ALL POSTS BY

Blog

New Business Models - IIoT for Manufacturing

Blog New Business Models - IIoT for Manufacturing

We had a great time meeting so many people at Google Cloud Next 18! One of the highlights for us was the invitation for our very own Guneet Bedi, VP of Sales and GM Americas, to join the Intelligent IoT for Manufacturing panel.

 

If you didn't get to see this panel in person, or would just like to check it out again, you can watch the video here:  

 

Read the transcript of Guneet's Portion here:

Thank you for taking the time. I'm here from relayr. I lead all the commercial activities in North America.

 

I want to talk to you about new business models. Like a lot of you, we've spent so much time in engineering and designing the right solutions, it's hard for an engineer like myself to take a step back and say... why? Why is everything that Jen said important, and why is it critical for our customers and the industrial manufacturing world to actually look at new business outcomes and new business models?

 

Show of quick hands... how many people from the industrial manufacturing world? Wow! That's almost like one third of the room, so I'm sure you relate to us. We have a booth down there if you want more detailed conversations. Happy if you stop by.

 

But, who is relayer? Why am I here? We're a very committed partner to Google. We have about ten years of experience in this space, which makes us veterans on how we actually get data and transform industrial and manufacturing environments. We are about 200 people strong, headquartered in Germany, from an R&D perspective, and of course spread out globally in the Western Hemisphere. What we focus on is enabling the industrial customers to stay relevant by creating OPEX (operational expense) models. So I'll talk probably one slide on technology and then I'll talk more about how we are actually using multiple technologies that Jen shared and what Ryan will share to deliver actual business transformation.

 

We focus a lot on the edge and getting data from actual machines on site. We focus on a lot of artificial intelligence, using tools like DCPs tools but more industrial focused, just to give you a sense.

Screen Shot 2018-08-06 at 1.10.45 PM

What we do, which is very interesting, I don't know if you've seen this before: On the top right, in the red, we talk about technology. We leverage a lot of Technology from the Google cloud platform, and we have some wrappers that we add on top of that which actually puts this together as a SaaS. There's a lot of infrastructures service and PaaS services which we encapsulate into a solution; Visualization AI, a lot of focus on the edge. That's where we started as a company, and we realized very quickly - and for all of you guys in the industry will relate to this - it's hard to do this transformation, because there are so many different problems to solve, especially the legacy equipment that you have.

 

So we actually developed a very large and qualified Professional Services team which is the other red on the bottom, to say hey you need to pick the right sensor kits, the right frequency, the right specifications. You need to actually customize all the tools which are available for your specific environment, and create an MVP and beyond.

 

The other two sections in gray, which is very very interesting is there is risk. From your management perspective or a C- level perspective, there is risk in doing this. They’re saying “I've been doing this for 140 years or more. Why should I change?” I think that's where one of our investors, the world’s largest reinsurance firm, and we actually create models to give the data - which Jen described beautifully - data as the new oil to insurance companies to smarten the actuarial models, to actually guarantee outcomes.  So if you're looking at an outcome with the backing of the insurance and the data that we have, we can guarantee it. I'll give you an example later in my presentation.

 

The second part, which a lot of you will appreciate, is that if you're going from a CAPEX model where you sell a lot of equipment upfront into as-a-service model, there is a decrease in volume in the short term. So how do you finance that, especially for smaller organizations? We're enabling the whole industrial and manufacturing space with the combination of Insurance, Finance, Services - or what we call Delivery, and Technology. A lot of the technology components which were discussed by Jen.

 

I think this is an interesting problem because we also focus on the mid-market. The top five challenges that we see with with CEOs or other C-levels is, like I said, there's a financial risk to do this. A lot of people in the industrial world are really good at what they do. They make machines or they  use machines to make widgets and make equipment, but they don't have digital and IOT technical skills. By the way, these skills are really new.

 

There's a lack of digital strategy, too. Because it really has to fundamentally change the way you're doing things, especially if you're trying to sell as-a-service, and I'll describe what that means. And, of course there's a lot of legacy infrastructure. So that's where it gets complicated. It's not only a software and cloud problem. How do you roll out these retrofit kits across millions of devices? How do you make sure if something goes bad in this asset do you go and replace it? So you need the whole infrastructure of distributors, people who can stock hardware, etc. It’s about getting all those pieces together, which a lot of industrial companies are not used to but a lot of the hardware networking vendors are.

 

Like I said, I'll show you one technical slide and get back to what I was saying. From a tech perspective, you see similar to what Jen showed, there's just fewer services.

Screen Shot 2018-08-06 at 1.21.42 PM

What we do today is we have an edge solution which is our IoT gateway agent, which we use almost 99% of the time. I'm so glad Google is actually thinking of the edge that seriously because almost every industrial customer needs it for latency and decision-making at the edge for cost and, for certain customers - specifically overseas, it's also anonymity. What kind of data do I want to send? We work with bottling plants, and Coca-Cola bottlers would never allow you to tell the number of bottles that were actually manufactured in a plant. That's just a no-no. That's their business model, I think that’s why edge is important, to actually make sure you're sending the right kind of information. We use IOT core as well as Pub/Sub a lot, and are starting to use a lot of other GCP services which sit on top of our software platform.

 

In all of that, I'm talking about what is equipment-as-a- service? Why are we why really doing this? I don't know if you knew this but, fun fact: 80% of the expense of a machine is for running the machine, not to purchase it. Anybody who uses a heavy industry machine in manufacturing environments or other environments, for example you buy a textile plant or you buy motors or pumps or elevators, you spend more than 80% of the total cost of ownership in the lifecycle of the asset. On maintaining it, on servicing it, on spare parts, on actually break/fix. And, of course, the machine builders are getting squeezed on margin up front.

Screen Shot 2018-08-06 at 1.19.16 PM

So, the idea is how do I plane that as-a-service/servitization model for that 80% additional revenue versus just selling the machine upfront? That's the problem statement. How we're solving it with IOT, as well as the business model, is how about you give your asset for free and you actually charged the customer on the utilization? It’s not a lease, just to clarify. It's a model where you actually charge per coffee cup made which is actually a true use case, per aluminum part treated by an automotive company like BMW or Ford. So how do you actually charge per utilization? Which means, SLA and warranties are important. Because now somebody else is maintaining your asset and your supply chain. You need to make sure that bloody well works, and works within 99.9 percent efficiency. You want to make sure you're getting data out of that, so you can remotely diagnose if any problem occurs. And you want utilization data, so you can actually bill a customer based on utilization.

 

That's where we see the real value of IOT, coming in to get all of that you know information, to do predictive maintenance, anomaly detection, make sure the asset runs, give SLA warranties, as well give you the real utilization so you can forecast your revenue a lot more predictably. Do you know why insurance comes in to play in this?  Because it’s important that an asset runs 99.9 percent or more, and that's where insurance can come and give guarantees. If they have that feed of data of how those assets are running. That's really what the big innovation we're seeing in the market using IOT and IOT core.

 

I'll leave you with you know one detailed example. I’ll spend a couple of minutes here, since it makes it real when you're looking at it. This is a company called Aluvation. We actually have a replica of this heat treatment plant in our booth. This is running on GCP today, and the whole idea here is previously customers had to ship their aluminum parts. By the way, fun fact: six years ago very few parts in cars were aluminum. Today it's 80 percent or more. So, there's a lot of aluminum in every car. But the whole idea was you ship the part somewhere, it gets treated, it comes back.

 

The first innovation our customer did is make micro-factories. They made these containerized factories, which can go on ships and trucks, which could treat the aluminum parts at their customers’ sites. The are two big challenges here. First challenge is how do you monitor these and make sure these are running because these are not in your factory now. These are on the OEMs - you know the BMWs and Fords - parking lots. Second challenge is the OEMS said they will pay just for the parts treated, since they can't forecast in their supply chain long term how their vehicle models will sell for 10 years anymore.

 

So, what we’re doing for Aluvation is basically Heat Treatment-as-a-Service. We're integrating with their existing infrastructure, which happens to be Siemens PLCs in this case because it's in Germany. We get all this data and from our edge solution in the Gateway and then we integrate with both the billing system and with AI to predict if something is gonna happen before it happens. We've increased the demand planning by almost 40% percent actually, and we've been able to diagnose two weeks in advance before something fails, resulting in almost 16% more uptime. So, it's a real case, and it's a real case in a very old-school industry. Mr. Belte, the CEO, expects he'll make about 40% more revenue just by selling this as-a-service or utilization in the next two years versus selling the micro-factories up front. These cost about $1 million, but instead of selling these $1 million up front he's gonna actually provide them for free and then charge utilization base pricing.

 

This gives you an example, you wouldn’t generally think of - heat treatment of all things. The last thought, then I’ll pass it back Jen, is lots of ways insurance can help. I think this is new for you guys, but you know when you and your management are thinking of implementing digital in IOT think about this model where insurance can back you up, to both give you a safety net for as-a- service model and take the financial risk away from your books, and guarantee outcomes be it SLAs, uptime, or whatever the outcome that you're looking for a business perspective. And we're so confident that we actually give you a money-back guarantee. So, if you want to do a project and get started and you know if that fails they're actually backing us up to give you money back as a customer. Insurance is playing a big role in this, which makes sense to us because we're in this world.

 

With that thought, I will pass it back to Jen.

Learn More: IIoT Enabled Business Outcomes in Manufacturing

Subscribe to Email Updates

Recent Posts