.32, two single experts were responsible for 24 out of 32 ‘remove’ responses. This affects the definition of the median threshold in each round, as the ‘remove’ option is excluded in round 2.

After this quantitative filter in R1, a qualitative consensus condition was implemented. Items with a relevant comment or suggestion from participants that could result in alterations led the item to be revised and included in R2. At the end of R2, a three-tier ranking system was applied using the score [62]:

• Ranking of criteria within each category.
• General ranking of categories.
• General ranking of themes.

For ‘1′, in case of ties, the following conditions were applied: standard deviation, % of responses in the Likert-scale from the highest to the lowest level, and median. For ‘2′ and ‘3′ there were no ties. These rankings are not intended to define the level of priority of criteria in the final framework, but the level of agreement regarding the importance of each criterion, category, and theme by participants of this study.

3. Results

3.1. Pre-Selection of Criteria

The mapping of the 45 frameworks resulted in a total of 1377 criteria, 730 from NSAT, and 640 from other frameworks. These criteria were then grouped into 27 categories, which were then arranged as a preliminary group of five themes, derived from the five pillars of the RC4BE conceptual model. For this analysis, some criteria have been classified in more than one category or theme.

Analysing categories (Figure 4), NSAT had a total of health and wellbeing criteria (146) far above the other categories. Resources (87), urban governance (82), cohesion and affordability (65), biosphere (57), and economy/business (52) were also in the top list. Systemic and lifecycle (4), landscape management (7), BE renovation (11), and digital, smart, information and communication technology (13) were in the bottom of the list.

As expected, most criteria from other frameworks, which have a circularity or regenerative focus, fall under the resources category (234 criteria). Other relevant categories include economy/business (89), urban governance and energy (47 each), and health and wellbeing (46). Aspects of landscape management (1), hydrological cycle (3), urban fabric (4), heritage and culture (5), green provision and design (6), and building design (7), had the smallest number of criteria.

Examining themes (Figure 5), NSAT have a more balanced distribution of their criteria among urban systems (324 criteria, 35.29%), governance and communities (275, 29.96%), circular metabolism (194, 21.13%), and bioconnections (121, 13.18%). Positive & systemic had the smallest number of criteria (4, 0.44%).

The majority of criteria from other frameworks fall into the circular metabolism theme (342 criteria, 47.7%), also as expected. This is followed by governance and communities (208, 29.01%), urban systems (112, 15.62%), and bioconnections (45, 6.28%). Surprisingly, the theme with the smallest number of criteria was positive & systemic (10, 1.39%).

3.2. Framework Proposed for Consultation

After the initial examination to identify topics and related criteria, and their distribution in categories and themes, the items were then analysed, merged, deleted if not relevant, and rewritten for the context of urban transitions under a regenerative circularity approach. This led to the preliminary framework proposed for the consultation process, which is organised in five levels (Figure 6). L1 refers to the pillars of the conceptual model (Section 1.2), which were adapted and derived into the themes from L2, and categories from L3. The scope of this study, the definition and validation of the 131 pre-selected criteria predefined, make L4. The final level, L5, will define the most appropriate indicators/metrics and performance levels in this research’s stage 5. The pillars were reorganised and renamed to better suit the structure of the framework. The ‘systemic approach and positive impact’ pillar, due to its transversality, is embedded into the five.

In the consultation process, round 1 presented 138 items for the evaluation of experts: six regarding structure and methodology, and 132 criteria. 19 items did not reach quantitative consensus, three structure/methodology and 16 criteria (Figure 7a). After applying the quantitative and qualitative conditions, a total of 45 items were reassessed on round 2, five structure and methodology, and 40 criteria (Figure 7b). In the end, a total of 143 items reached consensus, seven structure/methodology, and 136 criteria (Figure 7c), therefore increasing the initial number presented to participants. The results of the reliability analysis using Cronbach’s α were 0.9756 for round 1, and 0.9072 for round 2.

3.3. Demographics

The demographic profile (Figure 8) of the experts’ panel for round 1 (R1) and round 2 (R2) shows that most professionals are female (9a) and based in Australia and Brazil, with Chile being well represented as well (9b). The main age groups (9c) are 36–45 and 26–35, with no professionals below 26 or above 65. Most selected professionals have more than 10 years of engagement with the built environment sector (9d). The most represented BE subsectors (9e) were sustainability/environmental consultant, designer (architect, urban/landscape designer), and academia. About three quarters of participants have engaged with built environment sustainability tools, most just applying them in projects, and some being part of their development process (9f).

3.4. Criteria

Participants ranked the level of importance of the predetermined criteria, seeking to narrow the list to a set of relevant items.

Table 3. Score of themes and categories derived from criteria ranking average, and their respective definitions.

Themes	Theme’s Score	General Ranking	Categories	Cat. Score	General Ranking	Aiming to Redesign and Transform Precincts and Neighbourhood…
Flows and stocks	84.28%	1	Greenhouse gases	86.25%	7	into climate positive urban systems.
			Energy	81.05%	14	into positive energy urban systems.
			Water	86.73%	6	into positive water urban systems.
			Food system	77.88%	17	for facilitated access to regenerative food systems.
			Built environment stocks and flows	89.03%	3	for a circular and resource effective built environment stock.
			Resource sourcing	84.79%	10	for a regenerative and responsible sourcing of built environment resources.
			Material loops in operation	84.73%	11	with adequate infrastructure for circular and regenerative management of resources loops.
Urban systems	84.05%	2	Adaptive resilience	91.13%	2	to be future-proof through the adaptive resilience of urban systems and buildings.
			Access and diversity	87.50%	4	for a diverse, just, and universal access to services and infrastructure.
			Mobility infrastructure	82.93%	13	to provide widespread and regenerative mobility systems.
			Smart and digital systems	73.71%	23	for the uptake of technology and smart solutions towards the betterment of life for all citizens.
			Urban fabric	86.02%	8	to foster compact urban fabrics with optimal levels of quality density.
Outdoor Environmental Quality	78.59%	4	Thermal comfort	75.81%	20	to improve the local microclimate and provide conditions for the use of outdoor spaces under different weather conditions.
			Air quality	85.81%	9	to regenerate the local air quality.
			Soundscape	74.53%	22	for adequate noise levels and a pleasant soundscape.
			Visual comfort	64.52%	26	to reduce visual discomfort and light pollution, while maintaining the local safety and security.
			Beauty and sense of place	79.60%	15	to strengthen the sense of place by fostering beauty in its diversity, and honouring local culture and heritage.
			Safety and security	87.16%	5	to offer an increased sense of security through their physical characteristics.
			Proxemics and public space use	78.23%	16	to increase the use of public spaces through diverse opportunities for citizens to reconnect, interact, and isolate.
Bioconnectivity	81.31%	3	Biosphere conservation and regeneration	93.08%	1	to conserve, regenerate, and maintain ecosystems and biodiversity inside and outside urban boundaries.
			Impacts to biosphere and human health	75.08%	21	to reverse the impacts and achieve a positive impact into human and ecosystems health.
			Ecosystem services provision	76.34%	19	to enable and maximise the provision of ecosystem services and nature-based solutions.
			Green infrastructure maintenance	73.12%	24	to maximise the provision and optimise the operation of green infrastructure interventions in urban areas.
Local community and economy	77.48%	5	Governance and participation	83.06%	12	to achieve inclusive and just community governance and activate the required urban transition enablers.
			Education and awareness	76.61%	18	to spread awareness and to function as sources of knowledge.
			Local economies and businesses ecosystem	72.89%	25	to enhance local economies and prioritise business models that generate positive impact.

presents the categories’ aims, and the respective scores (determined by the sum of the two upper levels of importance) and general rankings for each theme and category. Each of these were calculated using the averages from the criteria score.

The heatmap in Figure 9 illustrates how criteria scores are distributed in a scale from 61.29% (the lowest score) to 100%. It allows the visualisation of which categories concentrate the lowest and highest values, and therefore deemed as less or more important. In the following subsections, the most relevant aspects for each theme are discussed. As the number of criteria is extensive, the complete list of criteria per category and theme is in Supplementary Materials File S1.

3.5. Results Per Theme

The ‘flows and stocks’ theme refers to the provision of a “regenerative, circular, healthy, and fair management of resources”, with a total of 42 criteria. The categories with the highest scores (Table 3) in the general top 10 ranking were built environment stocks and flows (3; 89.03%), water (6; 86.73%), climate change mitigation (7; 86.25%), and resource sourcing (10; 84.79%).

The importance of the above categories is reflected in the criteria scores, with three items reaching 100%: protect and regenerate natural water systems (category water), improve lifespan, flexibility, adaptability, repairability, and shared use of buildings and public spaces (category: built environment stocks and flows), and prioritise reusable resources (category: resource sourcing). In this category, three criteria did not make the final list:

• Reclaim underused spaces for free or low-cost healthy food production.
• Reclaim underused spaces for food production.
• Prioritise abundant resources.

‘Urban systems’ refers to the “access to high quality, diverse, and adaptive-resilient urban systems” with 29 final criteria in total. The categories with the highest scores (Table 3) in the general top 10 ranking were adaptive resilience (2, 91.13%), access and diversity (4, 87.50%), and urban fabric (8, 86.02%).

The criteria with the top three scores were increase affordability, socioeconomic plurality and diversity of housing (96.77%, in access and diversity), increase the adaptive resilience of critical infrastructure (96.78%, in adaptive resilience), and charging stations for electric vehicles and integrate electric mobility as a support system to support an energy positive neighbourhood (both in mobility infrastructure, tied with 93.75%).

Here, three criteria did not make the final list:

• Parking restrictions in public spaces.
• Provide public parking spaces dedicated to Personal Electric Vehicles.
• Free Wi-Fi in public areas.

The ‘outdoor environmental quality’ category refers to the “provision of liveable and inclusive outdoor public spaces that foster social reconnection and well-being”, with a total of 27 final criteria. The categories with the highest scores (Table 3) in the general ranking top 10 were safety and security (5, 87.16%), and air quality (9, 85.81%).

The criteria with the top three scores were eliminate local sources of air contamination (96.78%, in air quality), adopt an urban ‘safety and security’ approach based on women, children, and older adults (93.75%, in safety and security), and three others with tied scores (93.55%), variety of restorative open spaces for physical and mental health (in public use space, proxemics), and protect air quality during construction activities and landscape maintenance, and actively improve air quality/clean the air both in air quality).

Here, two criteria did not make the final list:

• Monitor/exhibit real-time local weather conditions to citizens.
• Monitor/exhibit real-time sound levels to citizens.

‘Bioconnectivity’ refers to the “adequate stewardship, maintenance and regeneration of urban biodiversity that enable the provision of ecosystem services sustainably into the future”, with 17 final criteria in total. The category with the highest score (Table 3) in the general ranking top 10 was biosphere conservation and regeneration (1, 93.08%).

The criteria with the top three scores were all in the same category (biosphere conservation and regeneration): restore/renature water bodies (100%), second-place tie between increase and regenerate green areas and corridors within and around the urban fabric, and prioritise infill, greyfield, and brownfield retrofit and remediation (93.75%), and a third-place tie with provide ecological buffers around sensitive areas, and regenerate and protect green areas outside the project (90.33%).

‘Local community and economy’ refers to “governance models based on inclusive and responsible management, community participation, and business ecosystems seeking to foster thriving communities and local economies”, totalling 21 final criteria. No category in this theme ranked in the top 10 ranking. The category with the highest score (Table 3) was local governance and participation (12, 83.06%).

Here, the criteria with the top three scores were all in the same category as well (local governance and participation): a first-place tie between regenerative and circular policies and guidelines for buildings and infrastructure, and ensure inclusive, equitable, and collaborative participation of urban stakeholders (90.32%), followed by support regenerative and circular initiatives developed by the community (87.10%), and a third-place tie between monitor and report performance/satisfaction (pre, during and post transition), and implement regenerative/circular procurement and contract practices (83.87%).

4. Discussion

Given the long list of items, in the following subsections, only selected aspects related to the most relevant criteria in each framework theme are discussed and contextualised in the literature.

4.1. Flows and Stocks (F&S)

The ‘built environment stocks and flows’ and the ‘resource sourcing’ categories are intrinsically related to a range of CE concepts, from an adequate sourcing of resources from responsible primary or secondary sources to an adequate design for the future [63] that allows for multiple cycles of use and extended lifetime, and the provision of supporting infrastructure, such as materials passports and marketplaces [64]. There was no consensus regarding abundant resources prioritisation. This may require a more widespread awareness about of the issues with non-abundant, over-exploited and critical resources, and the risks of supply shortage in the future [65,66], while also adopting an integrated approach with other measures to improve resource efficiency [67].

‘Water’, fundamental to life, is a crucial resource to be managed in a more systemic manner [68] through an implementation of water-sensitive principles [69] that could enable the water cycles inside and outside project boundaries [70], and at all phases, including the aspects of embodied water [71]. A systemic application of this framework should also consider how the use of nature-based solutions could enable the circularity of water systems [72], linking with category 4.3 ecosystem services provision.

The ideas in ‘greenhouse gases’ are in line with the three main strategies required to decarbonise cities, as indicated by the IPCC [1]: (1) reduction of energy consumption through efficient urban form and infrastructure, (2) transition of net zero emissions resources and electrification, and (3) increasing carbon uptake and stocks. The effectiveness of urban carbon sinks was questioned by some experts. The literature suggests urban trees may be considered a temporary carbon sequestration and storage (CCS) strategy, reversible in the long term when trees are cut or die, and the high potential of CCS of urban soils still needs more research [73]. Some cities, nevertheless, have been implementing the use of biochar [74], CO2 special concretes, and harvested wood products [75] as carbon sinks. Other comments suggested the unfeasibility of a net zero or climate positive embodied carbon in new constructions. However, there is already research exploring that possibility [76].

Besides the usual approach to energy efficiency and the consideration of renewable energies, one should note the synergies with 5.2 education and awareness to improve user behaviour as an enabler of better energy use patterns in buildings and cities [77,78].

Improving ‘food systems’ in cities through urban community gardens can contribute to improved food security, healthier lifestyles, elimination of food deserts, and reaching other sustainable development goals [79,80], which is of particular importance in pandemic and future-climate scenarios [81]. In this category, however, reclaiming underused spaces, therefore regenerating these areas, was not deemed as relevant. Experts mentioned that underused spaces may not be adequate in context-specific interventions, and that there may be soil contamination. That is certainly true, but regenerative and circular approaches always need to be context specific. An adequate use of horizontal and vertical neglected spaces for greening or food gardens can improve biological interactions and the local biodiversity [82]. If soils may be contaminated even in the backyard of urban houses [83], cleaning up contaminated areas in cities is an important process to regenerate and convert brownfields into productive land [84].

In all cases, decision making for a circular and regenerative approach should derive from a detailed analysis using material flow analysis and life cycle assessment methodologies (environmental, social, economic, and/or sustainability) [85,86,87].

4.2. Urban Systems (US)

A future-proof built environment is crucial in the prospect of increasing climate change impacts, particularly in the most vulnerable areas. Not only regarding physical elements of the city, but also social and economic factors. An ‘adaptive-resilience’ approach not only facilitates bouncing back from disruptive events, but also allows systems to bounce forward to new and better conditions by embedding system integrity, self-(re)organisation capability, and learning from events [88].

The ‘access and diversity’ category seeks to respond to the connections between the design and planning (or its lack thereof) of the city and aspects as social inequity, health, violence, and environmental injustice that usually take place in areas with predominant racial/ethnic minorities and low-income populations [89,90]. Pathways include a people- and place-centred urban design that improves access to green areas and urban services and amenities at walking distances [91], leading to safer [92] and healthier urban spaces [93].

The lack of access to quality water supply and sanitation services have strong impacts in the lives of citizens, particularly in informal settlements [94]. Eliminating energy poverty entails increasing access to affordable and clean energy, with positive impacts to health issues through a clean indoor and outdoor air quality [95,96], increased households’ thermal comfort [97,98] and reduced urban overheating events [99].

Free access to public internet was not considered relevant, despite being considered an essential instrument to support human rights and freedom of speech by the United Nations [100], and the many problems faced mainly by students living in underprivileged areas during the COVID-19 pandemic [101,102].

Regarding ‘urban fabric’, the characteristics of cities’ buildings and infrastructure systems have enduring effects that may last for centuries [103], including the strong linkage to resource use throughout all of their lifecycle [67]. While it may be difficult to implement drastic changes to the existing BE stock, it is possible to define better pathways when densifying or retrofitting urban areas. Nevertheless, the discussions around density and compactness usually provoke strong reactions among citizens. A lack of poor planning, citizen engagement, and understanding of what quality density is may lead to reduced social, ecological, and environmental benefits [104].

4.3. Outdoor Environmental Quality (OEQ)

Feeling safe is an important component in any environment. The notion of urban safety is complementary to crime prevention, as it embeds the comprehension that urban design features, governance, and social and territorial exclusion patterns are strongly related to crime and violence, and aims to identify and mitigate threats and vulnerabilities through participatory processes [105]. And while built environment characteristics affect the perception of safety by citizens [106], aspects of gender, age, race/ethnicity, income, sexual identity and orientation, among others, may strongly affect the experience of safety and the exposure to violence and crime [107,108,109], thus requiring an intersectional approach [110].

Improving air quality has been an increasing global movement. Air pollution has grave impacts on human health, ecological systems, and ecosystem services provision [111], with an expected increase in these effects due to climate change [112]. There are also strong linkages between urban pollution islands and the heat island effect [113]. More than eliminating polluting sources, an active approach to remove air pollutants is needed [114,115].

There is a seemingly misunderstanding and lack of awareness about the soundscape concept even among professionals. The notion of soundscape is strongly related to how citizens perceive their acoustical environment in context [116], which requires an approach that not only mitigates noise and vibration, but also seeks a balance between ensuring quietness and enhancing sounds that may cause pleasant experiences [117]. Carefully crafted soundscapes may produce outcomes as varied as improved health and wellbeing, nostalgic attachment, relaxation, excitement, liveliness, comfort, safety, among others [118].

4.4. Bioconnectivity (BIO)

A discussion of urban ecology needs to add ecology of cities to the notion of ecology in cities [119]. While the latter focus on how natural ecosystems are affected by the urban environment [120], the first adds humans and their activities as key factors in cities [121], and the dependency on the outside of urban boundaries [122], generally linked to high-emitting sectors as energy, agriculture and forestry, and resource extraction [123]. Wu [120] integrates both approaches into a ‘sustainability of cities’ approach, which aligns with the idea of humans as an integral part of nature, as posited by the regenerative approach [39]. The lesson is the importance of regenerating green areas inside and outside cities by adopting a systemic approach.

Large and interconnected urban green spaces (UGS) do have benefits, however, small and disconnected green areas can also play an important role in improving biodiversity [124], and the social and ecological boundaries of cities [125]. While prioritizing infill, greyfield, and brownfield areas for urban retrofit and densification may avoid the occupation of green and agricultural areas at the fringes of cities and the related impacts [126], if made inadequately they may also pose a threat to existing UGS. The notion that land should be seen as a finite resource was mentioned by one expert, and is supported by the literature [84]. The linkages with other categories, particularly urban fabric, need to be acknowledged for effective actions. Implementing ecological buffers may also restrict urban growth at the fringes of the city [127] and protect sensitive areas such as rivers and streams [128].

What history shows is that many cities have turned their backs to rivers by either rectifying, polluting, or covering them to build roads. In a regenerative approach for reconnecting humans with nature, renaturing rivers and other water bodies plays an important role, which may be enhanced by adopting nature-based solutions or NBS [129]. NBS refer to the “actions to protect, sustainably manage, and restore natural or modified ecosystems, that address societal challenges effectively and adaptively, simultaneously providing human well-being and biodiversity benefits” [130].

Despite not ranking high here, NBS are considered by the IPCC an important strategy for climate change mitigation and adaptation [131] with clear socioeconomic benefits and delivery of ecosystem services [132,133]. It is the only pathway, between two others (human development and reduced footprint), capable of generating positive results to counteract the impacts of climate change on ecosystem services [134].

The use of urban planning instruments to enforce the enhancement of high-quality UGS and a better delivery of ecosystem services based on scientific parameters is influential to regenerate cities based on bioconnections [125]. This is already being adopted by some cities, as Malmö’s Green Space Factor, Singapore’s Green Plot Ratio, and São Paulo’s Environmental Quota [135].

4.5. Local Community and Economy (LCE)

Cities should be made by citizens and for all citizens. Thus, there is an increasing importance of good urban governance and inclusive participation in urban planning [136] and regenerative approaches [137]. It seems, however, that this perception was not reflected in the perceptions of participant experts given the lower scores in this theme. This topic needs further exploration in the future for better comprehension. A circular governance model may be characterised by a set of values and principles: participatory, inclusive, transparent, accountable, collaborative, circular (focused and iterative), and fair and just [138]. More importantly, participatory processes should be genuine, making sure that the voices of all relevant and representative stakeholders are heard, rather than being mere consultation processes biased towards interest groups [139].

Improving aspects of education and awareness is essential for urban transitions as they are drivers of change [140], especially when the focus is at the community level. There’s still limited awareness about CE among the general public [141], and even professionals who may have heard of it do not possess a complete or accurate understanding [142]. This aspect, alongside other social issues, was explored with BE professionals in stage 03 of the larger research scope (Figure 2) this paper is a part of [52]. Transitions management requires vocational and academic skills as varied as thematic expertise, creativity, critical thinking, theory application, collaboration and teamwork, diplomacy, systems thinking, oral and communication skills, among others [143]. The provision of adequate facilities/infrastructure—both physical and digital, with equity and inclusivity in access, is also important to promote community-building [144] and enable education and awareness activities regarding technical and daily life knowledge.

Enabling local economies and business ecosystems in neighbourhoods is a key factor to increase economic activities and job availability, and contribute to health and wellbeing [145], with positive impacts to eradicate poverty and other socioeconomic outcomes [20]. This means prioritising small and medium business over large enterprises, which may create opportunities for ethnic minority groups, and whose profit will stay in the locality [91].

These businesses should also have support to transition towards regenerative and circular business models. Regenerative and circular business models, should enable the circular metabolism of resources [146], while moving beyond a focus on the business itself to generate positive impact for the society [147]. Enabling cross sectoral loops by linking local supply chains can facilitate industrial symbiosis at the local level, an important strategy for the circular management of resources [148]. This is applicable for any local business and also for the generation of an interconnected construction sector [149].

The provision of facilities and infrastructure for community building, as argued above, could also support local initiatives to leverage regenerative and circular practices, from banks of talents [150], to repair shops [151], local waste management through composting [152] and the partnership with waste picker cooperatives, extremely important to ensure the flow of resources towards recycling in developing nations [153].

5. Conclusions

Drawing upon the need of cities to respond to the many ecological and social emergencies, and existing gaps of the circular economy approach and neighbourhood sustainability assessment tools (NSAT), this study aimed at understanding in what ways we can implement a regenerative and circular transition of urban areas.

The chosen pathway is by validating a set of criteria based on the ‘Regenerative Circularity for the Built Environment’ (RC4BE) conceptual model proposed in [41] as a support for the transition of existing urban areas. The set of criteria proposed for a RC4BE framework was organised in five themes—(1) flows and stocks, (2) urban systems, (3) outdoor environmental quality, (4) bioconnectivity, and (5) local community and economy.

Some of the identified gaps that were addressed, even if partially, in this study, refer to the consultation of (expert) stakeholders through a transparent process, the inclusion of other aspects of sustainability beyond the aspects of circular management of resources, and a flexible set that may be applicable to different contexts. The proposed framework also responds to previous limitations regarding climate mitigation [15], which is complemented here by aspects of climate adaptation, and biodiversity conservation [17] through the various criteria in bioconnectivity.

Existing NSAT could particularly benefit of more flexible and systemic structures that are less prescriptive and more performance-based [52]. In the case of urban transitions, absolute performances may be less important than focusing on a distance-to-target approach conducted towards community-agreed goals [154].

The two-round Delphi consultation process with sustainability and circular economy experts in the built environment sector that was employed here led to 26 categories and 136 criteria. The results of this process with experts showed that most criteria were adequately defined right at the start, as only eight did not reach consensus after two rounds. That number even increased from the initial 132 to 136 in the end. The set of criteria, however, should not be seen as a fixed checklist of strategies as in other rigid frameworks, but as a guideline to support decision-making.

During the process, it was noticeable how context-biases may affect, not necessarily in a negative way, the responses from some experts. Many of the opinions and comments can be tracked to the context of the respondents’ countries. In a few cases, this was responsible for a lack of understanding that the framework required a broader and systemic perspective. This reinforces the need to clearly establish the synergies among criteria, categories, and themes in the context of each project. Overall, however, there was a high level of agreement. Having experts from nine different countries (even though mostly from Australia, Brazil, and Chile) indicates the criteria are applicable in more than one location with very different contexts, while flexibility for case-by-case adaptations will be needed for more effective implementations. Yet, this does not mean the framework could be applicable and adaptable to every context and situation. This would go against the basic principles of sustainability.

As we adopted a Delphi method, which usually focuses on experts, we were not able to consider the views of non-expert citizens who could certainly bring valuable inputs. We acknowledge that a pre-definition by experts may hinder the design of a fully participatory framework and influence stakeholders’ decisions [155]. However, the decision came because of the constraints in time and the social isolation brought during this study by the COVID-19 pandemic, which hindered the participation of citizens of existing communities. The consideration of the participatory process, nevertheless, should be embedded in the future steps of the framework design, when considering its implementation process. Rather than a consultation with non-experts for the generic framework, a more valid approach could be during each use of the framework, which will require different considerations.

The large number of preselected criteria was another limitation for the use of other methods for weighting items, as AHP; it may also increase the complexity for their execution. At the same time, the set of themes and categories seems flexible enough to allow for customisation by each project to respond to local context issues, and thus contribute to the broader use of the framework in different places and contexts.

The IPCC asserts that CE contributes to several Sustainable Development Goals: Clean Water and Sanitation (SDG6), Affordable Energy and Clean Energy (SDG7), Decent Work, and Economic Growth (SDG8), Responsible Production and Consumption (SDG12) and Climate Action (SDG13) [15]. The RC4BE approach has additional potential to directly contribute to Sustainable Cities and Communities (SDG11), and indirectly to a few others. It has also been designed to facilitate cities’ responses to the burdens imposed into the global and local ecological [4,156] and social [5] boundaries.

While ecological aspects may appear to have a stronger weight, the set of criteria addresses most of the social themes and sub-themes discussed in [157] for a social sustainability framework for neighbourhood design. Nevertheless, we acknowledge that the social aspects of a regenerative circularity may still deserve a more in-depth examination. The proposed set of criteria has given more emphasis to social issues related to urban design, community engagement, awareness, and local economies. This is seen mainly in the ‘local community and economy’ theme, but also considered in other themes, as ‘urban systems’ (access and diversity category) and ‘outdoor environmental quality’ (safety and security, beauty and sense of place, and proxemics and public space use). Including aspects of a social life cycle assessment [158] could provide a good improvement of the framework in the future.

Future steps of this research should include the development of:

• A better integration of social aspects, particularly through participatory planning and co-design,
• a set of metrics to support the implementation and performance monitoring, which should also evaluate if a scoring and weighting system to allow for benchmarking between projects should be adopted,
• a consideration of the ecological externalities, or the impacts from the community outside its boundaries [159]
• a matrix identifying synergies between criteria, and a detailed analysis of their linkages with relevant global initiatives, as the Sustainable Development Goals, and
• an action framework, to support the implementation and operation of regenerative and circular neighbourhoods, based on the idea that the complexity and ever-evolving nature of neighbourhood transitions requires a process-based approach [154].

Here, we sought to expand the circular economy approach in cities. Usually focused on resource cycles, we proposed its merging with the regenerative approach, so that physical resources, ecosystems, liveability, infrastructure, livelihoods, and other socioeconomic aspects of communities are considered. The results of this study may bring new perspective to the adoption of circular economy concepts to the urban scale, particularly in the retrofit of the existing built environment stock. It may also contribute to expand and complement the scope of actions for urban transitions when using tools like LEED, BREEAM, AQUA-HQE, and Green Star, to effectively generate positive impact for people and the planet. Many of the criteria derived from existing NSAT, either directly, or reshaped to fit a regenerative circularity approach. This should enable these tools have space for improvement. The DGNB system [160] is one of those with clear CE criteria.

We expect the ‘Regenerative Circularity for the Built Environment’ framework to be useful not only for built environment professionals, but also other urban stakeholders interested in regenerating their communities by going beyond current green approaches.

The set of criteria established and validated here will be used for the next stage or a larger scope, as previously discussed. There are many unaddressed issues to be solved, particularly due to the complexities of existing neighbourhoods, which may need a transitions management approach [161,162]. This paper, however, presents a new perspective to the adoption of circular economy principles in cities.