DFTSimulator

class DFTSimulator(
dft: DFT,
*,
seed: int = 42,
relevant_events: Iterable[str] | None = None,
)

Bases: object

High-level simulator for Dynamic Fault Trees.

Builds the random number generator and state-generation information automatically, and resolves failable elements to their BE/dependency names.

Create simulator.

Parameters:

  • dft – DFT to simulate.

  • seed – Seed for the pseudo-random number generator.

  • relevant_events – Additional names to mark as relevant events (beyond the top-level event), or None.

is_done() → bool

Whether the simulation has ended in a sink state. A sink state can either be that the DFT is failed, the state is invalid or no further failures are possible anymore.

Returns:

True iff the simulation has ended.

is_failed() → bool

Whether the DFT is failed.

Returns:

True iff if the top level event is failed.

is_next_dependency_failure() → bool

Returns whether the next failure is due to a dependency (or the BE failing on its own).

Returns:

True iff the failure is triggered by a dependency.

let_fail(
be: str,
*,
dependency_successful: bool = True,
) → SimulationStepResult

Let the given BE fail next. If the BE fails due to a probabilistic dependency, this failure forwarding can be either successful or unsuccessful.

Parameters:

  • be – Name of the BE which should fail next.

  • dependency_successful – Whether the failure forwarding of the dependency was successful.

Returns:

Result of the step (successful, unsuccessful, invalid).

Raises:

ValueError – If be cannot fail, or if dependency_successful=False is given for a failure that is not due to a dependency or that is due to an FDEP.

next_failures() → list[str]

Returns the BEs which can fail next.

Returns:

Names of the BEs which can fail next.

nr_next_failures() → int

Returns the number of possible BEs which can fail next.

Returns:

Number of possible BE failures.

random_fail() → SimulationStepResult

Let a random BE fail next. The next BE is chosen according their associated failure probability.

Returns:

Result of the step (successful, unsuccessful, invalid).

reset() → None

Reset the simulator to the initial state.

simulate_traces(timebound: float, nr_traces: int) → int

Simulate a number of traces via Monte Carlo simulation and check how many led to an overall failure within the given timebound.

Parameters:

  • timebound – Time bound up till which traces are simulated.

  • nr_traces – The number of traces to simulate.

Returns:

The number of traces which led to an overall failure.

status() → tuple[str, dict[DFTElement, str]]

Get current status of DFT elements.

Returns:

Tuple (overall status, dict element -> status).