3 minute read 20 Jun 2018
A man in a lab

Why the chemical sector should harness the power of machine learning


EY Americas

Multidisciplinary professional services organization

3 minute read 20 Jun 2018

This tool helps companies make critical decisions more quickly and gain deeper insights, including in ingredient quality.

What do you need to be a pioneer in the chemicals sector? For Ravi Joshi, an executive director in our Advisory service line, the answer lies in machine learning and predictive analytics, and they can be applied in ways you might not have considered.

One example: ingredient quality.

“In the chemical industry, one company’s finished product might be a raw material for another company,” Ravi said. “Any deviation in quality is going to have a significant impact on the supply chain. Machine learning can drive ingredient quality not just for that particular industry, but across the supply chain and throughout the life cycle of the product.”

In the near future, machines will be able to predict and then tell you in advance what decisions need to be made.
Ravi Joshi
Executive director in Advisory service line

That’s a process that may sound time-consuming — and it is, if you’re doing it in a low-tech way.

“If a set of quality results does not meet standards, you can build algorithms and start doing repeatability studies to predict when things might go wrong,” Ravi said. “Machine learning can be used for that. If you try doing the process analysis manually, it is going to take a lot of time. Machine learning algorithms can get that information quickly and minimize bad product quality by allowing leaders to be proactive and take corrective actions.”

And then there’s one perk that all sectors can seize upon: putting data into context right when it’s needed. “There is a lot of information needed right at that moment to make key decisions,” Ravi said. “It is not like I can make the decision after one month, or one week, or one day. Having people try to get that information right away takes a significant amount of time. That is when machine learning is going to come in handy. People can get to decisions rapidly using the information.”

That capability is crucial for companies as they move from data storage to data insights. 

“The tools are there, but it is a significant leap from just having the tools and having data analyzed,” Ravi said. “I want to see the end result of taking the leap to build those algorithms that drive the speed at which decisions are made. I think it is really important for organizations to be leaders in analyzing the data using the tools of machine learning.”

This plays into the broader topic of technology transformation in the section, and how machine learning can be one tool to drive it. 

“I have been in situations where I have used my mobile device and said, ‘Hey, we can do this one,’” Ravi said. “People say: ‘You are lightyears away. We do not have that technology; we aren’t there.’ Then, within a couple years, when we go back and talk to them again, they say, ‘Oh, we are either there or behind the industry.’ Especially in the chemical industry, there are a lot of applications and a lot of data that are all the way there.”

“In the near future,” Ravi said, “machines will be able to predict and then tell you in advance what decisions need to be made. In some cases, the decisions will be made by the machines, not by people taking a lot of time to review the data. When information is automatically processed and presented to you in the format you want, it becomes easy for you to make rapid decisions and be a leader in the sector.”


Machine learning can be an important component in how you gain a competitive edge and equip your company with the tools of the future as part of a broader transformation.

About this article


EY Americas

Multidisciplinary professional services organization