Sustainable & Inclusive Wellbeing (SIW)

• SIW is a framework that combines current wellbeing, inclusion, and sustainability to ensure equitable quality of life now and for future generations.
• SIW employs diverse methods—from generalized means to Bayesian networks—to capture non-compensatory, context-sensitive wellbeing indicators.
• Applications of SIW span digital governance, climate adaptation, education, and organizational settings, driving inclusive and sustainable policy development.

Searching arXiv for the core SIW-related papers and framing sources. Sustainable and Inclusive Wellbeing (SIW) denotes “a state of wellbeing that is sustainable, inclusive, and attentive to the future. It encompasses current wellbeing (multidimensional quality of life), inclusion (the distribution of wellbeing across people and places), and sustainability (the maintenance of conditions for future wellbeing)” (Vieira et al., 9 Jul 2026). In this sense, SIW is neither a synonym for GDP growth nor a reduction of social progress to subjective happiness alone. It joins current quality of life, distributive equity, and intergenerational viability, and it is therefore closely aligned with “Equitable and Sustainable Well-being” frameworks such as BES, with SDG-based thinking on equity and planetary boundaries, and with governance-centered accounts that treat institutions, social protection, and environmental quality as constitutive determinants of wellbeing rather than external background conditions (Onori et al., 2020, Lusseau et al., 2018, Pereira et al., 2024).

1. Conceptual foundations

SIW is built around three linked dimensions: current wellbeing, inclusion, and sustainability. Current wellbeing covers both objective and subjective conditions: health, education, employment, income, housing, safety, environmental quality, life satisfaction, happiness, and perceived quality of life. Inclusion concerns how wellbeing is distributed across individuals, social groups, and places, and extends to cross-border spillovers where one society’s wellbeing depends on ecological degradation or extraction elsewhere. Sustainability concerns the maintenance of the environmental, social, and institutional conditions that support future wellbeing (Vieira et al., 9 Jul 2026).

This orientation differs from GDP-centered evaluation. GDP measures economic output; SIW evaluates whether people can live good lives now and in the future without breaching ecological limits and without excluding disadvantaged groups. The review of wellbeing determinants explicitly argues that political systems and institutions significantly affect well-being and that subjective indicators should be incorporated into public policy decisions, because objective/material indicators alone are insufficient for a full understanding of well-being and governance (Pereira et al., 2024). The BES framework makes the same move institutionally by treating wellbeing as a multidimensional, equitable, and sustainable construct rather than as a residual of economic growth (Onori et al., 2020).

The theoretical bases are likewise plural. Needs theory is used to justify lower limits below which core dimensions cannot be traded away; strong sustainability is used to justify environmental ceilings and limited substitutability between natural systems and other forms of capital; and capability-oriented reasoning supports a view of development as expansion of real freedoms rather than accumulation of output (Vieira et al., 9 Jul 2026, Pereira et al., 2024). A related systems view appears in SDG interaction research, where sustainable success depends not on isolated progress but on the structure of synergies and conflicts among goals (Lusseau et al., 2018).

2. Core domains, determinants, and distributive structure

SIW is empirically organized around recurring domains. The BES system operationalizes this through domains including Health, Education, Work & Life Balance, Economic Well-being, Social Relationships, Security, Subjective Well-being, Landscape & Cultural Heritage, Environment, Politics & Institutions, Research, Innovation & Creativity, and Quality of Services (Onori et al., 2020). Other work lists comparable determinants: income and distribution, employment and job security, housing, health, education, environmental quality, social participation, trust in institutions, and work–family balance (Pereira et al., 2024).

Governance is a central determinant. Democracy, freedom, control of corruption, rule of law, and government effectiveness are treated as major drivers of life satisfaction and broader wellbeing; decentralization is associated with closer alignment of policies with local needs where governance quality is high (Pereira et al., 2024). In the BES Bayesian-network analysis, geographic macro-area is a major structuring factor: AREA is a root node with 21 outgoing edges and indirectly influences 41 of 61 indicators across 12 domains, indicating that territorial inequality is not peripheral but structural (Onori et al., 2020).

Work, education, and inequality recur as especially consequential. In the BES network, Employment Rate is a key systemic variable linking Low Work Intensity, Satisfied Employees, Childcare Services, and Income per Capita, while High School Diploma is a central educational threshold connecting University Degree, Lifelong Learning, and School Early Leavers (Onori et al., 2020). At a global scale, SDG interaction analysis shows that priorities differ by country income: “No Poverty” emerges as a structural priority in low-income countries, while “Reduced Inequalities” becomes the central leverage point in high-income countries, and prioritising poverty alleviation in low income countries and reducing inequalities in high income countries has compounded positive effects on all SDGs (Lusseau et al., 2018). This suggests that SIW is not a single universal policy recipe; its determinant structure is context-sensitive, especially with respect to inequality, poverty, and territorial distribution.

3. Measurement and aggregation

The central measurement problem in SIW is that it is a formative construct: indicators define the construct, are not required to be correlated, and cannot be collapsed without normative loss into a single latent factor. The comparative review therefore evaluates aggregation methods against nine conditions: limited substitutability, penalisation of imbalances, non-linear transformations, respect for environmental ceilings, respect for lower limits, a formative measurement model, no correlation requirement, distributional sensitivity, cross-border spillovers, and intertemporal aggregation (Vieira et al., 9 Jul 2026). No single reviewed method satisfies all nine conditions (Vieira et al., 9 Jul 2026).

The simplest aggregation family is the generalized mean,

$$M(p,w,x) = \left( \sum_{i=1}^{n} w_i x_i^p \right)^{1/p},$$

with the linear additive model

$LAM(x_1, \dots, x_n) = \sum_{i=1}^{n} w_i x_i$

as the fully compensatory case (Vieira et al., 9 Jul 2026). These methods are easy to interpret, but they permit complete compensation: severe environmental degradation or social exclusion can be offset by high scores elsewhere. Geometric means reduce compensation, and penalty-based indices such as AMPI explicitly penalize imbalanced profiles; outranking methods and veto rules go further by treating some thresholds as non-negotiable (Vieira et al., 9 Jul 2026).

Several concrete SIW-oriented measurement systems illustrate these issues. BES uses a large, domain-structured indicator system and Bayesian Networks to study conditional dependencies among indicators rather than assuming a single reflective latent variable. In the BN formalism,

$$P(X_1, \dots, X_n) = \prod_{i=1}^{n} P(X_i \mid \text{Pa}(X_i)),$$

which makes conditional independencies and mediation structures explicit (Onori et al., 2020). At the welfare-accounting level, the dynamic ISEW study shows that the Index of Sustainable Economic Welfare is better than GDP at capturing social and environmental policy effects, but it omits the full environmental costs of growth; the Doughnut, with social thresholds and biophysical boundaries, provides better guidance for policymakers striving for sustainable wellbeing (Dallinger et al., 25 Feb 2026). At the happiness–sustainability interface, Graphical Lasso identifies a direct conditional link between happiness and sustainability with a partial correlation of about 0.21, while Quantile-on-Quantile Regression shows that the relationship is modest, asymmetric, and context-dependent rather than uniform across countries (Chaouch et al., 13 Dec 2025).

A plausible implication is that SIW measurement is best understood as a layered problem. Indicator choice, normalization, accounting conventions, threshold design, and aggregation logic are all constitutive. A single headline number is possible only if these levels are combined deliberately rather than collapsed into a conventional weighted average (Vieira et al., 9 Jul 2026).

4. Organizational, educational, and lifecycle operationalizations

A substantial strand of SIW research treats wellbeing and inclusion as properties of socio-technical processes rather than outcomes appended after production. In software engineering, Sustainable DevOps is defined as an approach that “integrates sustainability into the DevOps process, automating reliable software delivery while considering environmental impact, individual well-being, social responsibility, economic viability, and technical efficiency,” thereby extending DevOps beyond economic and technical performance to environmental stewardship, workforce inclusion, and developer wellbeing (Herati et al., 11 Mar 2025). This explicitly links SIW to digital work: social sustainability includes equity, inclusion, accessibility, diversity, and ethical values, while individual sustainability includes health, empowerment, self-care, and the reduction of stress, depression, and anxiety (Herati et al., 11 Mar 2025).

In education, an SIW-like ethos appears in interventions that connect software engineering with well-being, mental health, and the UN SDGs. A combination of “well-being-focused software projects” and “brief classroom interventions such as short reflective discussions and team-building activities” produced qualitative outcomes including a more human-centred perspective, more team discussions about mental health, and greater awareness of ethical, social, and environmental responsibilities (Graßl et al., 25 Mar 2026). The six reported themes—broadening the perspective of software engineering, shift to human-centred focus, developer wellbeing and mental health awareness, interdisciplinarity and diverse perspectives, teamwork and social connection, and ethics, social responsibility, and societal impact—outline a concrete trajectory from technical training toward sustainable and inclusive professional formation (Graßl et al., 25 Mar 2026).

At the organizational level, the S-Assessment Tool treats social sustainability as a “forgotten pillar” and evaluates organizations across Health and Wellness, Gender Equality, Decent Work and Economic Growth, Reducing Inequalities, and Responsible Production and Consumption (Annarelli et al., 2024). These sections operationalize SIW through mental health, work–life balance, anti-discrimination, equal opportunities, minority rights, intersectionality, and environmental responsibility. In product-lifecycle research, the same logic is formalized more strongly. Sustainability Value is defined as

$$SV = (\text{Env}) \times (\text{Soc}) \times (\text{Eco}),$$

so a function receives $SV=1$ only when it contributes simultaneously to environmental, social, and economic sustainability (Yaldiz et al., 2024). Dignified Well-Being is defined as

$$\text{DWB}_{\text{stakeholder}} = (\text{number of function groups}) \times (\text{number of provided resources}) \times (\text{number of fulfilled needs by provider}),$$

thereby combining dignity, inclusion, capability, and need fulfilment in an inclusive lifecycle context (Yaldiz et al., 2024).

The complementary metrics paper identifies the inclusivity–empowerment pair most strongly associated across ten case studies. The inclusivity metric $I_3$ combines diversity, hierarchy, number of lifecycle phases, and number of interactions; the empowerment metric $E_4$ combines empowering impact, dependency level, and number of means. Their strong positive association indicates the kinds of inclusion that should lead to greater empowerment in product lifecycles (Yaldiz et al., 2024). This suggests that SIW requires more than participation counts: it requires role structure, power balance, access to means, and lifecycle-wide inclusion.

5. Computational modeling and decision support

SIW is increasingly operationalized through computational systems that connect physical environments, social behavior, and wellbeing outcomes. In climate adaptation, a reinforcement-learning framework integrates four interconnected components—long-term rainfall projections, flood modeling, transport accessibility, and wellbeing modeling—and uses an RL agent to identify spatial and temporal policy interventions that help sustain or enhance subjective wellbeing over time (Vandervoort et al., 14 Apr 2025). Climate adaptation is modeled as a Markov Decision Process with state transitions

$s_{t+1} \sim P(\cdot \mid s_t, a_t),$

and the agent seeks to maximize discounted cumulative reward,

$$V^{\pi}(s_0) = \mathbb{E}\left[ \sum_{t=0}^{\infty} \gamma^t R_t \,\middle|\, s_0, \pi \right],$$

where reward is derived from predicted life satisfaction (Vandervoort et al., 14 Apr 2025). The accessibility component is gravity-based,

$LAM(x_1, \dots, x_n) = \sum_{i=1}^{n} w_i x_i$0

and flood-induced mobility losses propagate into a cumulative link model of life satisfaction through motility variables (Vandervoort et al., 14 Apr 2025). This is SIW in operational form: environmental stress is evaluated not only by hazard metrics but by long-run subjective wellbeing.

At urban and regional scales, digital twins are proposed as governance tools for “inclusiveness, fairness, and resilience,” especially in small towns and non-urban environments (Pereira et al., 2023). Smart governance is described as improving quality of life while emphasizing citizens’ role in collaborative decision-making, and digital twins are positioned as evidence-based supports for sustainable local governance and for linking policy outcomes to inhabitants’ perceptions of local governance (Pereira et al., 2023). A related large-scale vision appears in FuturICT, whose Planetary Nervous System, Living Earth Simulator, and Global Participatory Platform are designed to promote social self-organization, self-regulation, well-being, sustainability, resilience, fairness, happiness, and opportunities for social, economic, and political participation (Helbing, 2013).

Behavioral sensing and social-influence studies extend this computational SIW agenda into everyday life. Smartphone-based monitoring of college students’ call duration and call count across places and times provides a behavioral substrate for sustainable mobile health and wellbeing solutions, with the goal of supporting adaptation to stress-inducing life transitions (Kim et al., 2020). In smart residential contexts, sustainability scores are computed as

$LAM(x_1, \dots, x_n) = \sum_{i=1}^{n} w_i x_i$1

allowing weekly feedback loops on food, mobility, consumption, resource use, and environmental citizenship (Fontan et al., 2023). These systems do not by themselves solve SIW, but they show how sensing, modeling, and participatory feedback can make sustainability and wellbeing jointly observable and governable.

6. Debates, limitations, and research frontiers

Several debates structure current SIW research. The first concerns aggregation. Conventional compensatory indices produce similar rankings, but their logic conflicts with strong sustainability and needs-based reasoning; at the other extreme, strict bottleneck methods may ignore meaningful partial improvements. The comparative review therefore concludes that a future SIW composite indicator will require combining methods across levels: non-linear normalisation, non-compensatory aggregation, and measurement-level choices for inclusiveness and spillovers (Vieira et al., 9 Jul 2026).

A second debate concerns scope. Some frameworks are rich in social and subjective content but weak on ecological ceilings; others are strong on environmental accounting but weak on dignity, inclusion, or mental health. The ISEW analysis explicitly shows that welfare can rise while planetary boundaries remain transgressed, which is why the Doughnut is treated as a better guide for sustainable wellbeing than GDP or ISEW alone (Dallinger et al., 25 Feb 2026). Conversely, the happiness–sustainability study shows that sustainability gains are not uniformly associated with happiness: effects are positive in low-happiness, high-sustainability contexts, negative in high-happiness, low-sustainability contexts, and essentially neutral elsewhere (Chaouch et al., 13 Dec 2025). This indicates that SIW cannot be reduced either to environmental performance or to subjective wellbeing alone.

A third debate concerns socio-technical transition risks. The paper on generative AI argues that stable, quality employment is the bedrock of societal strength and introduces the idea of an AI-capital-to-labour ratio threshold beyond which recessionary pressures, social disparities, reduced social cohesion, and diminished “Mental Wealth of nations” may become self-reinforcing (Occhipinti et al., 2024). This places decent work, social cohesion, and meaningful human contribution at the center of SIW under technological change. In a different register, social sustainability research in training and organizational settings repeatedly notes that the social dimension of sustainability has been comparatively neglected, even though engagement, mental health, diversity, and social capital are indispensable to sustainable development dynamics (Bebey, 23 Apr 2025, Annarelli et al., 2024).

Methodological and ethical limitations are likewise recurrent. Many frameworks remain conceptual or early-stage; several papers explicitly note the absence of validated metrics, the shortage of clear methods for operationalizing social sustainability, or the reliance on self-reports and small samples (Herati et al., 11 Mar 2025, Graßl et al., 25 Mar 2026, Annarelli et al., 2024). Digital SIW systems raise privacy, trust, and equity issues, especially when phone metadata, geolocation, or large-scale participatory data infrastructures are involved (Kim et al., 2020, Helbing, 2013). Cross-border spillovers and intertemporal aggregation remain especially underdeveloped: they are central in theory, but no single reviewed method adequately resolves them in practice (Vieira et al., 9 Jul 2026).

Taken together, the literature portrays SIW as a mature normative orientation but an evolving technical field. Its common commitments are clear: multidimensional wellbeing, distributive inclusion, environmental ceilings, social floors, and long-run institutional viability. Its open challenge is to convert these commitments into measurement systems, governance mechanisms, and socio-technical designs that are simultaneously non-reductionist, operational, and normatively explicit (Vieira et al., 9 Jul 2026, Pereira et al., 2024, Lusseau et al., 2018).