52 ISE Magazine | www.iise.org/ISEmagazine

Inside IISE Journals

research

Enhancing the grain storage

management capability by

harnessing the sensor data

Grain storage is a critical issue for the

national economies of a majority of de-

veloping countries. The quality of grains

will decrease if grains are not properly

preserved in granaries. Thermal man-

agement plays an essential role in grain

storage because temperature is among

the key factors that may directly inﬂu-

ence the quality of stored grains.

The thermal ﬁeld of a whole granary

is of great importance in grain thermal

management. The accurate acquisition of

the whole thermal ﬁeld information gen-

erates a meaningful index for grain qual-

ity surveillance and storage maintenance.

Therefore, a timely and effective thermal

ﬁeld estimation approach is necessary for

grain storage.

Doctoral student Di Wang and profes-

sor Xi Zhang from Peking University

and professor Kaibo Liu from the Uni-

versity of Wisconsin-Madison developed

an online thermal ﬁeld estimation ap-

proach in grain storage in their article,

“Modeling of a Three-Dimensional Dy-

namic Thermal Field Under Grid-Based

Sensor Networks in Grain Storage.” This

article combines a 3D nonlinear physical

dynamics model with a stochastic spatio-

temporal model to accurately obtain a

dynamic thermal ﬁeld during grain stor-

age by using the sensor data collected in

the granary.

The technique in this article has been

well applied in several national grain de-

pots in China and has achieved satisfac-

tory thermal ﬁeld estimation results. This

technique provides guidance for the de-

sign of engineering structures to conserve

energy and reduce preservation costs. For

example, an accurate estimation of the

grain thermal ﬁeld can help optimize the

layout and enact on-and-off strategies of

sensor networks in granaries. This tech-

nique has been honored in “Grain Sci-

ence and Technology Achievements of

China” in 2017 and 2018.

CONTACT: Xi Zhang; xi.zhang@pku.edu.cn; +

86-010-82524911; Department of Industrial Engi-

neering & Management, Peking University, Beijing

100871, China

Enabling ‘lab-to-fab’

transition of metal-based

additive manufacturing

Laser-based additive manufacturing

(LBAM) is a disruptive technology that

can greatly enhance future capabilities

for the commercial and defense indus-

tries. One major challenge of imple-

menting LBAM is that parts must often

meet very stringent requirements for

qualiﬁcation and certiﬁcation, requiring

This month we highlight two articles from IISE Transactions. The ﬁrst article studies the sensing system design and

estimation to provide a thermal ﬁeld of a whole granary for grain quality surveillance and storage maintenance. The

developed technique has been implemented in several national grain depots with great successes. The second article

develops machine learning methods for facilitating the laboratory to fabrication (lab-to-fab) transition of additive

manufacturing technology. The developed method provides the capability of using in-situ sensor data to predict defects in

the AM part, which speed up the transitioning “laboratory” research to component-level “fabrication.” These articles will

appear in the May 2019 issue of IISE Transactions (Volume 51, No. 5).

Doctoral student Di Wang from Peking University in Beijing stands in an empty grain

warehouse before sensor layout initiation.

April 2019 | ISE Magazine 53

long lead times for introducing AM parts

into mission-critical applications in de-

fense. The unique beneﬁts of rapid build

time and unique microstructural control

offered by AM processes cannot be fully

realized without real-time process tech-

nologies that accelerate performance-

based qualiﬁcation and certiﬁcation.

Mississippi State University research-

ers collaborated with the U.S. Army

Research Laboratory (ARL) to develop

machine learning methods for facilitat-

ing the laboratory to fabrication (lab-to-

fab) transition of AM technology. This

ongoing collaboration utilizes LBAM

to customize lightweight, complex parts

while maintaining mechanical proper-

ties and quality. The LBAM process

provides the capability to tailor the ap-

plication and reduce weight to ensure

the resilient platforms and systems that

meet the Army’s modernization needs.

This interdisciplinary team included

Ph.D. student Mojtaba Khanzadeh,

former Mississippi State student Su-

dipta Chowdhury, professors Haley R.

Doude, Mohammad Marufuzzaman

and Linkan Bian, all from the Bagley

College of Engineering at Mississippi

State University, and ARL scientist

Mark Tschopp, now regional lead of

ARL Central. In their research paper,

“In-Situ Monitoring of Melt Pool Imag-

es for Porosity Prediction in Laser-based

Additive Manufacturing Processes,” the

team developed a new machine-learn-

ing method that used sensor data (melt

pool thermal image streams from ex-

tensive experimental studies) to predict

defects in the resultant part (locations

of porosity measured via CT scan). The

study demonstrates the efﬁcacy of real-

time process monitoring for defects in

LBAM.

The success of this research signiﬁ-

cantly contributes to real-time process

monitoring and control technologies,

which will improve the quality, pro-

ductivity and cost reduction required

for wide adoption of metal AM pro-

cesses. This critical study has wide-

reaching implications for accelerating

the selection and optimization of build

parameters and developing a better fun-

damental understanding of the interac-

tion of material and process on LBAM

component properties. Both are neces-

sary components for reaching the goal

of “lab-to-fab,” efﬁciently transitioning

“laboratory” research to component-

level “fabrication” in industry.

CONTACT: Linkan Bian; bian@ise.msstate.edu;

(662) 325-0570; Mississippi State University, 260

McCain Building, Mississippi State University,

MS 39762.

A dynamic target volatility

strategy for asset allocation

Since the collapse of Lehman Brothers

in 2008, it has become increasingly im-

portant for investors to manage ﬁnancial

risk through tail risk hedging. This is

especially true of institutional inves-

tors with long-term investments such

as funds and insurance companies who

have established a risk-control strategy

known as “target volatility” to maintain

a predetermined level of volatility for a

portfolio.

The target volatility strategy system-

atically adjusts the investment propor-

tion of the portfolio between a risky and

a risk-free asset to maintain a desired

level of target volatility. Accurately fore-

casting the volatility of the risky asset

return is a critical task when implement-

ing a target volatility strategy.

In “A Dynamic Target Volatil-

ity Strategy for Asset Allocation using

Artiﬁcial Neural Networks,” authors

Youngmin Kim from Soonchunhy-

ang University and David Enke from

the Missouri University of Science and

Technology developed an intelligent

target volatility strategy for asset alloca-

tion. The strategy provides a volatility

forecast using artiﬁcial neural networks,

based in part on a stability-oriented ap-

proach (SOA), while also incorporating

target volatility.

The model is unique in that it ap-

plies an SOA based on statistical tests to

compare data from the current period to

a past set of data from a stable period.

This research summary highlights

an article from Volume 63, No.

4, of The Engineering Economist.

Many investors manage risk by

attempting to maintain a constant

target level of volatility in their

portfolios. In this article, the

authors use an artiﬁcial neural

network to forecast the volatility of a

risky asset based on historical data

and equivalent futures contracts.

Mississippi State University researchers involved in the “lab to fab” project for additive

manufacturing included (from left) Ph.D. student Mojtaba Khanzadeh and professors

Mohammad Marufuzzaman, Haley R. Doude and Linkan Bian.

Mark Tschopp Sudipta Chowdhury

54 ISE Magazine | www.iise.org/ISEmagazine

This provides a higher level of reliability

due to a more abundant source of stable

volatility data compared to traditional

approaches that often model on a smaller

amount of unstable drawdown data.

In addition to applying intelligent

classiﬁcation and forecasting techniques,

according to the market situation as ob-

tained from a SOA and a volatility index,

equivalent futures contracts are used

in the model instead of the underlying

portfolio assets. This approach is more

practical and cost-effective for managers

and investors compared to conventional

models for asset allocation.

Working together, the model acts as a

risk-management technique for achiev-

ing long-term investment and capital

allocation objectives. It provides bet-

ter portfolio performance compared to

benchmark strategies in terms of both

the downside risk of the portfolio (i.e.,

fat-tailed distributions) and a common

risk-adjusted measure of return (i.e.,

Sharpe ratio).

In the end, the hope is that intelligent

and dynamic risk models can help inves-

tors receive expected long-term returns,

even as the volatility of individual assets

and the markets change over time.

CONTACT: David Enke; enke@mst.edu; 221 Engi-

neering Management, Missouri University of Sci-

ence and Technology, Rolla, MO 65409.

Jianjun (Jan) Shi is the Carolyn J. Stewart

Chair and professor in the H. Milton Stewart

School of Industrial and Systems Engineer-

ing at the Georgia Institute of Technology. He

is editor-in-chief of IISE Transactions, an

academician of the International Academy for

Quality and a fellow of IISE, ASME and

INFORMS.

Sarah M. Ryan is the Joseph Walkup Profes-

sor of Industrial and Manufacturing Systems

Engineering at Iowa State University. She is

editor-in-chief of The Engineering Economist

and a fellow of IISE.

research

My fascination

with industrial

engineering comes from

that never-ending search

for optimization and

efﬁcacy. We don’t only

search for solutions to beneﬁt manufacturing,

we also facilitate and improve people’s lives by

making products more affordable. We can make

basic daily tools readily available to improve the

quality of products and services. Reducing waste

and inefﬁciency also saves people time, which

they can dedicate to their close ones, family

and themselves.”

Give Back

to the Future

of the Profession

Donate to the

IISE Scholarship Fund

The IISE Scholarship Fund recognizes industrial

and systems engineering students’ academic

excellence and campus leadership. Last year,

IISE awarded more than $90,000 in scholar-

ships. Eenne Bausta is pursuing his B.S. in

Industrial Engineering with a minor in Systems

Engineering at ITESM – Queretaro in Mexico.

He won IISE’s Dwight D. Gardner Scholarship.

Visit www.iise.org/PlannedGiving to make a

donaon today.

IISE Transactions is IISE’s ﬂagship research

journal and is published monthly. It aims

to foster exchange among researchers and

practitioners in the industrial engineering

community by publishing papers that are

grounded in science and mathematics and

motivated by engineering applications.

The Engineering Economist 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.

To subscribe, call (800) 494-0460 or (770)

449-0460.

About the journals

Youngmin Kim David Enke