# 045. Household appliance load profile

**Question:** Which hours have the highest typical appliance energy per recorded interval?

## Result

The largest computed metric is 190.4 for 18.0; the smallest is 48.24 for 3.0. Metric: mean_wh (Wh per 10-minute interval).

These are interval-energy values, not watts. This household may not represent another building.

![Household appliance load profile](outputs/chart.png)

| hour | intervals | median_wh | mean_wh | p90_wh |
| --- | --- | --- | --- | --- |
| 0 | 822 | 50 | 52.79 | 60 |
| 1 | 822 | 50 | 51.33 | 60 |
| 2 | 822 | 50 | 49.08 | 60 |
| 3 | 822 | 50 | 48.24 | 60 |
| 4 | 822 | 50 | 49.36 | 60 |
| 5 | 822 | 50 | 52.74 | 70 |
| 6 | 822 | 50 | 57.71 | 70 |
| 7 | 822 | 60 | 78.65 | 110 |
| 8 | 822 | 60 | 106.1 | 269 |
| 9 | 822 | 70 | 112.8 | 290 |
| 10 | 822 | 70 | 125.4 | 310 |
| 11 | 822 | 70 | 133.1 | 330 |
| 12 | 822 | 80 | 123.6 | 290 |
| 13 | 822 | 80 | 124.7 | 290 |

The chart shows 24 of 24 result rows; the table previews the first 14 in the analysis-defined order. [Download the full result table](outputs/results.csv). Numerical values are computed from the source; missing results stay unavailable.

## Method

Hour-of-day medians, means and sample counts.

The study uses shared source preparation and reusable statistical routines. Its specific transformations are in [analysis.py](analysis.py), and common model/evaluation code is in [portfolio/methods.py](../../portfolio/methods.py). The [notebook](analysis.ipynb) executes the study and displays the saved results.

## Decision and limitations

Compare actual schedules and occupancy records before proposing load shifting.

Measurements come from one home over a limited period. Energy is Wh per recorded 10-minute interval. This is not a representative household sample; tariffs and occupancy labels are unavailable. 

Related studies may share observations or holdouts. These are focused analytical studies, not independent replications or deployed business systems. Any model refinements informed by these results need new untouched evaluation data. No commercial impact is inferred from an association or backtest.

## Reproduce

From the repository root, after installing `requirements.txt`:

```powershell
python projects/045-household-appliance-load-profile/analysis.py
```

Source data are downloaded automatically if absent. Original archives are retained unchanged and checked by SHA-256. The cleaned cache normalizes column names; field-specific changes are visible in [data preparation](../../portfolio/data.py). Runtime evidence is in [receipt.json](outputs/receipt.json).

## Source

[Appliances Energy Prediction](https://archive.ics.uci.edu/dataset/374/appliances+energy+prediction), Candanedo (2017). [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Source data are transformed and aggregated in this study. [Source provenance](../../data/provenance/energy.json) and [prepared-data audit](../../data/provenance/energy_prepared.json) record the downloaded files, field coverage and hashes.