Project Showcase

Exploratory Data Analysis: Urban Energy Insights

Interactive EDA workspace revealing consumption clusters and behavioral archetypes.

Key Technologies

R Tidyverse pipelines and interactive Shiny components.
Observable Linked visual encodings for rapid hypothesis testing.
DuckDB Local analytical warehouse powering ad-hoc SQL exploration.

Live Demo & Visualizations

Problem Statement

Stakeholders needed to understand daily energy consumption behavior across city districts to prioritize infrastructure upgrades and demand response incentives.

Solution Approach

Consolidated smart meter readings into a unified DuckDB dataset accessible from R, Python, and SQL.
Developed a Shiny dashboard with drill-down charts, cluster analysis, and segmentation personas.
Embedded Observable notebooks to compare clustering algorithms and share reproducible narratives.

Results & Findings

Decision makers identified three actionable consumption personas and secured funding for targeted efficiency retrofits.

Stakeholder workshops delivered

Time-to-insight reduction

45%

Compared to static PDF reporting workflows.

Notebook reuse

18 teams

Number of cross-functional squads adopting the reusable notebooks.

Challenges & Lessons Learned

Harmonizing anonymization policies required building automated privacy reports and ensuring each visualization contained clear aggregation messaging.

Datasets & Sources

Hourly energy usage aggregated by district
Weather and mobility indicators aligned to the same temporal granularity

Model Performance

Clustering evaluated using silhouette scores, Calinski-Harabasz, and Davies-Bouldin indices to confirm segmentation stability across random seeds.

Future Work & Improvements

Incorporate indoor air-quality sensors and overlay socio-economic indicators to deepen neighborhood profiles.

Collaboration & Contributions

Contributions welcome via pull requests on the EDA notebooks repository.

Predictive Modeling: Energy Demand Forecasting
Forecasting project that operationalizes insights from this analysis.

BibTeX

@misc{ exploratory-energy-insights,
  title = { Exploratory Data Analysis: Urban Energy Insights },
  author = { Diogo Ribeiro },
  year = { 2026 },
  publisher = { DataLog | Data Science \& Research Theme },
  url = { https://diogoribeiro7.github.io/analytics-blog-jekyll/portfolio/exploratory-energy-insights/ }
}

RIS

TY  - COMP
TI  - Exploratory Data Analysis: Urban Energy Insights
AU  - Diogo Ribeiro
PY  - 2026
PB  - DataLog | Data Science & Research Theme
UR  - https://diogoribeiro7.github.io/analytics-blog-jekyll/portfolio/exploratory-energy-insights/
ER  -

EndNote

%0 Computer Program
%T Exploratory Data Analysis: Urban Energy Insights
%A Diogo Ribeiro
%D 2026
%I DataLog | Data Science & Research Theme
%U https://diogoribeiro7.github.io/analytics-blog-jekyll/portfolio/exploratory-energy-insights/

Open science & reproducibility badges

These badges highlight the transparency practices applied to this work. Hover or focus on each badge to learn more about the criteria.

Open Data Dataset and code repository published with permissive license. Public repository, DOI issued, README with reproduction steps.
Reproducible Workflow Containerized environment and automated tests provided. Continuous integration pipeline with reproducibility checks.
Transparent Peer Review Peer review reports archived with DOI and linked to article. Open peer review statement and archived reports on Zenodo.