52 ISE Magazine | www.iise.org/ISEmagazine

Inside IISE Journals

research

To explore, or not to explore:

That is the question in running digital twins

Exploration versus replication (or exploitation) has been a

constant struggle whenever global optimization is desired,

but the response surface is complex. Consider the multi-

armed bandit problem, a kind of sequential decision-mak-

ing problem that investigates whether the decision-maker

should move to another arm (exploration) or stick to the

current arm (replication) to minimize regret.

A similar issue arises when running digital twins, a vir-

tual representation of a real physical process or system. As

a part of the emerging trend of using digital technologies,

digital twins play a crucial role for monitoring and testing

the performance of actual systems in many applications.

In their article “Optimal Budget Allocation for Stochas-

tic Simulation With Importance Sampling: Exploration vs.

Replication,” professor Young Myoung Ko from Pohang

University of Science and Technology and professor Eun-

shin Byon from the University of Michigan discuss how

to strike a balance between exploration and replication for

effectively running digital twins.

Running a digital twin often takes a two-level simu-

lation procedure in which the ﬁrst level simulates a sys-

tem’s operating conditions and the second level invokes a

stochastic black-box computer model. Such procedure has

been widely used, including in renewable energy systems

and ﬁnancial engineering. Almost always, the bottleneck is

the limited computational resource, in particular when the

black-box computer model is computationally demanding.

Variance reduction techniques, such as importance sam-

pling, help reduce the computational budget requirement,

but they do not provide a general guideline on whether to

explore or replicate, i.e., whether to run simulations at new

operating conditions or replicate at already sampled condi-

tions to further exploit the stochastic output.

The authors show that simulation schemes that allow

both exploration and replication theoretically provide an

optimal trade-off, but the optimal strategy easily becomes

impractical to implement because it requires a noninte-

This month we highlight two articles in IISE Transactions. The ﬁrst article looks into how to effectively run digital twins

for monitoring and performance testing of actual physical systems. Speciﬁcally, the question is whether one would be

better off spending more resources exploring new input conditions or spending those resources exploiting the current

input condition. The authors found that an exploration-only scheme will do just ﬁne. They demonstrate the merit of their

exploration-only approach using the National Renewable Energy Laboratory’s wind turbine simulator. The second article

attempts to provide an understanding of how ﬂexibility enables better economic decisions in power system planning

expansion. Such understanding is provided by a set of decision rules solved through optimizing the time-based deployment

of a portfolio of generators that meet future energy demands at the lowest possible cost. Using a real-world power

generation expansion problem in the Midwestern United States, the authors show that ﬂexibility brings signiﬁcant cost

savings to complex energy systems design and planning. These articles will appear in the September 2022 issue of IISE

Transactions (Volume 54, No. 9).

Young Myoung Ko Eunshin Byon

August 2022 | ISE Magazine 53

ger number of simulation runs. To tackle the issue, they

propose an alternative scheme that does only exploration

and no replication. They prove that the exploration-only

scheme guarantees lower variance than the implementable

version of the theoretically optimal exploration-and-repli-

cation scheme.

A case study using the National Renewable Energy

Laboratory’s wind turbine simulator demonstrates that the

exploration-only simulation performs well and generates

robust results. The results provide critical insights for, as

well as practical guidelines in, using digital twins for trans-

forming system design and operations.

CONTACT: Eunshin Byon; ebyon@umich.edu; Industrial and Opera-

tions Engineering Department, University of Michigan, 1205 Beal Ave.,

Ann Arbor, MI 48109 and Young Myoung Ko; youngko@postech.ac.kr;

Department of Industrial and Management Engineering, Pohang Uni-

versity of Science and Technology, 77, Cheongam-ro, Nam-gu, Pohang,

Gyeongbuk, 37673, South Korea

Rethinking power systems expansion

planning: How ﬂexibility enables

better economic decisions and

sustainability in the face of uncertainty

As the world shifts toward environmentally friendlier

power generation systems

to ﬁght climate change in

more sustainable ways, the

intermittence of renewable

energy sources such as wind

and solar is becoming a cru-

cial contributor of uncer-

tainty, especially in meet-

ing ever-increasing energy

demands.

Making timely invest-

ments for long-term nation-

al grid expansion of gen-

eration capacities requires careful planning and ﬂexibility

to adapt to constantly changing conditions. Flexibility in

design is a paradigm derived from the theory of real op-

tions. It moves away from standard design approaches by

focusing on complex energy system designs that shift the

distribution of possible outcomes (e.g., net present value,

costs) toward better outcomes while reducing exposure to

downside risks.

In the article “Flexibility and Real Options Analysis in

Power System Generation Expansion Planning Under Un-

certainty,” Aakil Caunhye, lecturer and assistant professor

Aakil Caunhye

IISE publishes four peer-reviewed re-

search journals that address key topics

in the industrial and systems engineer-

ing world. Summaries from various is-

sues are featured in the Research sec-

tion each month in ISE. For information

on how to subscribe or submit articles,

visit iise.org/publications.

IISE Transactions (link.iise.org/iise-

transactions) is IISE’s ﬂagship research

journal and is published monthly. It aims to foster exchange

among researchers and practitioners in the industrial engi-

neering community by publishing papers that are grounded

in science and mathematics and motivated by engineering

applications.

IISE Transactions on Healthcare Systems Engineering (link.

iise.org/iisetransactions_healthcare) is a quarterly, refereed

journal that publishes papers about the application of indus-

trial engineering tools and techniques to healthcare systems.

The Engineering Economist (link.iise.org/engineeringecono-

mist) is a quarterly refereed journal published jointly by IISE

and the American Society of Engineering Education. Devoted

to issues of capital investment, its topics include economic

decision analysis, capital investment analysis, research and

development decisions, cost estimating and accounting, and

public policy analysis.

IISE Transactions on Occupational Ergonomics and Human

Factors (link.iise.org/iisetransactions_ergonomics) is de-

voted to compiling and disseminating knowledge on occu-

pational ergonomics and human factors theory, technology,

application and practice across diverse areas. You can follow

on Twitter at twitter.com/iisetoehf or @iisetoehf.

About the journals

54 ISE Magazine | www.iise.org/ISEmagazine

research

of business analytics at The University of Edinburgh Busi-

ness School; Michel-Alexandre Cardin, senior lecturer and

associate professor in computational aided engineering at

Imperial College London; and former National University

of Singapore undergraduate student Muhammad Rahmat,

now at the Ministry of Trade and Industry Singapore, in-

troduce an intuitive approach to optimize deployment of

power generation systems that allow better ﬂexibility, eco-

nomic performance and sustainability under renewable en-

ergy supply uncertainty.

Their work proposes a new mathematical model that op-

timizes the time-based deployment of a portfolio of genera-

tors to meet future energy demands at the lowest possible

cost. Embedded in the model is a set of decision rules that

allow the conditional adaptation of generator deployment

as uncertainty evolves over time. This deploys capacity only

if and when it is needed, thus contributing to better sus-

tainability. These rules are based on optimized thresholds

from historical observations of the capacities of intermittent

energy sources and allow a portfolio of generators to be de-

ployed only when those capacities fall below the thresholds.

The decision-making framework is demonstrated on the

analysis of a real-world power generation expansion prob-

lem in the U.S. Midwest. The authors show that ﬂexibility

brings signiﬁcant cost savings to complex energy systems

design and planning. This contributes to making better use

of limited economic resources while improving the sustain-

ability of future renewable energy systems.

CONTACT: Michel-Alexandre Cardin; m.cardin@imperial.ac.uk; Dyson

School of Design Engineering, Imperial College London, UK

Yu Ding is the Mike and Sugar Barnes Professor of Industrial

and Systems Engineering at Texas A&M University and Associ-

ate Director for Research Engagement at the Texas A&M Institute

of Data Science. He is editor-in-chief of IISE Transactions and a

fellow of IISE.

Grow mentorship during convenient times

Customize the duration of your partnership

Target partnership within specific societies, divisions or chapters

View your activities with summary reports

I enjoyed being a sounding board to others as well as having open and

uplifting discussions with my team. Being able to talk about work,

professional life, and goals was really helpful, especially in a non-com-

petitive and supportive environment. I liked the structure offered as well

as the option to expand beyond topics given. It was also really nice to

have a positive group to talk to during difficult COVID-19 times. My

group members and I continue to keep in touch, and I have continued

mentorship as well with my mentor.

It’s easy!

Questions? Contact Kathy Gamboa at KGamboa@iise.org.

- IISE member Elyse Stevens, who has served as both mentee and mentor

Share & Like IISE

Follow IISE pages, like posts and share them with

your network of contacts on social media. Watch

and enjoy IISE videos at www.youtube.com/iisechannel.

Go to www.iise.org/social for details on IISE’s social

networks. If you’re a member of IISE, chat with

fellow members on IISE Connect (connect.iise.org)

and its communities speciﬁc to your area of interest.

On-campus, online

and in Manhattan

binghamton.edu/ssie

Systems Science and

Industrial Engineering at

Binghamton University

WATSON COLLEGE

Systems Science and Industrial Engineering

SUNY Distinguished

Professors

tenured/tenure-

track faculty

SUNY Chancellor’s

Award recipients

$10

million

in research

expenditures

in 2020–21

Degree

programs

in Industrial and

Systems Engineering

(BS, MS, PhD), Systems

Science (MS, PhD) and

Healthcare Systems

Engineering (MS)

280

undergraduates

210

master’s

180

doctoral students

Michel-Alexandre Cardin Muhammad Rahmat