spychart 1.0.0

A Python library for Statistical Process Control (SPC) analytics.

spychart

spychart is a Python library for Statistical Process Control (SPC), developed to support analytics at Somerset Foundation Trust. It provides a centralised and flexible framework for monitoring healthcare time series data.

The package supports a range of SPC chart types (including X, moving range, and p-charts) and can be extended further with custom functions. It supports multiple changepoints, allowing control line re-calculation following system changes (e.g., ward reconfigurations). spychart can also account for seasonal variation by calculating separate control limits for each seasonal period, negating the need to separate data into multiple charts.

Installation

To install spychart using pip:

pip install spychart

Simple Usage

from spychart import SPC, spc_functions, spc_plots

# Instantiate SPC object
spc_example = SPC(data_in=df, # df is a Pandas.DataFrame (with DateTimeIndex)
                  target_col="value") # column name in data_in

# Process changepoint
spc_example.add_process_change_date(
    change_date="2020-08-01"
)

# Calculating control lines and storing results in a Pandas DataFrame
example_data = spc_example.calculate_spc(
    spc_calc_func=spc_functions.x_chart,  # importing X-chart
    rule_1=True) # SPC rules bool

# Pass in dataframe to create SPC chart
fig, ax = spc_plots.seaborn_chart(example_data)

For more examples, refer to the Example Notebook

Contributing

Pull requests are welcome. Otherwise, feel free to open an issue to discuss changes you would like to make.