IEEE Power & Energy Magazine - July/August 2016 - 21

figure 1. A wide variety of uncertainties in TEP suggests
many different management strategies, modeling techniques, key properties, and classifications.

opportunities offered by an in-depth analysis of uncertainties in TEP is presented.

Conceptual Framework
for Planning Under Uncertainty
Clear, broad, and practical definitions for the basic concepts
of uncertainty and risk are the first and foremost component
of the conceptual framework for TEP under uncertainty.
Uncertainty is defined as a state where it is impossible
to exactly describe the existing state or future outcome.
Because of the impossibility to accurately and precisely predict the future, planning under uncertainty inevitably carries
risk. Risk is defined as the impact of uncertainties in TEP,
commonly deemed as a hazard. Simply put, uncertainty is
the cause while risk is the consequence or impact on TEP
(which is usually negative, see Figure 2), and we refer to any
uncertain quantity or proposition as an uncertainty. For planning purposes, uncertain outcomes are to be modeled while
risk is to be quantified and managed or hedged against.
The quantification of risk relates to some measure of performance, most commonly costs and economic benefits, but

also reliability, sustainability, or other measures in a multicriteria framework.
The typology of TEP uncertainties consists of one conceptual foundation and three practical dimensions (Figure 3). The
conceptual foundation is the distinction between two extreme
natures of uncertainty, inherent variability or aleatory on the
one hand, and lack of knowledge or epistemic on the other.
Each of the three practical dimensions relates to a key property of TEP uncertainties and is useful in categorizing uncertainties but also in identifying similar practical considerations
regarding modeling and planning processes.

Nature of Uncertainty: Aleatory or Epistemic?
The core of the proposed typology is the conceptual distinction between two extreme natures of uncertainty, aleatory or
epistemic. Whereas aleatory uncertainty refers to a natural
variability inherent to a particular process (e.g., hourly wind
speed, human behavioral variability, and earthquake incidents), which cannot always be represented by a probability
distribution, epistemic uncertainty is produced by imperfect
knowledge (e.g., the model to describe competition in electricity markets, generation expansion, and failure rates for new
facilities) and, thus, could possibly be reduced by increased
inquiry. The distinction between epistemic and aleatory
nature has been regarded in sustainability literature as the
more general and fundamental categorization of uncertainty.
To assert the importance and extent of this distinction in
TEP, consider generation expansion in a deregulated industry
(Figure 4). On one extreme, TEP studies often assume that generation expansion is not described by probability distributions.
Hence, generation expansion is modeled by means of alternative scenarios, using a classic "what-if" analysis, minimax
regret, among others. On the other extreme, proactive TEP
models treat generation expansion by means of an equilibrium
model between generation firms. Thus generation expansion
is not uncertain due to the lack of knowledge of the underlying

Uncertainty (The Cause)
1

A State Where It Is Impossible to
Exactly Describe the Existing
State or Future Outcome

Structure
Modeling
Approach

Dimension of Uncertainty
Practical Significance

Nature
2
Risk (The Consequence)
The Impact of Uncertainties
in TEP, Commonly Deemed
as a Hazard

figure 2. The definition of uncertainty as the cause and
risk as the consequence.
22

ieee power & energy magazine

Conceptual
Foundation

Source
Timescale Conceptual
Management
foundaƟonStrategies
Scope of Actions

figure 3. The proposed typology of TEP uncertainties
consists of one conceptual foundation and three practical
dimensions.
july/august 2016

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - July/August 2016