quantify()-ing binary outcomes

quantify() evaluates whether the top event occurs under each scenario. With the default prob = FALSE, inputs are binary (failed / not failed) for each basic event.

This vignette covers:

Setup: formulate the top event

library(tidyfault)
library(dplyr)
library(purrr)
library(tibble)

data("db_nodes")
data("db_edges")

f <- curate(db_nodes, db_edges) |>
  equate() |>
  formulate()
sort(names(formals(f)))
#> [1] "AF"  "AUF" "BF"  "DC"  "HF"  "MF"  "NF"  "SF"

Single scenario: direct function call

Any scenario can be passed as named arguments matching formals(f):

f(DC = 1, AF = 1, SF = 0, BF = 0, NF = 0, AUF = 0, HF = 0, MF = 0)
#> [1] 2

Many scenarios: quantify() on a data frame

Bundled db_outcomes_binary holds several rows of 0/1 inputs:

data("db_outcomes_binary")
head(db_outcomes_binary)
#> # A tibble: 6 × 9
#>   scenario    DC    AF    SF    BF    NF   AUF    HF    MF
#>      <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1        1     0     1     0     1     0     0     0     1
#> 2        2     0     0     1     0     1     0     0     1
#> 3        3     1     0     0     1     0     0     0     0
#> 4        4     0     1     0     0     1     0     1     0
#> 5        5     0     0     0     1     0     1     0     1
#> 6        6     1     0     0     0     1     0     0     1

quantify() returns one logical top-event outcome per row:

mybinary <- quantify(f, newdata = db_outcomes_binary)
mybinary
#>  [1]  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE FALSE  TRUE  TRUE

Batch evaluation: IT security tree

For larger batches, build a tibble of draws (one column per basic event), then call quantify() once. The example below uses it_security_probs as Bernoulli parameters and fixes the seed for reproducibility.

data("it_security_nodes")
data("it_security_edges")
data("it_security_probs")

f_sec <- formulate(equate(curate(it_security_nodes, it_security_edges)))

set.seed(123)
n_binary <- 5000
batch_binary <- it_security_probs |>
  map(~ rbinom(n_binary, size = 1, prob = .x)) |>
  as_tibble()

outcomes_batch <- quantify(f_sec, newdata = batch_binary)
table(outcomes_batch)
#> outcomes_batch
#> FALSE  TRUE 
#>  4914    86
mean(outcomes_batch)
#> [1] 0.0172

Why vectorize?

Summary

Input quantify() usage
One scenario Named vector, list, or one-row data frame
Many scenarios Data frame / tibble with one column per basic event
Top event type Logical vector (TRUE = top event occurred)

For probabilities instead of 0/1 inputs, see quantify()-ing probabilities.