Step 1
Map the wijk
A stylised choropleth view makes local grid risk spatial before the planner starts tuning assumptions.
Neighborhood energy simulation
WijkWise Planner
A simulation-first planning concept that helps Dutch neighborhoods explore whether the local electricity grid can support electrification across heat pumps, solar, EV charging, and batteries.
Open the prototype to explore the wijk map, scenario builder, AI-agent simulation sequence, metric cards, driver explanations, and first-action recommendation.
Open standalone app →What it is
WijkWise Planner is a simulation-first concept tool that helps Dutch municipalities, housing associations, and neighborhood (wijk) energy planners answer one hard question: can the local electricity grid handle electrification, and what should the neighborhood do first?
The Netherlands is electrifying homes and transport fast, with more heat pumps, solar panels, EV chargers, and batteries every year. WijkWise lets a non-technical planner pick a neighborhood, adjust a few simple controls, and watch projected outcomes resolve for grid congestion risk, transformer headroom, heat-pump readiness, EV charging pressure, solar potential, and battery opportunity. The goal is to move people from uncertainty to action without needing grid engineering knowledge.
Why I started thinking about this
I started thinking about this in the context of the Netherlands after reading news coverage about how grid congestion is becoming a growing problem, and how planning is underway to mitigate it in the years ahead. Those plans are not only about infrastructure. They also involve people working together at the smallest local scale, in neighborhoods, to conserve energy, coordinate demand, and make electrification possible. WijkWise explores what that kind of shared planning could feel like.
The product frames neighborhood electrification as a decision-support loop: choose a wijk, tune adoption assumptions, run the simulation, and convert projected risk into a first action.
Scenario levers
- Heat pump adoption
- EV charging growth
- Smart charging enabled
- Solar expansion
- Shared battery installation
- Insulation-first planning
Planner questions
- How much transformer headroom remains?
- Which intervention lowers the peak fastest?
- Where does congestion risk become critical?
- What changed, and why?
- How confident is the model?
- What should the neighborhood do first?
How it works
The experience is built around a "what happens if…" loop. You explore questions like: what if more homes install heat pumps? What if EV adoption climbs? What if smart charging is enabled, solar expands, a shared battery is added, or insulation comes first?
Map · scenario · outcome. A stylised wijk choropleth map sits on the left, a Scenario Builder in the center, and a projected-outcome panel on the right. You build a scenario, hit Run simulation, and the result resolves spatially.
Named AI agents. When a simulation runs, four named agents animate in sequence: Demand Modeler → Congestion Forecaster → Driver Analyst → Action Advisor. The outcome then appears with a model confidence score that drops for far horizons and aggressive extrapolation.
Every result explains itself. The outcome leads with a plain-language verdict ("This scenario overloads Oud-West's grid"), then interprets each of the six metric cards in a sentence, breaks down what changed and why with colour-coded raises / lowers / shifts load drivers, and surfaces the single highest-impact recommended first action.
Reading the risk
Grid congestion risk runs on a warm scale from green to red:
- Low: comfortable headroom, the grid copes
- Medium: tightening, worth planning ahead
- High: little spare capacity at the peak
- Critical: demand outstrips the transformer; new connections may have to be queued
To teach the controls, the Scenario Builder opens with three one-click sample scenarios tuned to land on Low, Medium, and High outcomes. A first-time user immediately sees how the levers translate into grid risk, then tweaks from there.
Step 2
Build a scenario
Simple controls model heat pumps, EVs, smart charging, solar, batteries, and insulation choices.
Step 3
Run agents
Four named agents animate the demand, congestion, driver, and action-advice reasoning sequence.
Step 4
Explain the result
The outcome surfaces a plain-language verdict, driver breakdown, confidence score, and recommended first action.
Who it's for
It's aimed at the people making local electrification decisions: municipal energy teams, housing associations, and neighborhood planners who need to sense where the grid is heading before committing to a plan. The interface is intentionally warm and approachable rather than technical: English with Dutch terms, soft civic palette, friendly type.
This is a demo concept built with realistic sample data. It is not factual grid data and the figures should not be treated as real; everything in the app is flagged "Demo data · not real grid figures."