Ron Beck is the Director of Industry Marketing, Energy, at AspenTech. During the past decade, Beck has been responsible for marketing the Engineering & Construction sector, the Engineering product portfolio, Aspen Economic Evaluation and Aspen Basic Engineering software. He is also a company spokesperson for his subject matter areas. Beck has well over two decades of experience in evangelizing software solutions to process industries and over a decade of experience in commercializing chemical engineering technology. He has authored papers on key industry topics and had presented at several public industry events. Beck is a graduate of Princeton University in New Jersey in the US.
As corporations adapt to the “new normal” of extreme volatility, uncertainty, complexity and ambiguity (VUCA), many are turning to Industry 4.0 technologies and AI capabilities to automate fundamental industrial processes. Automate Asia spoke to Ron Beck, Marketing Strategy Director on how Industrial AI could give corporations greater resilience, flexibility and agility—and with that, the ability to respond to shifting market conditions and thrive even in times of uncertainty.
1. AspenTech serves as a solution provider that focuses on Artificial Intelligence (AI) and machine learning to oil and gas, chemical, and engineering companies. Could you briefly elaborate on this? What is AspenTech’s key to success since its establishment?
The key to AspenTech’s success over the past 39 years has been to focus on the industry’s challenges and value levers, coupled with strong domain expertise and engineering knowledge. AspenTech recently pioneered a new approach to AI and machine learning called ‘Industrial AI’. The approach combines the basic digital models which make current oil and chemical assets operate safely, which are known as first-principle models, with advanced analytics models (or AI machine learning), and industry domain expertise.
By combining these three models, practical solutions can be created to break new ground in the ability to model hard problems, predictive insight into what will happen, and intuitive usability to make these advanced solutions accessible to regular industry workers. Powerful process simulation, multivariate process control, and production planning are innovations we have delivered to industry over the past 39 years, and which 2,300 companies worldwide are using to create value. The R&D power behind these models has always involved strong collaboration between AspenTech, customer domain experts, and academia innovators with more than 2,000 man-years of chemical engineering and domain expertise
Process simulation models, introduced by AspenTech beginning in 1981, created the foundation for rapid growth and innovation in chemicals and oil and gas over the past 30 years. These included the ability to introduce new products and processes quickly, to scale production to meet the demands of the rapidly growing Southeast Asian middle class and economy; as well as to support safety and energy conservation in increasingly complex operations. To take the next leap, we have added some of the best AI and data science minds to our teams. The combination of chemical engineering and data science knowledge is creating exciting practical, Industrial AI results.
Over the past four years, we have introduced exciting Industrial AI technologies. The first was Aspen Mtell, which can predict equipment degradation up to 60 days in advance. Last month, we introduced Aspen Hybrid Models, which combines our award-winning Aspen HYSYS modeling system with machine learning; and similarly, we’ve just introduced new digital solutions based on our PIMS-AO planning modeling system and our DMC3 advanced process control system. We see tremendous potential to apply these innovations to sustainable oil & gas and chemical industrial companies.
2. How can Malaysian firms assess their industrial AI readiness and examine a framework for organizational preparedness? What are the critical factors for Industrial AI readiness?
To put it simply, all companies need to move quickly to begin to adopt elements of AI today, or what I have referred to above as Industrial AI. Malaysian companies have most of the elements in place to begin adopting AI solutions that will create value. But first, let me correct a few common misconceptions. A company does not need an army of data scientists to succeed in this endeavor. Nor does a company need very large quantities of data. The best AI-based solutions are designed to be used by today’s workers, operators, engineers, planners, traders, maintenance people, and managers to support strategic decision-making and agility.
These solutions will employ powerful analytical engines on the inside, but provide an intuitive workflow and interface on the outside to support the manner in which decisions are made and work is performed. Malaysian firms need two things for a successful industrial AI strategy. First is company leadership that understands and embraces that digital “disruption” (in the form of industrial AI solutions brought into their business mix). Such disruption must be vigorously pursued to be resilient and emerge as a leader in the current economic environment. Second is the selection of team leadership for this transformation who both understands the business and technology well.
Once those two things are in place, do ensure there are two ingredients of implementation and organizational readiness. The first is to develop a plan that will map implementation of AI against areas of the business challenge, or initiative, or need. The second is to understand the need to make the organization flatter, more collaborative, and more agile.
3. A major challenge for most industrial organizations is not a lack of data, but a lack of accessible and useful data for industrial AI. How can companies effectively address this key gap?
A key element of useful Industrial AI solutions is adding intelligence to the system to assist in the organization, aggregation, selection of useful data. I would argue that this needs to be thought of in reverse. First, business challenges need to be addressed and then understand which data is needed for insights relative to those challenges. This then will value different data as to whether it helps provide insight in solving those challenges. Hence which data areas are most productive to invest in.
4. What defines a ‘SelfOptimizing Plant’? How does it benefit companies to drive maximum profits, minimize environmental impact, and ensure greater reliability and efficiency?
The Self-Optimizing Plant is AspenTech’s strategy for taking customers on a journey to the future intelligent plant, asset, or set of assets. Our vision for the Self-Optimizing Plant (SOP) is a plant that is self-learning, self-adapting, and self-sustaining. By self-learning, we mean that the plant, as it operates, learns from each action it takes, s from the data streams reporting on the plant and from the digital twins providing operating insights.
Therefore, it improves its ability to reach its potential and even set the potential higher based on its learning. By self-adapting, we mean that the plant will continually adjust to changes in the condition of the asset itself, as well as to external factors, to change the objectives of the operation continually. By self-sustaining, we mean that the plant will intelligently monitor the health of its equipment, processes, and systems, based on data streams and insights from those data streams. It will then take corrective actions to ensure the integrity of the asset, and the health of the equipment, to avoid or minimize degradation, and to avoid missing customer targets.
Our target is to reach a capability for the asset to be self-sustaining. Some aspects of operations may become autonomous in the relatively short term. But broader autonomy is a longer-term goal. That choice of words is a conscious strategy decision on AspenTech’s part. Typically, oil and chemical assets are too complex to be able to run completely autonomously, at least within the next five to ten years. Instead, we are driving toward enabling a self-sustaining plant. With respect to sustainability and environmental impacts, these capabilities will be crucial to navigate the complicated technical, operational, and business trade-offs required to make energy and chemical assets move towards zero carbon and reduce levels of waste production and water usage.
Most companies in the chemical and energy fields are beginning to set ambitious sustainability targets, to ensure that they contribute to future sustainability and maintain access to capital. Achieving progress in areas such as carbon neutrality, a circular plastics economy, and water conservation, are complex challenges. It is a complex optimization challenge that requires looking at a company’s assets, its value chain, and its optionality. The self-optimizing plant will be critical to achieving these goals
5. What are AspenTech’s expansion plans, priority growth areas, and market outlook for the next 12 months in Malaysia and Asia Pacific region?
Beyond the current economic disruptions globally, all projections still show strong GDP growth across Southeast Asia over the next 10 years. Our level of engagement with industries and companies in Southeast Asia continues to grow. In fact, the level of technical talent in Malaysia, Thailand, Indonesia, and Vietnam is impressive. So, while the current climate remains to be turbulent, we are optimistic as to the opportunities and growth in the region. We are prepared to support the growth as it happens. We have already demonstrated our willingness and intent to put in place local talent as the opportunity presents itself.
6. Do you have any key advice for organizations that are exploring Industrial AI?
Industrial AI, data science, and analytics are key areas that are moving very quickly from a technology and solution viewpoint. Companies must select strong partners with the capability and strategy to “futureproof ” their solutions. It is important to work with a technology supplier who has a clear strategy and vision. Also, implement a technology architecture that can adapt rapidly to changes in the availability of the technology building blocks.
Create an AI plan that maps to your business strategies, key business challenges, and value levers. Industrial AI is already a prove