Composite Sustainability Score

• Composite sustainability score is a quantitative, multi-dimensional metric that aggregates impacts across environmental, economic, and social pillars using structured weighting and normalization methods.
• It involves a systematic process of indicator identification, scoring, and aggregation—often employing MCDA, PCA, or DEA frameworks to analyze trade-offs and conflicts.
• The score serves as a decision support tool by quantifying trade-offs and sustainability conflicts, demonstrated in case studies like medical device evaluations.

A composite sustainability score is a quantitative, multi-dimensional metric aggregating the sustainability implications of a product, firm, or decision across distinct pillars—most commonly environmental, economic, and social domains—through systematic identification, scoring, weighting, and aggregation of effects or indicators. The resulting scalar provides an operational basis for comparative assessment, trade-off analysis, and decision support in sustainability-sensitive contexts. Methodological rigor in input selection, normalization, weighting, and explicit treatment of trade-offs is central to all credible frameworks.

1. Conceptual Foundation and Pillar Structure

Composite sustainability scores formalize the multidimensionality of sustainability by quantifying impacts across three canonical pillars: Environmental (e.g., resource use, emissions), Economic (e.g., cost, value generation), and Social (e.g., worker safety, community benefit). Each effect or indicator is associated with one pillar, a defined sign (positive/negative), and a calibrated strength or severity level. In advanced frameworks, pillar weights encode stakeholder or regulatory priorities, reflecting the context-specific criticality of particular sustainability dimensions (Chakrabarti, 15 Dec 2025).

The general workflow consists of:

• Systematic enumeration of sustainability-relevant effects using methods such as life cycle analysis and cause–effect mapping.
• Explicit assignment of each effect to a pillar, with sign and quantitative magnitude.
• Definition of pillar-level weights $w_p$ to capture context-dependent prioritization.

2. Mathematical Formulation and Aggregation Frameworks

The aggregation of effects into a final composite score involves several canonical approaches, often instantiated via Multi-Criteria Decision Analysis (MCDA) techniques (Chakrabarti, 15 Dec 2025), or more sophisticated forms such as the hierarchical SMAA–Choquet integral (for synergies/redudancies) (Angilella et al., 2018), or PCA-based approaches in compositional data contexts (Rondós-Casas et al., 8 Sep 2025).

A typical MCDA-based composite score is computed as:

$$P_p = \sum_{j: p_j = p,\, \text{sign}_j = +} \xi_j,\quad N_p = \sum_{j: p_j=p,\, \text{sign}_j = -} \xi_j$$

$P^{(w)}_p = w_p P_p,\quad N^{(w)}_p = w_p N_p$

$P = \sum_p P^{(w)}_p,\quad N = \sum_p N^{(w)}_p$

$T = P + N,\quad R = \frac{N}{T}$

Here, $\xi_j$ denotes the effect strength (e.g., 0.25/0.5/0.75), and $R \in [0,1]$ is the composite sustainability ratio: $R \rightarrow 0$ is fully sustainable (all positives), $R \rightarrow 1$ is fully unsustainable (all negatives) (Chakrabarti, 15 Dec 2025).

Alternate data-fusion frameworks specify:

• Weighted arithmetic/geometric aggregations with normalization and statistically or subjectively derived weights (Konak, 12 Jul 2025).
• Principal-component extracted weights in PCA-CODA for aggregating ratios (Rondós-Casas et al., 8 Sep 2025).
• DEA/BoD-based composite efficiency scores for purely output-driven sustainability measurement (Kapelko et al., 2023).

Table: Canonical Aggregation Schemes

Approach	Formula for Score	Notable Features
Weighted Sum (MCDA)	$\sum_i w_i s_i$	Compensatory, transparent
Choquet Integral	$\sum_i (x_{(i)}-x_{(i-1)}) v(\{(i),…,n\})$	Interaction-aware
PCA-based	$\sum_k w_k t_{ik}$	Correlation-aware
DEA (BoD)	Maximize $\sum_r w_r y_{r,0}$ s.t. constraints	Unit-invariant, non-compensatory

3. Input Identification, Life Cycle Analysis, and Effect Formalization

Identification of scoring inputs proceeds via a comprehensive life cycle analysis, which decomposes the object of study (e.g., medical device) into discrete stages (raw materials, manufacturing, transport, use, end-of-life). At each stage, cause–effect mapping enumerates all relevant decisions and their direct/indirect impacts per pillar.

Each effect is systematically codified with:

• A precise pillar assignment ($p_j$),
• A signed impact (positive or negative),
• A severity grade ($\xi_j$), calibrated as categorical–numerical (e.g., Low = 0.25, Medium = 0.5, High = 0.75).

This formalization ensures completeness and granularity and constructs an explicit database of sustainability trade-offs for downstream scoring (Chakrabarti, 15 Dec 2025).

4. Weighting, Normalization, and Domain Prioritization

Assignment of pillar and indicator weights is central to both MCDA-based and advanced aggregation frameworks. Weights may be determined by:

• Pairwise priority exercises or stakeholder elicitation (Chakrabarti, 15 Dec 2025),
• Analytic Hierarchy Process (AHP),
• Statistical methods (variance-based, entropy-based, PCA),
• Optimization to enhance predictive links with external variables (e.g., risk measures in ESGM construction) (Sahin et al., 2021).

Normalization methods vary according to domain:

• Min-max scaling to $[0,1]$ is standard.
• Centered log-ratio transformation for compositional data (Rondós-Casas et al., 8 Sep 2025).
• Z-score normalization for aggregating over heterogeneously distributed indicators (Konak, 12 Jul 2025).

Explicit handling of missingness, through statistical imputation or via dedicated “missingness” pillars (M) with associated weights, is recommended for transparent risk attribution (Sahin et al., 2021).

5. Trade-off Quantification and Conflict Resolution

Quantitative decomposition of sustainability trade-offs is critical for actionable interpretation. By assigning effect-specific strengths and separately aggregating positive ($P$) and negative ($N$) contributions per pillar, the frameworks render the sustainability conflict space numerically explicit. The total impact magnitude $T$ serves as a proxy for overall design complexity or conflict intensity (Chakrabarti, 15 Dec 2025).

Resolution strategies include:

• Redesign to attenuate high-strength negative effects,
• Enhancing compensating positives,
• Adjusting pillar weights to reflect new strategic or regulatory priorities.

Design alternatives may thus be compared by their $R$ score, and conflict-intensity metric $T$, over full parametric sweeps (Chakrabarti, 15 Dec 2025).

6. Application: Medical Device Case Study

In the explicit medical device implementation (Chakrabarti, 15 Dec 2025):

• The oxygen concentrator was mapped to five core sustainability conflicts.
• Raw negative/positive impact sums were computed by pillar.
• Chosen weights: $w_{\text{env}}$=0.50, $w_{\text{econ}}$=0.25, $w_{\text{soc}}$=0.75.
• Weighted negative and positive impacts aggregated as $N = 2.625 + 0.125 + 1.3125 = 4.0625$, $P = 0 + 0.625 + 1.125 = 1.75$.
• Sustainability ratio $R = N/(P+N) \approx 0.6982$, categorizing the design as “Highly Unsustainable.”

7. Strengths, Limitations, and Practical Considerations

Strengths:

• Full numeric transparency of trade-offs and pillar contributions.
• Domain-prioritized weighting aligns with regulatory and stakeholder imperatives.
• Robustness to missing data through explicit modeling or risk-driven weight adjustment.

Limitations:

• Sensitivity to subjective weight selection and severity mapping.
• Exclusion of nuanced, non-explicit trade-offs if cause–effect mapping is incomplete.
• Necessity for high-quality, comprehensive LCA and effect data.

The composite sustainability score paradigm thus provides a robust, extensible apparatus for quantitative sustainability assessment, under rigorous methodological controls. It supports systematic comparison, conflict analysis, and optimization of sustainability in the design and evaluation of products, processes, and organizations (Chakrabarti, 15 Dec 2025).