Cross-Domain Usage

This page summarizes cross-domain examples for applying voiage outside health economics.

voiage’s core VOI methodology is domain-agnostic. While the library was initially developed for health technology assessment (HTA), the same EVPI, EVPPI, EVSI, and ENBS calculations apply to any decision problem under uncertainty.

Supported Domains

Healthcare — cost-effectiveness analysis, HTA, clinical trial design
Finance — investment decisions, portfolio optimization, risk management
Environmental — environmental policy, ecosystem management, climate adaptation
Engineering — infrastructure decisions, reliability analysis, design optimization
Public Policy — regulatory decisions, program evaluation

For runnable cross-domain examples, see:

A minimal cross-domain workflow

The public API accepts a sampled net-benefit matrix, so the same workflow can be used for engineering designs, investment strategies, environmental policies, or other decisions. Keep the domain model in the callback that produces net benefits and let the Rust-backed analysis layer evaluate the information value:

import numpy as np

from voiage.analysis import DecisionAnalysis
from voiage.schema import ValueArray

rng = np.random.default_rng(42)
uncertainty = rng.normal(size=(1_000, 3))
net_benefits = np.column_stack(
    [
        100 + uncertainty[:, 0] * 8,
        105 + uncertainty[:, 0] * 4 - uncertainty[:, 1] * 3,
        98 + uncertainty[:, 2] * 12,
    ]
)

analysis = DecisionAnalysis(
    nb_array=ValueArray.from_numpy(net_benefits, ["Design A", "Design B", "Design C"]),
    parameter_samples={"uncertainty": uncertainty},
)
print(analysis.evpi())

The full page covers engineering design and portfolio examples, parameter modelling, interpretation, and reproducibility considerations.