Hybrid models are a game changer in process engineering , says Sanjeev Mullick, VP – Sales, Asia Pacific and Japan, AspenTech

New process models and Machine learning technologies are a major step forward in transforming the plants operations aspenONE V12

  • November 21, 2020
Sanjeev Mullick, Vice President - Sales, Asia Pacific and Japan, Aspen Technology spoke exclusively to Pravin Prashant, Editor, Indian Chemical News on Asset Optimization Software market, USP of aspenONE V12, level of customization with respect to Version 12 and India market strategy for Asset Optimization Software. Excerpts of the interview: 
Globally, how big is the market of asset optimization software and how big is it in India? What is the growth in asset optimization software?
The world global manufacturing capacity is about $14 trillion and it is estimated by the National Association of Manufacturers in the US that 10% of this global manufacturing capacity is lost due to unplanned downtime and other factors which is about $1.4 trillion. These modern software solutions, aid manufacturing and supply chains, they really have the potential of helping the world industry recover as much as that $1.4 trillion. In the end, you may not recover all of it but that gives you the size of the price that is out there.
Now, as far as Aspen Technology is concerned, we estimate that every year we are actually creating about $50 billion worth of value for customers through the use of our solutions. We believe that with greater digitalization, we could easily take it to much higher levels and some estimates are that this will add up to $200 billion worth of incremental value for our customers.
Out of $1.4 trillion that you are talking about how big is Asia Pacific? 
The $1.4 trillion total available market size is estimated by the National Association of Manufacturers in the US, so we wouldn’t have the exact breakdown on hand, for the Asia Pacific market. However, even within Asia Pacific, some countries may be better in terms of managing their assets and having better maintenance and less downtime. Others might still have more to improve. So, it will be somewhere in that plus minus 10% range.
You are calling it AOS (Asset Optimization Software). Nowadays, people say platform instead of software. Is it more of an Asset Optimization Platform or Asset Optimization Software? 
Yeah, that is a good point. If you look at all the software solutions we provide, the Industrial AI solutions, they are backed by a very, very thorough technology stack, which is based on the latest IT platforms. So yes, behind the scenes, we offer a complete platform. We have built our solutions on top of it for asset optimization, hence, we are actually a complete system. 
As I mentioned earlier, we are independent of any particular operating or hardware systems thus in that way, we are very agnostic in terms of sitting on top of whatever infrastructure our customers might have. So, you can say we are both a platform as well as the software. If the customer already has a certain platform, and they've invested in that we can sit on top of it or if the customer wants our complete platform and our solutions, we can bring that in as well.
You talked about customers already utilizing your different solutions can upgrade by going for aspenONE V12. But for those customers who have not yet deployed your product, how can they go about deploying asset optimization software platforms?
Worldwide, we have 25 plus offices to be closer to our customers. In terms of the existing customer, if they want to try out a new product out of our suite, we actually engage very deeply with our customers to understand the business problems and in a very consultative way, we help them understand how they can architect our solutions into their infrastructure and what kind of value they can get. But it all starts with understanding their priorities, their initiatives, their business problems, their manufacturing challenges. We can combine all the different technologies; we have to come up with a complete solution for them. We provide implementation services, as well as we have a whole network of partners there in India as well and other parts of Asia Pacific, where they help our customers implement and maintain and sustain these applications. So that's how I would say we go about helping our customers acquire our solutions and be successful with it. 
What level of customization can be done in the new version? 
I would like to use a different word. I don't like to use the word customization because what we offer is with each release, it's a packaged solution. We like to describe it as configuration, we configure our solutions to a customer's unique business needs. What that means is our solutions are very flexible. They can be configured, and we don't ever have to go back to the source code to do any customization.
So, we're not creating bespoke solutions for our customers but we give them standard solutions and what the benefit of that is, when the next release comes out, it's much easier for them to upgrade. You asked me when can customers in India use Version 12 and I said today, that's because if they're using a standard version, then upgrading them is a very straightforward path because we ensure that from one version to another, our customers can seamlessly upgrade.
Would you talk about some of the new features incorporated in aspenONE V12?  
Okay, so some of the new features that are really taking the world by storm is an area called Hybrid Modelling. One of the ways that I should also point out to you, the way AspenTech has been developing the solutions in the  past few years is, we have advisory boards, customer advisory boards, we have also something called the innovation club, where leading customers from around the world work with us to do all kinds of pre-release work, whether it is doing usability testing, whether it's doing beta testing. In the case of Hybrid Modelling, we had over 100 companies participate, it received a very enthusiastic response from all over the world, including India and other parts of Asia Pacific. The result was that they came up with all kinds of very interesting use cases, because AspenTech has always taken pride in the fact that we do a lot of first principles modelling, which is based on the physics and the chemistry and the engineering fundamentals and 40 years of domain expertise. 
There is always a class of business problems that is hard to model using first principles. A case in point would be, how do you model the colour of a chemical compound, depending upon some impurities or some other processing steps, the colour might be different from one batch to another, or from one set of conditions to another. However, using data-based models, and using Machine Learning (ML), you can model these kinds of unique attributes. What we have done today is that you can take all of those and develop an Artificial Intelligence (AI) model, and then embed that into your first principles simulator like Aspen Plus or Aspen HYSYS. So, in that form, we are enabling our customers to solve all sorts of new types of problems that in the past have escaped attention or have been hard to solve.
Hybrid models are a major step forward in bringing together AspenTech’s process models and ML and are a game changer in process engineering and plant improvement. 
We have also come out with many other features, for example, there is a concept called multi-case, where you can use a simulator to run thousands of runs and use them in ways to give you additional insights. We are also coming out with advances in cognitive guidance in planning, with this release we have already come out so customers can basically verify the validity of their plans for the refinery or the petrochemical plants that they have. 
So, these are some of the new advances that we have come out with which are very, very unique, and nobody else in the world has offered any solution of this kind which is why the customers said it's a game changer.
Does the solution cater to brownfield or is it only restricted to greenfield operations?
We serve the existing assets and the new assets, upstream refining petrochemicals further downstream, all the way to pharmaceuticals, which is now the need of the hour. You know, we have a lot of pharmaceutical companies who are using our software and we're beginning to see more and more uptake for that in places like India as well. So, we cover the whole range - Brownfield and Greenfield.
You are using almost a majority of in-depth technologies, which is artificial intelligence, high performance computing, cloud computing, 4.0, IoT and machine learning. So how do you convert these technologies to reduce the losses?  
One of the biggest drains of the industry is unplanned downtime. So, a piece of equipment or critical equipment has some kind of a degradation and it shuts down in an unannounced unplanned fashion. And it has two major impacts to the industry. One is, of course, when the machine is down, there is lost production. So that's a big financial impact, it hurts your supply chain, you may have orders to fulfil, but now you are not able to manufacture your product to fulfil that order. And the other impact is you have to, under the circumstances of unplanned downtime do emergency repairs, which are always more expensive than planned maintenance activities. So, in that fashion with our software, we are able to actually help our customers minimize or avoid downtime, because using machine learning algorithms, we pick up these early signs of future failures and we are able to warn or alert the customers.
They can actually then organize when they want to take that piece of equipment down for maintenance and avoid the supply chain shock that it creates, avoid the loss of production, they can inform the people who are scheduling the facility how to work around that constraint that may be emerging in the next few days. 
The other example and I will elaborate is - if you look at advanced process control, we are helping our customers run their plants in a more stable fashion, produce products of quality with the correct yields and minimize emissions. In fact, we have customers in India, BPCL for example.
Sox emissions, and their success story has been published and they have been able to reduce their Sulphur Sox emissions by a very significant amount. For example, at BPCL Kochi Refinery with the introduction of dynamic limits regulated, it is necessary to move emission limits close to the operating value of units. The environmental team created a refinery-wide emission model via the use of Aspen HYSYS; Aspen Online; Aspen InfoPlus.21 and aspenONE V12 Process Explorer. A key outcome includes improving margins by reducing emissions. For example, in the refining industry based on typical energy consumption, a 1% improvement in thermal efficiency translates into energy savings of $600,000 per year.
So those are the kinds of things we are helping our customers, be compliant, be better stewards of the environment, if you will, while actually increasing the profitability.
How many man hours did it take for aspenONE V12 software to become commercial?
I don't know whether I can put a dollar number or man hours to it. All I can say is, we have a very large R&D organization and a significant percentage of our annual revenues in double digits is rolled back into R&D. So, I would say it's in millions of dollars of annual R&D expense with hundreds of very specialized scientists and engineers with very high qualifications working on this, all domain experts in their fields. 
So, yes, it is a significant effort, and we are very proud of what we do for the process industry worldwide, the chemical industry including those in India. 
Do you also do some software development in India or is it completely US based software development that you do?
Our software development centres are in the US but what we do out of India is provide a lot of implementation services and support services to our customers worldwide.
And how big is the team in India?
It ranges from 35 to 40 people and includes support services, solution consultants and sales team.
With respect to the Asia Pacific and the Indian market, how do you look at the India market post COVID-19 with respect to asset optimization software? 
I think the COVID-19 pandemic has been an eye opener for pretty much everybody in the world. And one of the things is that companies have already embarked on their digitalization journeys and have fared better during the pandemic. Work from home is an example - companies that had already started using cloud computing, had already enabled people to work in distributed teams, they were actually able to, with very little hiccup, transition to work from home. 
Another example, I would like to quote is companies that had started looking at more modern ways of planning and scheduling their facilities - what they found was that when global supply chains were disrupted, for example, supply of a certain raw material that you might be getting from another part of the world was disrupted because that manufacturing facility or that port was no longer functioning. That whole concept of ‘just in time’ manufacturing was upset. So, customers who could quickly reposition themselves in terms of where to source the raw materials from how to schedule that distribution, or raw material acquisition as well as scheduling the plants, they actually fared better.
So, post the pandemic I think the lesson learned by companies in India and the rest of the world is that they will be better off and better prepared for future events such as this by embracing more digitalization and more automation.

Register Now to Attend Navigating the Path to Net-zero in the Chemicals Industry on Friday, 26 July 2024