Talking Supply Chain With Stemcell Technologies

Hello and welcome to the Flowcast, a podcast from Demand Driven Technologies where we chat with industry leaders about how to build more responsive and reliable supply chains. I’m Leah Hoffmann and I’m here today with Juan Abbud of Stemcell Technologies. Welcome, Juan, and can you say a few words about yourself and your role at the company?

My name is Juan Abbud. I work for STEMCELL Technologies. I’m a Senior Business Architect for Demand Driven Operation, and my focus is on finding opportunities for optimization in supply chain and helping us be more adaptive, more resilient, and better serve our customers.

Can you tell me a little bit about STEMCELL? I understand your niche in the life science industry is with cell culture media and cell separation products?

The mission and focus of the company is providing reagents for experiments in the life science space. So we have a variety of product lines that support the different research workloads from our customers. There’s a cell separation product line. There’s a media and cell culture product line. There’s an instrumentation product line. There’s education offerings. There are fresh tissue offerings, as well. We’re sort of the foundational level of the life sciences industry.

Let’s set the stage by talking about why your supply chain is difficult from a materials planning perspective. You’ve got long lead times, I would imagine, raw materials that are pretty complex…

I think some of the things that make our supply chain slightly different than other industries is that we have deep bills of materials with cumulative lead times that can be very long. And there are certain levels of biological variability in many of the components that add uncertainty to the supply. So even when you get the material, it may not work as expected. You need a really good understanding of where the risks are to the continuity of your supply and develop strategies to manage against that—and basically make your investments as efficient as possible so that you can have good service at a reasonable investment level.

How did the pandemic affect those issues, and how did you react?

One of the other things that I didn’t mention earlier is that a lot of our products have shelf life. That also adds complexity: you have to make sure that the product is fresh. In terms of the pandemic, we didn’t really start seeing an effect until all the lockdowns started to take place. Once people started working from home and couldn’t go to the labs to do their research, we saw a drop in demand. That started to happen around maybe mid-March or April, and we needed to respond very quickly.

And the question that came to us was how can we contract so that we’re not producing things that are going to expire and consuming raw materials that may be difficult to get as we continue to get into the pandemic. At the same time, we need to be able to set ourselves up to be responsive once demand begins to pick up. 

I’m a little surprised to hear that there was a drop in demand—I’ve read that your cell culture medium was used by Chinese researchers to isolate the Coronavirus, and once the magnitude of the threat became clear, it was also pretty clear that the scientific world would have to devote its attention to finding better treatments and vaccines.

If you think about our product lines, we support all the different tissues in the human body, right? And there’s such a wide variety of products. Most of the research that we support is focused on immunology and cancer. The vaccine-related research is a small piece of the pie. So as people were no longer able to access the lab space and continue doing their research, and universities began to close campuses, all these things started to have an effect. 

So how did you go about reacting to the disruption and where did DDMRP come in?

One of the things that we really appreciated about DDMRP was the fact that, since most of our parts are buffered, if demand stops, by default, DDMRP is not going to recommend that we continue to make more products. So we weren’t at risk of over-producing very quickly. 

Just to clarify, you’re referring to the inventory buffers that are built into Replenishment Plus to help planners see which materials are in demand and where they might need to invest to maintain a smooth flow.

Right. However, there is a lag between when the buffers respond to decreased demand and our current signals to produce. So what we did is to introduce some adjustment factors to reduce the size of our buffers in the near term. We also shortened the horizons to make the buffers more responsive. Once we did that, we lowered our buffers and started to consume the inventory that we had while having them set to be much more sensitive to changes in demand and keeping a close eye on what they were doing.

So that set you up so that you wouldn’t overproduce while at the same time being responsive.

And then the next step was to do an analysis of our bill of materials and really identify, which are the parts that are most critical for our supply chain to have continuity? And which of these parts are a potential risk to supply? There were many materials—as you can imagine, everything that’s needed for vaccine experimentation and production—that we’re also using. Personal protective equipment was an issue. We have a clean room, and that’s where we manufacture everything, and if we don’t have PPE, we can’t make any product. 

So it was just sort of like, how do I identify where all our key raw materials and supplies are for us to be able to have continuity in business and how do we rank them and prioritize them so that we can invest and then put the proper strategies for those materials. So we did that, that worked out well. And then demand started to pick up after two or three months, and it has been picking up steadily since. Once people started to learn how to work with the virus and stay safe, things started to return to normal, and we’re running the system as we were running it before.

Did geography figure into those supply chain risk assessments? I know a lot of companies in the life science space are very heavily concentrated in Asia for ingredients and materials.

Yeah, we were not as dependent. We draw materials from all over the world, so we don’t really have a regional concentration for raw materials per se. But one of the things—and we’ve always kept a close eye on this, but I think this helped us sharpen our practice—we try to have a really clear understanding of how the different raw materials at the very base layer of our deep BOM correspond to the products we ship to our customers. How can we get more leverage? How can we get secondary sources of supply, what other improvements can we make to make our product more stable, more reliable, and less risk of supply chain disruption?

Was transportation an issue for you guys? 

No, not directly. the biggest issues that we’re facing right now are around items that are common to the vaccines. So for example, vials are hard to come by plastic, water bottles, all those things are in high demand. So it’s been a bit, harder to get them than normally. So for us, it’s very much looking at what is the stage of the vaccine experimentation, what’s the stage of the trials, when do they start manufacturing, how are they distributing them? What are the things that they consume that we consume, and how can we ensure that we have supply of those materials?

Let’s go back to the point you made about prioritizing the materials you need to maintain business continuity. There’s this idea that the pandemic exposed a flaw in many companies’ supply chains, where this relentless push for cost efficiency made it hard for many organizations to adapt to the disruptions. I wonder how you see those trade-offs and has the pandemic changed that at all?

Yeah, I think it’s something that we already kept a close eye on. Our orientation has never really been to run lean on the elements that are critical to our business. And I think that’s one of the lessons that we’ve taken to heart from our whole DDMRP strategy. We’ve identified the key areas of our business, and we make sure that they are appropriately stocked so that we can have some resiliency and be reliable. There’s variability because of shelf life, there’s biological variability. Our materials are subject to so many sources of variation that we tend to carry a sufficient amount of stock to ensure business continuity, but we don’t do it across the board. We’re very strategic about where we invest in inventory in that manner.

Have there been other changes to the way your team operates?

One thing that’s been a bit of a silver lining is just the fact that we needed to be so attentive. That has helped us improve our practice in DDMRP and be more focused on how we’re using it and maintaining the system. Once you get the system running, it doesn’t really need that much attention to keep running well. Having said that, it always benefits from attention. So the more attention that you pay to it, the more that you look into the details, the more opportunities for improvements that you can find. I think that this has gotten us into a really good rhythm of staying on top of the system and continuously improving it.

How have you structured that process? Do you have different goals, do you revisit it on a more frequent cadence?

Yeah, we do. In the past, our reviews tended to be periodical—every quarter, we’ll look at the system and we see which parts are doing well, which parts aren’t doing well. Now, we’ve started pushing ourselves to see how can we continuously monitor how different parts are performing, and when their performance gets out of certain ranges, we review them. There’s a certain series of processes that we follow to update them—inventory drifting above target rapid changes to the daily consumption, things of that nature. There are also things in our business where you could have an abnormally large order comes in or a batch that gets scrapped and has to be remade. Do you consider that for the ADU or not? How do you take all those events and weed them out so that your consumption rate is correct?

You’ve mentioned elsewhere that one of the things that drew you to DDMRP was the idea of not having to constantly spend time reconciling data and more time doing exactly that kind of analysis. So I wonder if you could maybe pull back for a moment and speak to how DDMRP supports you in that, in that journey.

Sure. When we were running with MRP and were dependent on a detailed forecast at the finished-good level, there’s so much error that builds in the forecast when it starts to aggregate into common raw materials. And people have different tolerance for risk and they hedge their bets in different ways. So you get all these different approaches to forecasting at the topline for different product lines, coming from different people, with different criteria or different expectations of what their products may or may not do. And a lot of them eventually converge in common, raw materials. So there was just so much error built in, and it was always hard to reconcile, when you had too much, or you didn’t have enough, where the error had come from.

In the past, the only avenue we had to deal with it was to try to be more precise with our forecast, but that really wasn’t getting us anywhere. So I really liked the perspective of identifying the key points of influence within your bill of materials, and how that relates to the variability in supply and demand for each of those materials and your ability to take action to address it. That has given us much better results, by focusing on efforts where we can make a difference.

How long have you been using DDMRP and Demand Driven Tech?

Let me think back when we implemented that, I would say somewhere around maybe 2014, somewhere around there.

And are there other lessons that you take from the pandemic in terms of how your industry should be working now or how it will have to work in the future?

I think one thing that comes to mind is don’t overreact. We want to make sure. And that’s again where we really like the idea of being able to just focus on a few key areas of our supply chain that we know determine how adaptive and responsive we can be. We definitely want to be prepared for risk and variability—that is a natural state of our business, whether the pandemic is there or not. So in a way that had us better positioned to respond to it, but we have to be careful of not over-correcting and going too far in the other direction. 

For us, one of the benefits that we see as a business is that there’s more alignment around how important it is to be responsive from a supply chain perspective. That gives us the ability to draw corporate resources to execute on certain projects, like lead time optimization and reduction on our entire BOM. How do you start to incorporate these concepts into the product development life cycle, and how can you work with your customers to make them aware of shorter lead times and what benefits that may have for them? If we can shorten our lead times, we’re going to have a more responsive supply chain without having to carry additional inventory. That helps us bring focus to the importance of being responsive and minimizing risk.

You use SAP systems to manage your global business. Can you talk about how DDMRP fits into that broader system and what drew you to Demand Driven Technologies?

When we implemented SAP, we were committed to being a demand driven organization, so we made sure the ERPs we were considering had DDMRP functionality. The fact that SAP did was one of the factors that steered us in that direction. Now, having said that, at the time we went live with SAP, the solution in SAP didn’t have the level of functionality for the business processes that we’re running or the maturity that we needed. So we ended up integrating our R+ DDMRP system and our DBR scheduling system with SAP, and currently planning and scheduling take place inside R+ and DBR.