GridWise

GridWise helps utilities balance AI demand with renewables and storage, turning grid data into real-time actions.

Comment 1

Inspiration

The rise of AI and data center electricity demand is creating unprecedented stress on utilities already balancing renewables, storage, and outage risks. We wanted to design a tool that helps grid operators, executives, and regulators make real-time, data-informed decisions without needing deep technical expertise. Our inspiration came from seeing how traditional dashboards fall short in surfacing actionable insights when demand spikes or outages loom.

What it does

Our project, GridWise, is an agentic, low-code Tableau Next dashboard that:

  • Forecasts and monitors baseline + AI/data center demand
  • Compares demand against solar, wind, and storage availability
  • Calculates net load and outage risk by region and time
  • Enables scenario-driven what-if analysis (e.g., add 1GW solar, increase storage)
  • Surfaces contextual next-best actions like dispatching storage, triggering demand response, or notifying regulators
  • Quantifies avoided penalties, cost savings, and downtime reductions

How we built it

  • Designed four core datasets (demand.csv, renewables.csv, outage_risk.csv, scenarios.csv) optimized for Tableau’s relational model
  • Cleaned and joined datasets on datetime and scenario metadata
  • Created calculated fields for net load, renewables %, risk classification, and scenario metrics
  • Built a Control Panel with sliders, toggles, and selectors
  • Developed KPI cards, demand/renewables multi-line chart, stacked renewable bars, interactive outage risk heatmap, and a scenario comparison view
  • Implemented agentic recommendations with Tableau’s low-code calculated fields

Challenges we ran into

  • Making synthetic data both realistic and scenario-flexible
  • Designing a dashboard that works equally well for operators (control room view) and executives (strategic view)
  • Implementing conditional recommendation logic in Tableau without external code
  • Ensuring accessibility and clarity under hackathon time constraints

Accomplishments that we're proud of

  • A fully functional Tableau Next prototype with real-time what-if interactivity
  • Agentic recommendation panel that automatically updates with scenario changes
  • Intuitive outage risk visualization across regions and time
  • Storyboard flow that highlights both business value and operational insights

What we learned

  • Tableau Next’s low-code interactivity can handle complex scenario modeling more effectively than we expected
  • Scenario-driven design keeps dashboards engaging for hackathons while being useful for real-world adoption
  • Agentic panels can turn dashboards from passive visualization into active decision-support systems

What's next for GridWise

  • Integrate real-time data feeds from utility APIs
  • Add market price modeling to simulate clean vs. fossil energy dispatch
  • Extend agentic actions into direct dispatch through webhooks or REST APIs
  • Scale beyond synthetic data into utility-grade operational datasets

Built With

Share this project:

Updates