Cathy Reese, Partner, IBM Advanced Analytics Global Practice Leader, discusses continuous intelligence in the chemical industry, including how one chemical company is transforming customer service by analyzing historical data for context and streaming data to make up-to-the-minute recommendations to their customers.
Adrian Bowles: We’re at IBM today with Cathy Reese, a Partner in Advanced Analytics in your Services group. I’m going to talk about advanced analytics, and what you’re doing, what you’re seeing with your clients. So maybe tell me a little bit about the practice itself. What do you cover? What areas?
Cathy Reese: Sure. We have a practice of consultants and we cover data science, data engineering.
That’s really our purpose and our mission is to help our clients really move forward their agendas in AI and data. So we are the premium services integrator for data and AI.
AB: Okay. And I know that you’ve been with IBM for about 15 years.
CR: Yes.
AB: Obviously, you weren’t doing this 15 years ago. Tell me how you sort of migrated into this role.
CR: Sure. I started building data warehouses for Basel compliance. Projects back then, in the ETL days, used to take four years to build one data warehouse so by the time you finished, requirements were totally different than when you started. But as technology has evolved and as AI and machine learning have come up to speed, our projects are now shorter and delivering value faster to our clients.
AB: Okay, great. Maybe you can talk about one industry or one example where you’re working with a customer.
CR: Sure. We’re working with a chemical company right now, and they’re really trying to be a digital-driven company and customer service is their number one thing that they want to focus on. So in order to serve their customers in this new digital world, we’re helping establish a flexible architecture for them that we can add on use cases and add on pieces.
If there’s a defect that comes in on these chemicals, they’re able to think about who else could be impacted by that defect. So they’re going to look at the weather data for that day, they’re going to look at the sales data, they’re going to look at the actual defect data, and they’re going to run all this analysis to be able to get alerts to those clients who could possibly be affected by that defect as quickly as possible.
AB: So you have a lot of different data sources using historical data, anything real time or streaming as it’s coming in?
CR: For that particular use case, the streaming data is the kind of quality you’ve got coming in, because think about when you have a defect in a chemical, that is something that you can’t just wait for the data to batch at the end of the day or the next day. It’s something that needs to be processed as quickly as possible to be able to get impacted customers aware right away.
AB: What stage is that project in now? Is that something where you’re working on it now or-
CR: Yeah, we’ve just finished doing the roadmap for them. We’ve identified, as I said, all those different value-driven use cases. It’s a series of eight use cases. Each of those are broken down into which ones are the quick wins that we need to work on first, and the key to everything we’re doing with them is to make sure that we have a flexible architecture. So we’re using IBM Cloud Pak for Data and we can add on and turn on the pieces of that architecture as they’re needed for each of those use cases.
AB: Using Cloud Pak, when you say that you can turn on the different pieces, I know that’s a very full-featured system and you could probably have a number of applications only use a small part of it, but you’re creating it and doing the architecture so that you can grow as you get those quick wins.
CR: Yes.
AB: Okay.
CR: We have it mapped out in three phases, but IBM Cloud Pak for Data is allowing us to map out phase one of the architecture, phase two, and phase three, and the client is comfortable knowing how all those features can turn on and turn off and come together as one package, and how everything is connected in together, and the fact that a lot of what is in IBM Cloud Pak for Data uses open source and uses newer technologies that keeps them excited and they’re not as worried about vendor lock-in.
Use the follow links to learn more about IBM Cloud Pak and Continuous Intelligence
