Building Composite Indicators with GIS Integration

Reliable decision-making increasingly depends on the ability to combine diverse data into clear, interpretable metrics. Whether assessing urban air quality, socio-economic vulnerability, or sustainability performance, decision-makers often need a single composite measure that reflects multiple dimensions at once. However, constructing robust composite indices can be complex, requiring careful data preparation, normalization, weighting, evaluation and interpretation.

To support this challenge, Hossein Hassani, Leila Marvian Mashhad, Steve Macfeely, Petra Kynclova, Nour Barnat, Fernando Cantu Bazaldua, have introduced the Composite Index Builder (CIB), a new open-source R package, accompanied by a user-friendly Shiny application, designed to guide users through the full process of developing composite indices using standardized and transparent methods.

Guest Authors:

Hossein Hassani (IIASA), Leila Marvian Mashhad (consultant), Steve Macfeely (OECD), Petra Kynclova (UNCTAD), Nour Barnat (UNCTAD), Fernando Cantu Bazaldua (UNIDO)

Getting Started

The R package is available on CRAN:

🔗 https://cran.r-project.org/web/packages/compIndexBuilder/

The Shiny app can be launched locally using:

install.packages("compIndexBuilder")

library(compIndexBuilder)

Documentation and tutorial examples are included within the package.

Why Composite Index Builder?

Composite Index Builder was developed in collaboration with international statistical experts and aligns with best practices recommended by global statistical standards. The tool provides:

  • Step-by-step workflows from raw data to final composite scores

  • Multiple normalization, weighting and aggregation methods

  • Automated quality checks and indicator summaries

  • Interactive visual analytics including rankings, correlations and time series

  • Geospatial (GIS) visualization for spatial comparison and mapping

This means users can construct composite indices in a reproducible, transparent, and interpretable way, critical for official statistics, research institutions, and policy agencies.

 

Example: Assessing Urban Air Quality Across Cities

To demonstrate the tool in practice, we applied the Composite Index Builder to air pollution data from cities worldwide. The dataset includes PM2.5, PM10, and NO2 concentrations, covering multiple regions and years.

Workflow Overview

  1. Upload Data
    Users can load CSV or Excel files containing any numeric indicators and geographic identifiers.
  2. Data Processing & Normalization
    Missing values are flagged; indicators may be normalized using Min-Max or Z-score scaling.
    For pollution, where lower is better, the direction of indicators was automatically inverted.
  3. Composite Index Construction
    The tool allows:
    • Equal or custom indicator weights
    • Linear or geometric aggregation
  4. Results & Visualization
    Outputs include:
    • Rankings of cities
    • Correlation matrix of indicators
    • Interactive geospatial mapping

Interpretation of Results

The rankings showed that cities with lower pollutant concentrations consistently scored higher in the composite index, confirming expected patterns. Correlation results indicated strong relationships between PM2.5 and PM10 in most contexts, while NO2 varied more by traffic and industrial intensity.

The global interactive map provides a spatial overview of air quality performance, allowing users to identify:

  • high-performing cities (dark green)
  • moderate performance areas (yellow)
  • pollution hotspots (red)

This geospatial perspective is particularly valuable for:

  • Urban planning
  • Public health interventions
  • Monitoring progress toward SDG 11.6.2 (air pollution)

 GIS Integration: Making Data Visible and Actionable

One of the strengths of the Composite Index Builder is the optional GIS mapping module, which automatically plots composite scores on an interactive global or regional map.
Users can zoom into cities, hover for details, compare time periods, and export maps for reporting.

This visualization layer transforms numerical results into intuitive insights, improving communication across government agencies, technical users, and the public.


Wider Applications

Although this demonstration focused on air pollution, the Composite Index Builder can support a wide range of assessment needs, including:

  • Quality of life and wellbeing indicators
  • Health or education performance indices
  • Economic resilience and competitiveness indexes
  • Digital transformation and innovation capability metrics
  • Environmental and sustainability assessments

Because the workflow is data-agnostic, it adapts to any domain where multiple indicators need to be aggregated into a meaningful summary measure.