bahamas

Submodules

Attributes

`SDLC_stages`
`ODC_types`
`UCA_types`

Classes

`BBN`	Bayesian belief network for reliability analysis of software
`CCCG`	Compute CCCGs for the given list of components of diversity and redundancy system
`Workflow`	Workflow manager for BAHAMAS calculation

Functions

`sdlc_stage_hep_calculation`(excel_file_path, ...[, ...])
`sdlc_stage_hep_calculation_approx`(excel_file_path, ...)	This function reads in the human error probability (mean and std) from a
`get_hemd_from_spreadsheet`(spreadsheet_file[, ...])
`get_stage_odc_dist`(excel_file[, distribution, sheet_name])	Get the distribution of each ODC defect at the given SDLC stage
`stage_dcp_calculation`(excel_file, sheet_name)	Defect conditional probability calculation for each SDLC stage
`get_uca_defect_correlation_dist`(excel_file[, ...])	Assign distribution for each UCA defect correlation term
`plot_histogram`(data_dict, title[, save, show])	Plot histogram for sampled data
`validate_toml`(config)	Validate TOML input file

Package Contents

class bahamas.BBN(defect_file, task_file, num_samples=1000, approx=False, data=None, seed=42)[source]

Bases: object

Bayesian belief network for reliability analysis of software

num_samples = 1000

_approx = False

_data = None

_uca = ['UCA-A', 'UCA-B', 'UCA-C', 'UCA-D']

_odc = ['Algorithm', 'Assignment', 'Checking', 'Documentation', 'Function', 'Interface',...

_sdlc = ['Concept', 'Requirement', 'Design', 'Implementation', 'Testing', 'Install and Maintenance']

_defect

_task

prob_stage

prob_odc

prob_dcp

prob_total = None

prob_uca

_hemd_dist

prob_stage_odc

prob_uca_correlation

_G = 0.25

review_trigger_factor

initialize_stage()[source]

calculate()[source]: Calculate software failure probability based on BBN

get_total_failure_probability()[source]

Get total failure probability

Returns:: mean, sigma, and samples
Return type:: tuple

get_uca(uca_type)[source]

Get UCA probability

Parameters:: uca_type (str) – Type of UCA
Returns:: mean, sigma, and samples
Return type:: tuple

plot(type='all', save=False, show=True)[source]

Plot calculated data

Parameters:

type (str, optional) – Type of plots to plot (defaults to ‘all’)
save (bool, optional) – Save plots into .png file if True (defaults to False)

Returns:

plotly figure object

Return type:

list or figure object

bahamas.sdlc_stage_hep_calculation(excel_file_path, sheet_name, hemd, num_samples=100, distribution='lognorm')[source]

Parameters:

excel_file_path (str) – Filename of the spreadsheet with the number of actions and types
sheet_name (str) – Sheet name in the spreadsheet with the data
hemd (dict) – Dictionary of rvs functions, keyed by action type
num_samples (int) – Number of samples to generate for each action
distribution (str) – Type of distribution to use (currently always “lognorm”)

Returns:

total, fitted (numpy.array, dict) – The samples for the SDLC stage by considering all human error propagations, and a dictionary of the fitted mu and sigma parameters.
This function reads in the number of human actions and human error modes from a
spreadsheet, and calculates the human error probability distributions.

bahamas.sdlc_stage_hep_calculation_approx(excel_file_path, sheet_name, num_samples=100, distribution='norm')[source]

This function reads in the human error probability (mean and std) from a spreadsheet and calculates the human error probability distributions.

Parameters:

excel_file_path (str) – Filename of the spreadsheet with the number of actions and types
sheet_name (str) – Sheet name in the spreadsheet with the data
num_samples (int) – Number of samples to generate for each action
distribution (str) – Type of distribution to use (currently always “norm”)

Returns:

total, None – Samples for the SDLC stage by considering all human error propagations

Return type:

numpy.array, None

bahamas.get_hemd_from_spreadsheet(spreadsheet_file, sheet_name='HEMD', distribution='lognorm')[source]

Parameters:

spreadsheet_file (str) – Filename of the spreadsheet to read in
sheet_name (str) – Name of the sheet with the human error mode distribution data
distribution (str) – Type of distribution to use (currently only “lognorm” is supported)

Returns:

hemd, dist_dict (dict, dict) – The first dictionary contains the rvs sampling function for the action type; the second dictionary contains the distribution dictionary — both are keyed by the action type.
gets the human error mode distributions from a spreadsheet

bahamas.get_stage_odc_dist(excel_file, distribution='beta', sheet_name='ODC')[source]

Get the distribution of each ODC defect at the given SDLC stage P(Defect type|SDLC Stage) = dist_dict[‘SDLC Stage’][‘Defect type’]

Parameters:

excel_file (str) – Filename of the excel file to read in
distribution (str, optional) – Type of distribution to use (defaults to “beta”)
sheet_name (str, optional) – Name of the sheet with the ODC data (defaults to “ODC”)

bahamas.stage_dcp_calculation(excel_file, sheet_name)[source]

Defect conditional probability calculation for each SDLC stage

Parameters:

excel_file (str) – The filename of the spreadsheet with the review number and trigger coverage
sheet_name (str) – The sheet name in the spreadsheet with the data (i.e., the SDLC stage name)

Returns:

defect conditional probability

Return type:

dcp (float)

bahamas.get_uca_defect_correlation_dist(excel_file, distribution='norm', sheet_name='UCA Correlation')[source]

Assign distribution for each UCA defect correlation term

Parameters:

excel_file (str) – Excel file to read in
distribution (str, optional) – Type of distribution (defaults to “norm”)
sheet_name (str, optional) – Name of the sheet with UCA correlation data (defaults to “UCA Correlation”)

Returns:

UCA defect correlation distribution, {‘UCA type’:{‘ODC type’:Dist},…}

Return type:

dict

bahamas.SDLC_stages = ['Concept', 'Requirement', 'Design', 'Implementation', 'Testing', 'Install and Maintenance'][source]

bahamas.ODC_types = ['Algorithm', 'Assignment', 'Checking', 'Documentation', 'Function', 'Interface',...[source]

bahamas.UCA_types = ['UCA-A', 'UCA-B', 'UCA-C', 'UCA-D'][source]

bahamas.plot_histogram(data_dict, title, save=False, show=True)[source]

Plot histogram for sampled data

Parameters:

data_dict (dict) – dictionary of sampled data
title (str) – title for the plot
save (bool, optional) – Save plot into .png file if True (defaults to False)

Returns:

plotly figure object

Return type:

figure object

class bahamas.CCCG(file)[source]

Bases: object

Compute CCCGs for the given list of components of diversity and redundancy system

_sys_diagram

_cccg_final = []

_cccg_function = []

_cccg_input = []

_cccg_design = []

_cccg_single = []

_cccg_function_input = []

_cccg_function_design = []

_cccg_input_function = []

_cccg_input_design = []

_cccg_design_input = []

_cccg_design_function = []

_cccg_double = []

_cccg_function_input_design = []

_cccg_function_design_input = []

_cccg_input_design_function = []

_cccg_input_function_design = []

_cccg_design_input_function = []

_cccg_design_function_input = []

_cccg_triple = []

_function_all = []

_input_all = []

_design_all = []

generate(config=None)[source]

Generate CCCGs based on three coupling factors (i.e., Function, Input, and Design)

Parameters:: config (dict, optional) – config file for output control. Defaults to None.

get(name)[source]

Get CCCGs

Parameters:: name (str) – name for CCCGs group
Returns:: list of CCCGs
Return type:: list (list of Pandas.DataFrame)

write(data, fname, style='csv')[source]

Dump data

Parameters:

data (list of pandas.DataFrame) – output data to dump
fname (str) – file name to save the data
style (str, optional) – type of file. Defaults to “csv”.

aggregate(cccgs)[source]

Aggregate expanded Function, Input, and Design into Function_Config, Input_Config, and Design_Config

Parameters:: cccgs (list) – list of pd.DataFrame (i.e., CCCGs)
Returns:: list of aggregated CCCGs
Return type:: list

class bahamas.Workflow(config)[source]

Workflow manager for BAHAMAS calculation

_config

_bbn_config = None

_ccf_config = None

update_config(config, name)[source]

update default config with user provided data

Parameters:

config (dict) – User-provided config file
name (str) – Name for the config file (i.e., BBN or CCF)

Raises:

IOError – If name is valid, raise the error

Returns:

updated config file

Return type:

dict

initialize_bbn()[source]: Initialize BBN calculation

initialize_ccf()[source]: Initialize CCF calculation

run_bbn()[source]

Run BBN Calculation

Raises:: IOError – Error out if invalid input for analysis type is provided

run_ccf()[source]: Run CCF calculation

run()[source]: Execute the workflow

write(data, fname, style='csv')[source]

Dump data

Parameters:

data (pandas.DataFrame) – Output data to dump
fname (str) – File name for saving the data
style (str, optional) – Type of file (defaults to “csv”)

visualize()[source]

reset()[source]

_validate(config)[source]

Validate input file using JSON schema

Parameters:: config (dict) – Dictionary for input
Raises:: IOError – Error out if invalid

bahamas.validate_toml(config)[source]

Validate TOML input file

Parameters:: config (dict) – Loaded TOML input
Returns:: True if valid
Return type:: bool