Digital Twins: Essential to Driving Sustainable

Operations for Chemical Producers

by Paige Marie Morse, PhD, Sustainability Advisor, Aspen Technology, Inc.
by Geeta Pherwani, Sr. Product Marketing Manager, Aspen Technology, Inc.

White Paper
Introduction
Asset intensive industries are facing a challenging future. They must
manage the fragile balance of adopting sustainable operations while
meeting shareholder expectations for profitability. At the same time,
volatile and uncertain market conditions and a rapidly changing
workforce are contributing to a loss of domain knowledge and expertise,
and creating complicated operating scenarios that demand new solutions.

For chemical producers, digital technologies continue to be an essential

enabler for kickstarting sustainability initiatives and reaching targets. For
example, digital solutions can help organizations more easily integrate
data from isolated systems to a centralized location, providing insights to
drive improved business decisions for sustainable operations.

Additionally, digitalization further enables organizations to assess and

analyze multiple simulations in parallel by creating variations of original
simulation models, known as digital twins. These detailed simulations
provide new levels of operational excellence through enterprise-wide
insights that drive improved business and sustainability operations.
Artificial intelligence (AI) can provide additional sophistication to the
digital twins by increasing the level of prediction accuracy in the models
that enhance safety, reliability and intuitiveness of the processes.

2
Sustainability Initiatives Challenge Current Business Models
Three main levers drive sustainability In addition, the high energy intensity of many
initiatives for asset-intensive companies: chemical processes is forcing many chemical
Resource efficiency, circular economy and companies to engage directly in the energy
energy transition. transition, sometimes in partnership with
energy companies.
For many companies, the immediate focus is
Resource Efficiency: the efficient use of energy, Overall this shift is forcing a fundamental
water and feedstock in current operations. rethinking of current business models as well
Longer term, many companies are challenged as the tools and capabilities required to meet
by the unique demands of the circular economy, new objectives for success. The sheer volume
where economic activity is decoupled from the of information to be analyzed and need for
consumption of finite resources and waste is agility and scenario development create a
re-integrated into processes. unique challenge.

Sustainability Is Driving Significant Initiatives Across Industry Segments

Specialty Existential impact

Chemicals 36% 36% 27%

Significant impact driving

Energy 42% 26% 17% 10% 4% current inititatives

Engineering 30% 38% 25% 8% Significant impact driving future

investments

Chemicals 29% 21% 25% 18% 7% Low impact but growing

Polymer 28% 21% 38% 10% 3% Marginal impact •

3
Kickstart Sustainability Journey
Using Digital Twins
Digital twins provide valuable models that either connect to plant data The industry-leading digital twin from AspenTech is one such solution
for continuous monitoring or run offline to safely explore what if analyses that can help you meet these needs. With 40 years of process industry
without risking the plant performance. These simulations of actual experience, AspenTech is uniquely positioned to provide the entire
operations can provide companies with comprehensive solution that can solution building from chemicals simulation expertise to vertically
optimize asset performance across multiple dimensions of sustainability, integrated production and the end-to-end value chain.
safety and profitability through adaptive models, shared data and
advanced visualization. AspenTech Digital Twins Accelerate
There are a number of digital twin solutions available on the market, and
the Sustainability Journey
it can be difficult to know where to begin when choosing the one that’s
nsition
right for your needs. Here are three “needs” you might want to consider
y Tra
for your business before making any decisions: rg
s
ne New
in

Tw
Process

l
Validation

ita
•
• Do you need to plan and optimize operations

cy

Dig
throughout the value chain to build a robust Bio-based Carbon

ien
Capture

it a li z a ti o n
Feedstocks

ainability
sustainability program? ource Effic

rid M o d els •
• Is there a need for online/offline monitoring across Energy & KPI
Polymer Water Visibility
an entire asset for meeting sustainability targets or Efficiency
Product
testing new solutions? Specs
Flare
• Res

• Do you have the insights for enhanced collaboration Optimization

t
Emission
s

Dig

Hyb
to make informed decisions at every business level? Tracking
u
Plastics
S

Recycling Operator
ny

Training

L •
mo

Equipment

-M
o
Ec

Performance-

AI
lar Lifecycle
cu
Cir

Figure 2. AspenTech’s digital twin solutions to meet sustainability objectives. 4

SABIC, a global chemical manufacturer, in
conjunction with its partner company Kemya “Aspen Utilities Planner brings a wide range of visibility
(50-50 joint venture between SABIC and intoutilitiesoperationstoenablenewlevelsofsophistication,
ExxonMobil affiliate Exxcon Chemical Arabia) energy opportunity identification and improvement.”
used AspenTech’s digital twin solution to plan
its energy management system. ­ — Azzam Adel Homeida, Energy and Sustainability Engineer, Kemya

In 2009, SABIC launched a sustainability

program to reduce greenhouse gas emissions,
and energy and water consumption, by 25
percent and reduce the material loss by 50
percent by 2025. Over the last five years,
strategic initiatives have helped SABIC improve
emissions and plant efficiencies, leading to
10-15 percent improvement towards reaching
the sustainability initiatives.

SABIC developed utilities digital twin models

for energy management using the Aspen
Utilities PlannerTM solution to accelerate its
sustainability journey. The solution has helped
SABIC identify energy losses at the equipment
level and perform an overall utility system
optimization to maximize energy gain. The
utilities digital twin helped SABIC obtain a
comprehensive view of the utility system
and identify site opportunities for efficient
equipment selection and optimization of the
overall steam system. Since then, SABIC has
created utility models for 10 plants, obtaining
numerous benefits to maximize energy gains
and reduce GHG emissions. 5
SABIC’s Optimized Utility System
Illustration Scenario: U&O -1 U&O -2 Process Steam Import
and Export:

Figure 3. Illustration of Sabic’s Optimized Utility System.

Digitalization continues to play a vital role in Actual and Forcasted Energy Gain
SABIC’s ability to accelerate its sustainability
journey. Specifically, the Aspen Utilities
Planner solution helped SABIC gain visibility
on consumption patterns, enabling them
to regulate energy and utility losses at the
equipment level. During a 30-day period,
SABIC gained nearly 130 GJ/h for a single site,
equivalent to nearly 60,000 tons CO2 reduction
per year (assumes natural gas usage).1 Figure 4. Sabic’s actual and forecasted energy gains with Aspen Utilities Planner for a 30-day period.

6
Digital Twins Enable Key Insights for Sustainable Operations
Digital twin models can be used as an CO, SO2 and other pollutant emissions
alternative to sensor and testing approach to generated from a number of refinery sources.2
track emissions of CO2 and other pollutants.
The combination of process data and simulation BPCL’s model received an ASSOCHAM
tools such as Aspen HYSYS® and Aspen Plus® Innovators Excellence Award in 2020. In
provide the means to convert the information acknowledging the award, Murali Madhavan,
captured in the process databases into true Executive Director of the Kochi Refinery,
process knowledge. This enables companies specifically thanked AspenTech for guiding the
to mitigate the faulty analyzers and accurately company during the online implementation of
estimate the emissions of the whole plant, the model.
including the non-measurable emission points
ensuring plant-wide emissions compliance to
increasing regulations. System Architecture for Emissions Monitoring Digital Twin
Bharat Petroleum Corporation Limited (BPCL),
an Indian refining and petrochemical company,
developed its ‘Digital Twin Refinery Emission
Model’ for its Kochi Refinery. It improves
monitoring transparency and maintains
compliance under dynamic emissions rules, and
helps the operations team choose an optimum
fuel mix to reduce emissions while balancing
profitability. The team created steady state
emissions prediction models using Aspen HYSYS.
The models were validated and connected to
the data historian for online real-time data
collection, and then used to create customized
visualization dashboards to monitor CO2, NOx,

Figure 5. System architecture for BPCL’s emissions monitoring digital twin.

7
Digital dashboards provided enhanced visibility of sustainability metrics
for easy charting, thus highlighting progress on sustainability. This
reporting increased employee awareness of sustainability practices and
enabled better transparency to outside stakeholders.

Emissions Quick Access Digital Dashboard for BPCL Site

Figure 6. BPCL’s emissions quick access digital dashboard.

Digital Twins Accelerate Sustainability Efforts with Integrated Operations

Supply chain efficiencies gained through digitalization were highlighted “Digital technologies improve supply chain visibility, which offers the
as a key tool to improve sustainability progress in a recent report opportunity to improve scheduling, to eliminate activities that are not
published by the European Petrochemical Association. The report, adding value, and to optimize the use of the assets and the resources.” 3
‘Digitalisation as an Enabler for a Sustainable Future’, states that

8
FP Corporation (FPCO), Japan’s largest different units and disciplines that results
manufacturer of food containers, works in a truly circular solution. Digital twins
to provide a stable and responsive provide insights into these processes allowing
food distribution in an efficient and organizations to work collaboratively and
environmentally-friendly way. Its complex holistically to compare existing operations with
production and distribution network with future alternatives.
10,000 products demanded a detailed solution
to optimize transportation and inventory costs BASF Group successfully reduced CO2
while also integrating post-consumer plastic emissions from 40 metric tons in 1990 to 22
back into its operations. metric tons in 2018, while simultaneously
doubling its sales volume. The company has
Working with AspenTech partner Time further targeted carbon neutral growth by
Commerce, FPCO used aspenONE® Supply 2030 with its carbon management system
Chain Management (SCM) to reduce CO2 —including process optimization, energy
emissions by 160,000 MTs and waste by management, technology research for clean
443,000 MTs for fiscal year 2019. AspenTech fuel and purchase of renewable energy—to
and FPCO won the 2020 Supply and Demand reduce the company’s overall carbon footprint.
Chain Executives Green Supply Chain Award
for this effort in creating a sustainable,
recycling-oriented society.1

Digital Twins Enable

the Circular Economy
The circular economy approach not only
decreases waste production but also enables
organizations to evaluate new sustainable
operations without the additional cost of
raw materials. One of the biggest challenges
with this approach currently is ensuring
collaborative decision-making between
9
For one of the initiatives in its sustainability Absolute Greenhouse Gas (GHG) Emissions
program, BASF redesigned its methanol
Indexed (1990 = 100)
production process to cut CO 2 emissions by
160
recycling the carbon containing off-gases back Assumed
140 Global GHG emissions policy shift
into the process. Engineers used an offline
World
digital twin to evaluate the new application 120
36.4 billion t
for the redesign. The complex digital twin was 100
BASF business
developed using Aspen Plus to streamline as usual
80
the steps of the process and at the same time BASF
Carbon
60 40.1 million t
obtain a holistic view of the whole process. The management
BASF GHG emissions
model helped them save energy by avoiding 40 2030 target:
No increase
usage of fired heater, typical for methanol 20 of emissions
synthesis.
0
1990 1995 2000 2005 2010 2015 2020 2025 2030

Keys steps included methanol synthesis and Figure 7. BASF’s greenhouse gas emissions 1990 to 2030 (projected).

distillation, OxyFuel boiler, and BASFs patented

process for carbon capture and storage, OASE®
blue. Reinvention of processes and integration
BASFs CO Emission Free Methanol Production Process
of raw materials, products across these 2
CO2 Recycle
processes enabled significant productivity gains
low
and emissions reductions.1 pressure
additional residual O2-
flash gas
H2 Hydrogenation

PSA Oxygen
tail-gas
Inerts to
Syngas Fluegas OASE blue Atmosphere
Methanol OxyFuel
Synthesis dist. CO2 recovery
off-gas
Process Patent application filed
raw Methanol pure
Methanol Distillation Methanol

Methanol Synthesis and Distillation Process practically unchanged

Figure 8. BASF’s CO2 emission free methanol production process.

10
Digital twins can also be used for unit optimization and operational improvements. Hanwha Solutions
Corporation has a corporate initiative to build a smart factory to become more competitive by maximizing
profits and minimizing risks while achieving its carbon neutrality objectives. This multinational energy
services, petrochemical, and real estate development company headquartered in Seoul, South Korea, was
working towards implementing a solution in its processes that can help operations, safety, environment,
and training to help prepare operators to handle actual scenarios and potentially dangerous situations
without directly experiencing them.

Hanwha identified the benefit of a digital twin solution to help with its challenges. The digital twin of the
hydrocarbon resin manufacture process was modeled with Aspen Plus and Aspen Plus Dynamics. The
models were used to simulate both steady state and dynamic conditions to improve operations through
modeling transitions, catalyst deactivation and product quality improvement strategies.

Hanwha applied the Aspen Plus-based digital twin model to various applications, including one to
optimize solvent recovery tower in the polymerization section and run feasibility studies of solvent
selection for the hydrogenation section of the process. The use of digital twin model offered insights into
solvent management and enabled the reduction of solvent waste by 29 percent in four months and the
new solvent offered savings of $0.5 million per year for the hydrocarbon resin manufacture process.1

11
Hanwha’s Target Process for Digital Twins

Figure 9. Hanwha’s target process for digital twins.

Hanwha plans to further improve the models integrating domain knowledge and engineering
by applying AI, specifically machine learning, constraints to build an enriched model without
to key product quality indicators. Machine requiring the user to have deep process or
learning leverages simulation or plant data, AI expertise.

12
Hybrid Models Boost Digital Operations for Sustainability
One of the leading challenges with digitalization is the need to shift new processes that have less
impact on the natural environment. The enhanced accuracy gained with hybrid models can help
companies make faster progress toward these new requirements. Hybrid models combine AI and
first principles concepts to deliver a comprehensive, accurate models more quickly. The accuracy of
empirical models and the strength of first principles models, leveraging the power of AI along with
domain expertise, creates a more predictive model. AI can also be a key enabler to run multiple
digital twins, required for site-wide optimization and monitoring.

13
Nissan Chemical Corporation needed to reduce costs in its ammonia first principles knowledge and machine learning from plant data, using
process. Beginning with the steam methane reformer step, engineers a neural network to perform the calculations. The data was imported to
were challenged by the furnace temperature distribution, limiting the Aspen Plus to create the neural network after selection of dependent and
application of a rigorous first-principles simulation. The Japanese independent variables.
company decided to apply the hybrid model approach by combining

First Principles-Driven Hybrid Model

Natural gas Plug Flow Reactor

Steam

Recycle H 2 T, P Reaction rate constants

1st Principles model knowledge
(Process engineering)
Calculator block
(Neural Network)
Flows, compositions

Machine learning from data

for unknown phenomena

Figure 10. Nissan’s ammonia processing hybrid model schematic.

Using the hybrid model, the Nissan team was able to fit the kinetic The model accuracy optimized the steam input and reduced the steam
parameters and reproduce real plant data more accurately than the consumption by one percent.
conventional reformer model. This offline digital twin is a highly accurate
model was created in half the time of the conventional model.1
14
Better Fitting Data With First Principles-Based Hybrid Models

Conventional Model Conventional Model

Hybrid Model Hybrid Model
Predicted

Predicted
Observed Observed

.Figure 11. Conventional models vs hybrid models.

“Usingthe[AspenTech]hybridmodel,wewereabletocreateamodelthatcanreproducereal
plantdatamoreaccuratelythantheconventionalreformermodel.Wewereabletocreateahighly
accurate model in a short period of time.”
—Mr. Takuto Nakai, Production Department, Nissan Chemical Corp.
15
Conclusion
Global efforts to deliver products through processes that have less impact on the
environment are driving a strategic shift in business metrics and practices. Digital
twins are an important tool in this process, enabling improved process design,
greater manufacturing insight, and better operational integrity. Many forward-looking
companies have already begun this process, investing to build new capabilities, and
developing innovative technologies and business models to achieve new targets.

The integration of sustainability targets with business goals will be transformational

for chemical companies, especially as these topics are being raised in shareholder
meetings, by large institutional investors and by external stakeholders. Furthermore,
the target-oriented approach also positions businesses toward renewed growth and
long-term business success. The Ellen MacArthur Foundation highlights that products
developed for the circular economy offer as much as $1 trillion in new business
opportunity for companies that meet these objectives.4

Achieving the fragile balance of sustainability goals—equally considering people,

planet, and profit— is a considerable challenge for organizations, but one that must
be addressed to be competitive in the energy and chemical markets of today and
tomorrow. Digital technologies and Industrial AI are critical capabilities that will
separate market leaders from niche players of the future.

Citations
1
Originally presented by the customer as part of OPTIMIZE ‘21 conference, held in May 2021. Contact your AspenTech representative for more details.
2
www.aspentech.com/en/resources/white-papers/digital-emission-and-efficiency-monitoring-through-modeling-improves-sustainability-at-bpcl
3
Report, Digital Cafe Workshop, EPCA 53rd Annual Meeting, Berlin, October 2019, ‘Digitalisation as an enabler for a sustainable future’
4
The Circular Design Guide, The Ellen MacArthur Foundation 16
About AspenTechnology
Aspen Technology (AspenTech) is a global leader in asset optimization software.
Its solutions address complex, industrial environments where it is critical to optimize the asset
design, operation and maintenance lifecycle. AspenTech uniquely combines decades of process
modeling expertise with artificial intelligence. Its purpose-built software platform automates
knowledge work and builds sustainable competitive advantage by delivering high returns over the
entire asset lifecycle. As a result, companies in capital-intensive industries can maximize uptime
and push the limits of performance, running their assets safer, greener, longer and faster.

aspentech.com

© 2021 Aspen Technology, Inc. All rights reserved. AT-07361