API

Calculations

manufacturing.suggest_specification_limits(data: (typing.List[int], typing.List[float], <class 'pandas.core.series.Series'>, <class 'numpy.ndarray'>), sigma_level: float = 3.0)

Given a data set and a sigma level, will return a dict containing the upper_control_limit and lower_control_limit. values

Parameters

  • data – the data to be analyzed

  • sigma_level – the sigma level; the default value is 3.0, but some users may prefer a higher sigma level for their process

Returns

a tuple containing the upper_specification_limit and upper_specification_limit keys

manufacturing.calc_ppk(data: (typing.List[int], typing.List[float], <class 'pandas.core.series.Series'>, <class 'numpy.ndarray'>), upper_specification_limit: (<class 'int'>, <class 'float'>), lower_specification_limit: (<class 'int'>, <class 'float'>))

Calculate and return the Pp (upper) of the provided dataset given the upper control limit.

Parameters

  • data – the data to be analyzed

  • upper_specification_limit – the upper specification limit

  • lower_specification_limit – the lower control limit

Returns

the ppk level

manufacturing.calc_pp(data: (typing.List[int], typing.List[float], <class 'pandas.core.series.Series'>, <class 'numpy.ndarray'>), upper_specification_limit: (<class 'int'>, <class 'float'>), lower_specification_limit: (<class 'int'>, <class 'float'>))

Calculate and return the Pp of the provided dataset given the control limits.

Parameters

  • data – the data to be analyzed

  • upper_specification_limit – the upper control limit

  • lower_specification_limit – the lower control limit

Returns

the pp level

Plotting & Charting

manufacturing.ppk_plot(data: manufacturing.visual.ListValues, upper_specification_limit: (<class 'int'>, <class 'float'>), lower_specification_limit: (<class 'int'>, <class 'float'>), parameter_name: typing.Optional[str] = None, threshold_percent: float = 0.001, is_subset: bool = False, show_dppm: bool = False, figure: typing.Optional[matplotlib.figure.Figure] = None)

Shows the statistical distribution of the data along with CPK and limits.

Parameters

  • data – a list, pandas.Series, or numpy.ndarray representing the data set

  • upper_specification_limit – an integer or float which represents the upper control limit, commonly called the UCL

  • lower_specification_limit – an integer or float which represents the upper control limit, commonly called the UCL

  • parameter_name – a string that shows up in the title

  • threshold_percent – the threshold at which % of units above/below the number will display on the plot

  • is_subset – False if the data represents a complete dataset, else True; determines if Ppk or Cpk are in the titles

  • show_dppm – True if defective parts-per-million are to be shown

  • figure – an instance of matplotlib.figure.Figure

Returns

an instance of matplotlib.figure.Figure

_images/ppk_plot.png
manufacturing.cpk_plot(data: manufacturing.visual.ListValues, upper_specification_limit: (<class 'int'>, <class 'float'>), lower_specification_limit: (<class 'int'>, <class 'float'>), parameter_name: typing.Optional[str] = None, subgroup_size: int = 30, max_subgroups: int = 10, figure: typing.Optional[matplotlib.figure.Figure] = None) → matplotlib.figure.Figure

Boxplot the Cpk in subgroups os size subgroup_size.

Parameters

  • data – a list, pandas.Series, or numpy.ndarray representing the data set

  • upper_specification_limit – an integer or float which represents the upper specification limit, commonly called the USL

  • lower_specification_limit – an integer or float which represents the upper specification limit, commonly called the LSL

  • parameter_name – the name of the parameter that will be displayed on the plot

  • subgroup_size – the number of samples to include in each subgroup

  • max_subgroups – the maximum number of subgroups to display

  • figure – an instance of matplotlib.figure.Figure

Returns

an instance of matplotlib.figure.Figure

_images/cpk_plot.png
manufacturing.control_chart(data: (typing.List[int], typing.List[float], <class 'pandas.core.series.Series'>, <class 'numpy.ndarray'>), parameter_name: typing.Optional[str] = None, x_upper_control_limit: typing.Optional[typing.Union[float, int]] = None, x_lower_control_limit: typing.Optional[typing.Union[float, int]] = None, r_upper_control_limit: typing.Optional[typing.Union[float, int]] = None, r_lower_control_limit: typing.Optional[typing.Union[float, int]] = None, highlight_beyond_limits: bool = True, highlight_zone_a: bool = True, highlight_zone_b: bool = True, highlight_zone_c: bool = True, highlight_trend: bool = True, highlight_mixture: bool = False, highlight_stratification: bool = False, highlight_overcontrol: bool = False, max_points: typing.Optional[int] = 60, figure: typing.Optional[matplotlib.figure.Figure] = None) → matplotlib.figure.Figure

Automatically selects the most appropriate type of control chart, based on the number of samples supplied in the data and the max_points and returns a matplotlib.figure.Figure containing the control chart(s).

Parameters

  • data – (List[int], List[float], pd.Series, np.ndarray),

  • parameter_name – a string representing the parameter name

  • x_upper_control_limit – an optional parameter which, when present, will override the internally calculated upper control limit for the X plot; note that this is NOT the specification limit!

  • x_lower_control_limit – an optional parameter which, when present, will override the internally calculated lower control limit for the X plot; note that this is NOT the specification limit!

  • r_upper_control_limit – an optional parameter which, when present, will override the internally calculated upper control limit for the R/S plot; note that this is NOT the specification limit!

  • r_lower_control_limit – an optional parameter which, when present, will override the internally calculated lower control limitfor the R/S plot; note that this is NOT the specification limit!

  • highlight_beyond_limits – True if points beyond limits are to be highlighted

  • highlight_zone_a – True if points that are zone A violations are to be highlighted

  • highlight_zone_b – True if points that are zone B violations are to be highlighted

  • highlight_zone_c – True if points that are zone C violations are to be highlighted

  • highlight_trend – True if points that are trend violations are to be highlighted

  • highlight_mixture – True if points that are mixture violations are to be highlighted

  • highlight_stratification – True if points that are stratification violations are to be highlighted

  • highlight_overcontrol – True if points that are overcontrol violations are to be hightlighted

  • max_points – the maximum number of points to display (‘None’ to display all)

  • figure – an instance of matplotlib.figure.Figure

Returns

an instance of matplotlib.figure.Figure

manufacturing.run_chart(data: (typing.List[int], typing.List[float], <class 'pandas.core.series.Series'>, <class 'numpy.ndarray'>), parameter_name: typing.Optional[str] = None, upper_control_limit: typing.Optional[typing.Union[float, int]] = None, lower_control_limit: typing.Optional[typing.Union[float, int]] = None, highlight_beyond_limits=True, max_points: typing.Optional[int] = 60, figure: typing.Optional[matplotlib.figure.Figure] = None) → matplotlib.figure.Figure

Create a pre-control chart based on the input data. Pre-control charts contain less information and are more suitable for direct usage by production personnel.

Parameters

  • data – a list, pandas.Series, or numpy.ndarray representing the data set

  • parameter_name – a string representing the parameter name

  • upper_control_limit – an optional parameter which, when present, will override the internally calculated upper control limit; note that this is NOT the specification limit!

  • lower_control_limit – an optional parameter which, when present, will override the internally calculated lower control limit; note that this is NOT the specification limit!

  • highlight_beyond_limits – True if points beyond limits are to be highlighted

  • max_points – the maximum number of points to display (‘None’ to display all)

  • figure – an instance of matplotlib.figure.Figure

Returns

an instance of matplotlib.figure.Figure

manufacturing.x_mr_chart(data: (typing.List[int], typing.List[float], <class 'pandas.core.series.Series'>, <class 'numpy.ndarray'>), parameter_name: typing.Optional[str] = None, x_upper_control_limit: typing.Optional[typing.Union[float, int]] = None, x_lower_control_limit: typing.Optional[typing.Union[float, int]] = None, mr_upper_control_limit: typing.Optional[typing.Union[float, int]] = None, mr_lower_control_limit: typing.Optional[typing.Union[float, int]] = None, highlight_beyond_limits: bool = True, highlight_zone_a: bool = True, highlight_zone_b: bool = True, highlight_zone_c: bool = True, highlight_trend: bool = True, highlight_mixture: bool = False, highlight_stratification: bool = False, highlight_overcontrol: bool = False, max_points: typing.Optional[int] = 60, figure: typing.Optional[matplotlib.figure.Figure] = None) → matplotlib.figure.Figure

Create a \(X-mR\) control plot based on the input data.

Parameters

  • data – a list, pandas.Series, or numpy.ndarray representing the data set

  • parameter_name – a string representing the parameter name

  • x_upper_control_limit – an optional parameter which, when present, will override the internally calculated upper control limit for the X plot; note that this is NOT the specification limit!

  • x_lower_control_limit – an optional parameter which, when present, will override the internally calculated lower control limit for the X plot; note that this is NOT the specification limit!

  • mr_upper_control_limit – an optional parameter which, when present, will override the internally calculated upper control limit for the mR plot; note that this is NOT the specification limit!

  • mr_lower_control_limit – an optional parameter which, when present, will override the internally calculated lower control limitfor the mR plot; note that this is NOT the specification limit!

  • highlight_beyond_limits – True if points beyond limits are to be highlighted

  • highlight_zone_a – True if points that are zone A violations are to be highlighted

  • highlight_zone_b – True if points that are zone B violations are to be highlighted

  • highlight_zone_c – True if points that are zone C violations are to be highlighted

  • highlight_trend – True if points that are trend violations are to be highlighted

  • highlight_mixture – True if points that are mixture violations are to be highlighted

  • highlight_stratification – True if points that are stratification violations are to be highlighted

  • highlight_overcontrol – True if points that are overcontrol violations are to be hightlighted

  • max_points – the maximum number of points to display (‘None’ to display all)

  • figure – an instance of matplotlib.figure.Figure

Returns

an instance of matplotlib.figure.Figure

_images/xmr_chart.png
manufacturing.xbar_r_chart(data: (typing.List[int], typing.List[float], <class 'pandas.core.series.Series'>, <class 'numpy.ndarray'>), subgroup_size: int = 4, parameter_name: typing.Optional[str] = None, xbar_upper_control_limit: typing.Optional[typing.Union[float, int]] = None, xbar_lower_control_limit: typing.Optional[typing.Union[float, int]] = None, r_upper_control_limit: typing.Optional[typing.Union[float, int]] = None, r_lower_control_limit: typing.Optional[typing.Union[float, int]] = None, highlight_beyond_limits: bool = True, highlight_zone_a: bool = True, highlight_zone_b: bool = True, highlight_zone_c: bool = True, highlight_trend: bool = True, highlight_mixture: bool = False, highlight_stratification: bool = False, highlight_overcontrol: bool = False, max_points: typing.Optional[int] = 60, figure: typing.Optional[matplotlib.figure.Figure] = None) → matplotlib.figure.Figure

Create a \(\bar{X}-R\) control plot based on the input data.

Parameters

  • data – a list, pandas.Series, or numpy.ndarray representing the data set

  • subgroup_size – an integer that determines the subgroup size

  • parameter_name – a string representing the parameter name

  • xbar_upper_control_limit – an optional parameter which, when present, will override the internally calculated upper control limit for the X plot; note that this is NOT the specification limit!

  • xbar_lower_control_limit – an optional parameter which, when present, will override the internally calculated lower control limit for the X plot; note that this is NOT the specification limit!

  • r_upper_control_limit – an optional parameter which, when present, will override the internally calculated upper control limit for the R plot; note that this is NOT the specification limit!

  • r_lower_control_limit – an optional parameter which, when present, will override the internally calculated lower control limitfor the R plot; note that this is NOT the specification limit!

  • highlight_beyond_limits – True if points beyond limits are to be highlighted

  • highlight_zone_a – True if points that are zone A violations are to be highlighted

  • highlight_zone_b – True if points that are zone B violations are to be highlighted

  • highlight_zone_c – True if points that are zone C violations are to be highlighted

  • highlight_trend – True if points that are trend violations are to be highlighted

  • highlight_mixture – True if points that are mixture violations are to be highlighted

  • highlight_stratification – True if points that are stratification violations are to be highlighted

  • highlight_overcontrol – True if points that are overcontrol violations are to be hightlighted

  • max_points – the maximum number of points to display (‘None’ to display all)

  • figure – an instance of matplotlib.figure.Figure

Returns

an instance of matplotlib.figure.Figure

_images/xbarr_chart.png
manufacturing.xbar_s_chart(data: (typing.List[int], typing.List[float], <class 'pandas.core.series.Series'>, <class 'numpy.ndarray'>), subgroup_size: int = 12, parameter_name: typing.Optional[str] = None, xbar_upper_control_limit: typing.Optional[typing.Union[float, int]] = None, xbar_lower_control_limit: typing.Optional[typing.Union[float, int]] = None, s_upper_control_limit: typing.Optional[typing.Union[float, int]] = None, s_lower_control_limit: typing.Optional[typing.Union[float, int]] = None, highlight_beyond_limits: bool = True, highlight_zone_a: bool = True, highlight_zone_b: bool = True, highlight_zone_c: bool = True, highlight_trend: bool = True, highlight_mixture: bool = False, highlight_stratification: bool = False, highlight_overcontrol: bool = False, max_points: typing.Optional[int] = 60, figure: typing.Optional[matplotlib.figure.Figure] = None) → matplotlib.figure.Figure

Create a moving \(\bar{X}-S\) control plot based on the input data. Recommended for datasets which are to be grouped in subsets exceeding 11pcs each.

Parameters

  • data – a list, pandas.Series, or numpy.ndarray representing the data set

  • subgroup_size – an integer that determines the subgroup size

  • parameter_name – a string representing the parameter name

  • xbar_upper_control_limit – an optional parameter which, when present, will override the internally calculated upper control limit for the X plot; note that this is NOT the specification limit!

  • xbar_lower_control_limit – an optional parameter which, when present, will override the internally calculated lower control limit for the X plot; note that this is NOT the specification limit!

  • s_upper_control_limit – an optional parameter which, when present, will override the internally calculated upper control limit for the R plot; note that this is NOT the specification limit!

  • s_lower_control_limit – an optional parameter which, when present, will override the internally calculated lower control limitfor the R plot; note that this is NOT the specification limit!

  • highlight_beyond_limits – True if points beyond limits are to be highlighted

  • highlight_zone_a – True if points that are zone A violations are to be highlighted

  • highlight_zone_b – True if points that are zone B violations are to be highlighted

  • highlight_zone_c – True if points that are zone C violations are to be highlighted

  • highlight_trend – True if points that are trend violations are to be highlighted

  • highlight_mixture – True if points that are mixture violations are to be highlighted

  • highlight_stratification – True if points that are stratification violations are to be highlighted

  • highlight_overcontrol – True if points that are overcontrol violations are to be hightlighted

  • max_points – the maximum number of points to display (‘None’ to display all)

  • figure – an instance of matplotlib.figure.Figure

Returns

an instance of matplotlib.figure.Figure

_images/xbars_chart.png
manufacturing.p_chart(data: (typing.List[int], typing.List[float], <class 'pandas.core.series.Series'>, <class 'numpy.ndarray'>), parameter_name: typing.Optional[str] = None, highlight_beyond_limits: bool = True, figure: typing.Optional[matplotlib.figure.Figure] = None) → matplotlib.figure.Figure

Create a p-chart based on the provided data. The data must be a dataframe which contains the following columns:

  • pass, which contains a True/False or 1/0 indication of pass/fail status of a test sequence

  • lotid or datetime, either of which will be used to create subgroups; if lotid is provided, then data will be subgrouped into the defined lots; if datetime is provided, then lot sizes will be based on time units (hour, day, week, year) and will automatically be chosen to ensure that some defects are present in each lot size

Assumptions:

  • The probability of non-conformance is the same for each item

  • Each unit is independent of the other

  • The testing procedure is the same for each lot

Parameters

  • data – a dataframe containing two columns, pass and lotid or datetime

  • parameter_name – a string representing the parameter name

  • highlight_beyond_limits – True or False

  • figure – an instance of matplotlib.figure.Figure

Returns

an instance of matplotlib.figure.Figure

_images/p_chart.png
manufacturing.np_chart(data: pandas.core.series.Series, num_of_lots: int = 100, parameter_name: Optional[str] = None, highlight_beyond_limits: bool = True, figure: Optional[matplotlib.figure.Figure] = None) → matplotlib.figure.Figure

Create a np-chart based on the provided data.

Assumptions:

  • True represents conformance to specification and False represents non-conformance

  • The probability of non-conformance is the same for each item

  • Each unit is independent of the other (no duplicates!)

  • The testing procedure should be the same for each lot

Parameters

  • data – a list, pandas.Series, or numpy.ndarray representing the data set and consisting of only True/False values

  • num_of_lots – the number of lots into which to split the data; as this number goes up, the number of units in each lot will decrease

  • parameter_name – a string representing the parameter name

  • highlight_beyond_limits – True if points beyond limits are to be highlighted

  • figure – an instance of matplotlib.figure.Figure

Returns

an instance of matplotlib.figure.Figure

_images/np_chart.png

Import & Export

manufacturing.import_csv(file_path: (<class 'str'>, <class 'pathlib.Path'>), columnname: str, **kwargs) → Union[dict, pandas.core.series.Series]

Imports data from a csv file and outputs the specified column of data as a pandas.Series

Parameters

  • file_path – the path to the file on the local file system

  • columnname – the column name to which the data is associated

  • kwargs – keyword arguments to be passed directly into pandas.read_csv()

Returns

a dict containing a pandas series and the limits of the data to be analyzed

manufacturing.import_excel(file_path: (<class 'str'>, <class 'pathlib.Path'>), columnname: str, **kwargs) → Union[dict, pandas.core.series.Series]

Imports data from an excel file and outputs the specified column of data as a pandas.Series

Parameters

  • file_path – the path to the file on the local file system

  • columnname – the column name to which the data is associated

  • kwargs – keyword arguments to be passed directly into pandas.read_excel()

Returns

a pandas series of the data which is to be analyzed

manufacturing.data_import.parse_col_for_limits(columnname: str)

Return the upper and lower specification limits embedded into the column header.

Examples:

  • speed - just a column header called “speed”

  • speed (lsl=10.4) - a column header with a lower specification limit built-in (no upper)

  • speed (lsl=10.4 usl=15.1) - a column header with both lower and upper specification limits

Parameters

columnname – the column name to parse

Returns

2-value tuple of the form (lsl, usl); returns (None, None) if values not found