Modelling Under Risk and Uncertainty: An Introduction to Statistical, Phenomenological and Computational Methods

8.1 Getting the Costs Inside Risk Modelling – from Engineering Economics to Financial Modelling

8.1.1 Moving to Costs as Output Variables of Interest – Elementary Engineering Economics

Taking Costs into Account in Risk Assessment

In risk assessment stages, system models are generally studied through variables of interest characterising their technical performance such as, for the flood example, the overflow or plant failure. In decision stages, it is often useful to add a cost component to that type assessment through a comprehensive investment + detriment cost figure in order to help answer a fundamental question:

‘What actions (system reliable design, protection design, maintenance) should be taken to best balance costs (& benefits) in a risky/uncertain future?’ (Question 1)

Mathematically, with actions still being represented by vector d, suppose that technical performance at the future time of interest t_f is embedded inside the uncertain variables of interest (noted Z^t) being outputs of the technical system model. As was introduced in the flood example in Chapter 3, the engineering cost components generally comprise:

• Controlled (investment) costs c_i(d): typically the cost of preventive actions, that depend on d while not (or not much) on the performance v.i. Z^t. Think of the dike investment, or of the plant protection system.
• Uncertain (detrimental) costs c_d(d, z): typically the costs implied by a detrimental technical performance, such as damage costs associated with undesired ...