Visualization¶

EPM outputs are standard CSV files and a GDX binary — you can load them into any tool you prefer. Three ready-made options are available if you don't want to start from scratch.

Options at a glance¶

Option	Best for	Requires
EPM Dashboard	Quick interactive exploration, non-technical users	Web browser
Python	Custom analysis, scripting, reproducible figures	Python + Jupyter
Tableau	Polished stakeholder dashboards, scenario comparison	Tableau Creator license

EPM Dashboard¶

The built-in web dashboard lets you explore results interactively — without writing any code. It can also be used to launch model runs and configure scenarios directly (see How EPM Works).

The dashboard reads the CSV outputs from output_csv/ automatically once a run completes.

Dashboard overview Results overview — capacity, energy, and cost indicators across scenarios.

Dashboard dispatch Dispatch view — hourly generation by technology.

The dashboard is in continuous development. Feedback and contributions are welcome.

Python Postprocessing¶

For users who want full control over analysis and figures. The postprocessing library is at epm/postprocessing/utils.py and is designed to be used from Jupyter notebooks.

Recommended for: custom plots, aggregation across runs, Monte Carlo analysis, publication-quality figures.

Key functions¶

Data loading

Function	What it does
`extract_gdx(path)`	Reads `epmresults.gdx` into a dict of DataFrames
`extract_epm_folder(folder)`	Loads results from multiple scenarios at once
`process_simulation_results(folder)`	Full pipeline: inputs + outputs, ready to plot
`generate_summary(folder)`	Creates `summary.csv` from all scenario results

Plotting

Function	Chart type
`make_stacked_barplot()`	Capacity/energy evolution across years and scenarios
`make_stacked_areaplot()`	Generation dispatch area chart
`dispatch_plot()`	Combined stacked area + demand line
`bar_plot()` / `line_plot()`	Standard bar and time-series charts
`make_capacity_mix_map()`	Regional map with pie chart overlays
`make_interconnection_map()`	Transmission capacity map
`create_interactive_map()`	Interactive Folium map (HTML)

Quick start¶

from postprocessing.utils import process_simulation_results, make_stacked_barplot

results = process_simulation_results("output/simulations_run_20250101_120000")
make_stacked_barplot(results["pCapacityFuel"], title="Capacity Mix Evolution")

Tip

Use extract_epm_folder(..., save_to_csv=True) to convert any GDX-only variable to CSV on demand.

Example notebook¶

Coming soon

An example Jupyter notebook walking through a full postprocessing workflow will be added here.

Example outputs¶

Utils overview Overview of results across scenarios.

make_stacked_barplot() — capacity mix evolution across scenarios.

dispatch_plot() — hourly dispatch for representative days.

Tableau¶

Interactive dashboards using the EPM Tableau template. Best suited for sharing results with stakeholders or comparing multiple scenarios side-by-side.

Example: SAPP Dashboard on Tableau Public

License required

A Tableau Creator license is needed to create or edit dashboards. Contact the energy planning team for access.

Setup¶

1. Prepare the folder structure

tableau/
├── EPM_Report.twb             # Download template (link below)
├── ESMAP_logo.png
├── linestring_countries.geojson
└── scenarios/
    ├── baseline/              # One scenario must be named "baseline"
    │   └── output_csv/
    │       └── *.csv
    └── scenario_1/
        └── output_csv/

Download the template: EPM_Report.twb

2. Generate the GeoJSON

Update geojson_to_epm.csv with your zone names, then run:

cd epm/postprocessing
python create_geojson.py --folder tableau --geojson geojson_to_epm.csv --zcmap zcmap.csv

This produces linestring_countries.geojson for geographic visualizations.

3. Open in Tableau

Upload the tableau/ folder to OneDrive, connect to the shared VDI (Tableau pre-installed), and open EPM_Report.twb.

4. Extract data

Tableau opens in live mode by default, which is slow. Extract each dataset (Database_Compare, Main_Database, pCostSummaryWeightedAverageCountry, Plant DB, pSummary) via Data → Extract Data → Save settings.

5. Publish

Go to Server → Tableau Public → Save to Tableau Public as to share your dashboard.

Updating for new scenarios¶

Replace CSV files in scenarios/ with new run outputs (keep baseline folder name).
For each dataset: Extract → Remove → then re-extract (Step 4).
Re-publish to Tableau Public.

Tip

If nothing shows up: check the folder structure, verify filters are not hiding data, and confirm extraction completed successfully.