Make the next 72 outdoor hours readable across many cities without turning a communication index into advice or an alert.
3 · What it produces
24 hourly cells per city, 0–100 comfort score, comfort hours/best window, dominant issue, wind and waves where available, 72h cards and IT/EN threads.
The cards expose multiple views of the same workflow so result, method, evidence and history remain distinct. Site preview images use synthetic data (simulated data); operational runs acquire real provider forecasts.
4 · How it works
Multi-city hourly forecast + transparent heuristic score + semantic priority + graceful degradation
1Open-Meteo forecast and marine: apparent/air temperature, RH, dew point, precipitation, snow, wind/gust/direction, WMO code, radiation and waves for compatible coastal cities
2transparent 0–100 heuristic score; semantic priority for thunderstorm, snow, rain, heat, cold, wind, fog, humidity and comfort; best contiguous window
3independent city/day validation, batch→recent cache→single retry, grey cells for missing data, publication only with 4 core cities and at least 80% of the basket
4Persistence, rendering and governed publishing.
5 · Data and signals
| Element | Detail |
|---|---|
| Data and signals | Open-Meteo forecast and marine: apparent/air temperature, RH, dew point, precipitation, snow, wind/gust/direction, WMO code, radiation and waves for compatible coastal cities |
| Industrial transferability | outdoor planning, territorial communication, tourism and non-operational weather dashboards |
| Research | penalty sensitivity, class robustness and comparison with appropriate bioclimatic indices |
| Interpretation condition | not a medical index, PMV/PPD, UTCI, weather alert or normative indoor-comfort metric |
transparent 0–100 heuristic score; semantic priority for thunderstorm, snow, rain, heat, cold, wind, fog, humidity and comfort; best contiguous window
independent city/day validation, batch→recent cache→single retry, grey cells for missing data, publication only with 4 core cities and at least 80% of the basket
Python, requests, pytz, Pillow, GitHub Actions, AT Protocol/Bluesky
treating the score as clinical risk, meteoropathy, official warning or indoor comfort; local microclimate and personal variables are not modelled
Workflow technical dossier
Methods, checks, stack and transferability conditions.
Data and signals
Open-Meteo forecast and marine: apparent/air temperature, RH, dew point, precipitation, snow, wind/gust/direction, WMO code, radiation and waves for compatible coastal cities
Inference / calculation
transparent 0–100 heuristic score; semantic priority for thunderstorm, snow, rain, heat, cold, wind, fog, humidity and comfort; best contiguous window
Validation
independent city/day validation, batch→recent cache→single retry, grey cells for missing data, publication only with 4 core cities and at least 80% of the basket
Stack
Python, requests, pytz, Pillow, GitHub Actions, AT Protocol/Bluesky
Limitations
treating the score as clinical risk, meteoropathy, official warning or indoor comfort; local microclimate and personal variables are not modelled
See recent outputs
The Publications page collects local previews and, on request, the public feed.