Daintree Networks
Daintree national spark: GE’s Australian acquisition
Daintree also describes itself as a startup, and its 13-year road to success has been built on a breakthrough product, ControlScope. This unique open system allows companies and building managers, retail chains, parking stations and any building or group of buildings to wirelessly control and monitor their lighting, heating and air conditioning and electrical devices from a central point.
A determination to adhere to open standards first led Daintree’s founders and founding engineers to contribute to ZigBee, a networking protocol for connecting things wirelessly, and to persist, despite many challenges, in using ZigBee to develop a building-control system that will work with virtually any lighting or electrical fixtures. As a result, ControlScope can be used worldwide to achieve savings of up to 70% on lighting bills, and to illuminate opportunities for operational efficiencies in buildings that use hundreds of electrical devices, or in facilities and organisations that use many thousands of devices.
ControlScope’s principles are closely aligned with GE’s vision for the Industrial Internet: both harness data collected from sensor-enabled infrastructure to drive incremental efficiencies for business and industries.
Daintree engineering will form an essential digital stairway in GE’s plan to connect ground control to the Predix Cloud.
GE Reports spoke with Niall Mai, vice president of engineering and technology at Daintree Networks, about the brilliance that led this small Australian company onto the global stage, and about Daintree’s hopes for the future.
GE Reports: When Daintree started, 13 years ago, it was visionary for an Australian company to develop a product like this that is applicable to the global market. How did that thinking come about?
Niall: We always had a global vision in mind. Daintree was founded in Australia, but right from the very beginning we established a legal entity in the US. There was always the expectation that certainly the US market, but also the global market was really where we wanted to play. We were never building a solution focussed on what we could do here in Australia. It was always about building something for the world.
People weren’t thinking about wireless then either. In our very early days, we would go to major lighting events, or be talking to companies, and we were trying to convince everyone that wireless controls were the way of the future. In those first couple of years everyone said, “Wireless? That’s so risky. Wireless can’t work. My WiFi doesn’t work at home. My Bluetooth headset drops out all the time. There’s no way wireless is ever going to work for building control.” So that was the initial perception around wireless. That changed and people accepted that wireless was a legitimate way forward, but it then became a question of “Why ZigBee?”
GE Reports: Why ZigBee?
Niall: Number one is that it’s a wireless network that can scale to really large numbers. If you were just connecting devices in a home or in a small building you could get away with other technologies such as WiFi. But if you want a solution that can scale to really large buildings where you’re connecting many thousands of devices on a wireless network, ZigBee is a much better protocol.
The other thing is that ZigBee has a great protocol designed just for this purpose, so it has custom messages for lights, for switches, for thermostats, for sensors; for everything you can think of in the building, there is standard messaging for those types of devices, which means that we haven’t had to create anything proprietary in terms of the network messages to be able to work with lots of different devices.
The end devices are dumb, there’s nothing significant about them. Companies manufacturing electrical devices don’t have to put special Daintree software in there or anything. They just have to make a ZigBee, open-standard product and we can connect to it and work with it on our system.
The vast majority of the world’s buildings don’t have control systems because the cost of retrofitting wired systems, ripping through walls and ceilings, is very high. A wireless system has a huge advantage.
GE Reports: Was persistence essential to your success?
Niall: That’s something we’re really proud of. We have persisted with the approach of keeping to the open standard. In the early days when we were saying that we wanted an open system that would work with any brand and product, there wasn’t actually any product to work with.
Just two weeks ago I was in San Diego for Light Fair, the biggest lighting conference in North America and on our display we had a table of a couple of dozen different wireless devices all from different manufacturers and there was a moment of recognition that this was the story we’ve been telling for so many years: that we were building this open wireless solution that would work with lots of brands and finally, here we were, being able to show that there are lots of different devices now available from many different brands, which can work with an open system like ours.
GE Reports: So it’s ControlScope’s open system that makes it applicable to a world market?
Niall: That’s definitely key, because throughout the world every different region has different types of equipment. Even something as simple as lights are very different in Australia to the US, to Europe. They’re often different drivers running on different systems. Therefore, unless you’re a big enough company to build a range of hardware and devices to work in all those different regions, you need a networking product that can work with a broad range.
The ControlScope solution also has a tiered architecture. All of the endpoints, the lights, sensors and switches, don’t do a lot of processing themselves, so they’re cheap because they’re simple. We then have our gateway or our wireless area controller or WAC as we call it. It is very powerful and does all of our edge processing, controlling all the real-time operations: it guarantees that lights are turned off and on when required and adjusts temperatures and so on. Then there’s the server that sits up behind that and is predominantly in the cloud and is very much aligned with GE’s Predix industrial-scale cloud platform: it’s collating data to generate reports and analytics, but it’s not crucial to the real-time operation of the building.
That tiering means we’ve solved a number of problems: the end devices are as cheap as they possibly can be—you need a lot of them in a building; the operation of the building is fast, efficient and guaranteed because of that edge processing we do in the WAC; and thirdly we’ve got scale to enterprise level by having servers that are collecting the data and able to do the reports and analytics at a higher level.
You can look at it now and say. “Well, that’s quite obvious that that’s the model that would work, but it wasn’t obvious when we started, and other companies that have attempted to do it don’t have the same model, and as a result I think they’re restricted as to their growth and their scale and what they can do.
GE Reports: What kinds of capabilities will ControlScope enable in the Cloud? What can companies do with that data?
Niall: A lot of interesting operational efficiencies come from it. Take a bank that has branches everywhere. The traditional model is that each bank has a fleet of maintenance vehicles that routinely drive out to every single branch to check whether any lights are out, and whether the heating or air conditioning are operating at the right temperature. They roll from building to building to check these things—most of the time to find that there’s no need to be there. The data we’re collecting lets us produce analytics that indicate when a heating unit is operating outside of temperature, or when a light has failed in one of the offices, or the outside signage isn’t working. There are real operational benefits in being able to send maintenance crews only where they’re needed..
Our engineers and the team here are passionate about saving energy and we feel very proud of our work.
Another interesting area is in understanding occupancy patterns and space utilisation. We’re able to put sensors everywhere as part of the control system, and in addition to controlling devices for energy savings we can use that information to tell us about occupancy patterns. In a warehouse, for example, we can produce analytics that say that Aisle 10 in your warehouse has the most amount of traffic, so for more efficient operations, you should move that stock up to Aisle 1 to make it more quickly accessible.
GE Reports: You now have around 10 million square metres of floorspace under management in the US. When did ControlScope start achieving success? When did you know you were onto something?
Niall: That’s an interesting question. I think there was never a single point in time where I would say we thought we’d made it. We were still challenged around the issue of the cash flow: How we could balance how many engineers we had and how fast we could develop the product with our sales income? In some ways, this acquisition by GE is the point in time when we can say, “Wow, well, we have made it, because we’ve been recognised for the position that we hold and what we do.”
GE Reports: How did the purchase come about?
Niall: We’d spoken to GE a few times over the years about working with them and helping them with wireless devices and other bits and pieces. Also, as a startup, we were in the market last year looking for further investors. We got into negotiations with GE about becoming an investor in Daintree, and once that investment conversation got under way it very quickly transformed into an acquisition path.
GE Reports: Was it a catalyst for the acquisition that Current had just been set up as its own entity, as a startup arm of GE?
Niall: Absolutely. The vision that Current has and what it’s trying to achieve within GE, both in being a fast-paced company and in trying to utilise this platform of data and connect to everything … Once they started talking to us and understanding our proposition and what we had, we became the perfect fit for Current. We’re two companies that are perfectly aligned: same visions, same ideals, talking about doing the same things but starting at different levels. GE with Current had started at the cloud end. We’d started more in the buildings, in the connected space. Our products are perfectly complementary.
GE Reports: But what does Daintree have that GE then didn’t?
Niall: Daintree provides a middle platform, that can connect all the devices in the building both from a control point of view and as a means of connecting the data to Predix. It’s that essential middle strata of how you connect to the building and connect to the devices in the building that Current didn’t yet have a broad solution for.
And ControlScope doesn’t yet talk to Predix, but we’re madly working on those integration plans. We’ll have data feeding into Predix platform within the next couple of months.
GE Reports: What dreams will being part of GE allow you to realiset?
Niall: The first thing is that we’re part of a much bigger, more stable organisation, so that will change some of the dynamics of how we operate.
Then there’s the ability to access some of GE’s resources: we may be able to utilise what some of the teams in Current are working on in analytics and the smart software in the cloud, and focus our own efforts on other areas—a really nice way to accelerate the product offering to broader markets.
The other thing that’s really exciting is the reach that GE has from the sales point of view—being able to access a global sales network.
GE Reports: Is there an Australian way of thinking that is inherent to how Daintree operates?
Niall: I think so. One of the things that I like about Australians is the attitude and the approach to work. We’re dynamic and flexible. As a startup you need to be prepared to change direction. Australian people and engineers are really suited to that model of being fast paced and being able to go with whatever needs to be done. I also think that creative problem solving comes really naturally to Australians.