Driving Decarbonization with Digital Tech

Jan. 25, 2024
When process modeling and data analytics techniques incorporate the cost of CO2 emissions, companies can optimize operations for both profitability and sustainability.

Operators in the processing and manufacturing industries constantly seek to improve efficiency and  minimize energy use by optimizing existing operations. Now, many of these actions must also align with  sustainability goals to reduce a facility’s carbon footprint. 

Fortunately, it’s possible to seek decarbonization while continuing to reach traditional operations  optimization targets. Operations can be analyzed for a production line, at the plant level and even  throughout the value chain. This requires important digital capabilities but results in the ability for users  to reduce emissions while improving margins.

Whether companies are seeking the optimal operating setpoints for a specific asset, the best running  conditions for a plant, or other business opportunities, the answers can be non-intuitive and difficult to  obtain using traditional methods. That’s why companies need the tools to implement data gathering,  analytics and digital twins to drive optimizations that align business and decarbonization goals.

Road to digital decarbonization

Although existing operations are likely to be digitalized to some extent, the most common applications  have been focused on automation and safety. Applying digital transformation for efficiency gains and  decarbonization can build upon these existing efforts, but additional data and analysis is generally  required.

A priority here is to establish a fundamental platform that can connect to all sources of information to  provide immediate access to high-fidelity historical, real-time and predictive data. From this starting  point, hierarchies can be applied that reflect operating and reporting environments with built-in  analytics to support deeper analysis.

Once a solid strategy for gathering and sharing data is in place, simple calculations and basic analytics  configured on the operational data management system can help workers to monitor key performance  indicators (KPIs) and quickly identify areas of improvement to increase efficiency and reduce  greenhouse gas (GHG) emissions. 

An important aspect of any decarbonization strategy is to ensure maximum asset reliability and safety.  Optimizing the asset maintenance strategy prevents excessive fuel consumption as well as accidents  that could lead to undesired emissions. Using historical data, it is possible to apply machine learning  (ML) and artificial intelligence (AI) techniques to train models capable of predicting abnormal asset  behavior. Predictive analytics applications provide early warnings that enable the operations team to  plan corrective actions and allocate the best possible resources for corrective actions. The most  advanced predictive analytics solutions combine ML with rigorous modeling, increasing the accuracy and  early-response of anomaly identification.

Another approach is to use advanced models to recommend adjustments to the control system set  points or optimize the production planning strategy. Advanced models can handle several aspects such  as operational assets, process behavior, resource consumption and economic factors. 

The goal is to investigate various scenarios to determine the optimal operation of scopes 1, 2 and 3  emissions: minimizing operational CO2 emissions, maximizing clean energy use and better managing the  value chain (see chart).

By building emissions into the modeling, a unified and carbon-aware strategy can be developed for  running equipment, production plants, and the enterprise.

Implementing carbon-aware optimization

One of the world’s largest auto manufacturers chose to pursue green goals by reducing energy  consumption at its production plants, thereby also reducing CO2 emissions. However, there was no  standard for data monitoring capabilities and, in some cases, data was still being processed in a manual,  paper-based method at the plants. 

The team implemented a unified data model at each facility capable of connecting process and energy  data to create a centralized energy monitoring system. With this data model they were able to easily  build an asset framework for monitoring and contextualizing available data allowing them to analyze  gas, electricity and compressed air use. 

Through the combination of real-time data with reporting and visualization tools, the team can now  quickly evaluate different electricity usage scenarios, and even reveal the impact of weather on  operations to choose the right energy strategy and power source selection for reducing overall energy  use.

In another case, a large international energy company wanted to use advanced cloud computing to  better manage its supply chain—including provisions for tracking and managing CO2 emissions. They  wanted the ability to quickly evaluate strategies for improving margins and reducing carbon emissions at  company facilities worldwide.

Using a unified approach for production planning in the cloud, they were able to realize fast data access  and use optimization models to simplify and improve supply chain management. They also reduced  decision time from two days to less than two hours and rapidly deployed new CO2 modeling functions  for applications across their plants worldwide. 

Standardizing this structure enabled meaningful comparisons and shared insights among multiple  facilities, supporting the company’s sustainability goals.

Decarbonizing the future

By following and extending the data-driven approaches mentioned here, companies can reduce their  entire carbon footprint throughout the value chain. This can include monitoring and managing business  aspects, such as carbon offsets, carbon credits and carbon negative options.

These examples demonstrate the ability of digital twins and associated technologies to help  manufacturers improve both profitability and sustainability. In some cases, this will be achieved by  tuning equipment and processes to reduce CO2 emissions or maximize clean energy use. In other cases,  advanced analytics may identify specific facility upgrades or other changes to achieve the desired  results. In all situations, a robust data management approach with easy information access and sharing  is the most reliable way to reveal the best path forward.

Fernanda Martins is the industry marketing director at Aveva and a chemical engineer from the  University of São Paulo. She has more than two decades of engineering and industrial software  experience. Today her work supports Aveva´s strategy in developing industrial software to build a  sustainable connected future.