Robert is the founder & chief executive of Slingshot Simulations, a firm he describes as a “decision intelligence” provider _ particularly to cities and other places that need to better understand their interconnected systems and the impacts of policy decisions on sustainability and behaviours. In this talk, Robert provides a snapshot of the current state of play in this field, and provides insights from the last 20 years of place-based modelling.
For more info on their work, please visit: https://www.slingshotsimulations.com
This talk was presented at #SmartSheffield24 on January 29th 2024 – please see http://smartsheffield.city for videos of all the talks and to sign up to receive notification of future events.
Many thanks to the Department of Computer Science at The University of Sheffield for hosting, and to our other sponsors: Arup, Sheffield Hallam University and Sheffield Digital.
So my name is Rob howood I’m I’m going to talk a little bit about some of the work that we’re doing in place-based digital twins um with specific focus on Net Zero and climate resilience but I also want to compare and contrast with where the bulk of my background has
Been in which has been in advanced product design and development where digital Twins or whatever you want to call them that connected cyber physical system in one way shape or form has been used for decades and just talk about some of the lessons or some of the
Things and differences that we might see but I thought I’d start with a little bit of interactivity I don’t know if any of you follow the the Gartner hype cycle does that mean anything to you well the last one I could find that’s free anyway
And the last one I could find was the 2018 version and digital Twins were right at the top of the hype cycle which means we’re just on the precipice of cascading over to everybody going Christ not another digital twin right so I just thought i’ just have a a little show of
Hands and this may work or it may not so my organization or yourself depending if you’re just representing yourself do you know what digital twin is yes show hands most people it might be a self- selecting audience given the topic is digital twins but you
Know we know what one is but we’re not investing in it quite a few nodding heads we’re trying to figure out how to invest in it we know we need one but we’ have no idea what to do some nodding heads we’re running a pilot yeah some nodding heads we’ve got
Operational digital twins who’s actually got them operationalized in their business I’m not seeing any I’m not seeing any nodding heads we’ve tried and we’ve said no way no okay so most so we’re still early we’re still early in that adoption we’re still in that learning phase that’s interesting all right
So apologies for the ey chat and you don’t have to read all the words I’ll try my best to to translate so I would argue that advanced product design and development leads the way in the use of digital Twins and I thought I’d try and pick a classic British institution
That’s not too far from Sheffield I know dab is wrong side of the county but it’s close enough so rolls Roy right and I don’t know if anybody’s seen are familiar with the a maturity model I’m sure there are several out there but starting with you know data have we got
Our stuff together all the way through descriptive diagnostic predictive prescriptive and now with the Advent of AI and ml we’re getting into this whole cognitive way of dealing with analytics right and if you look at if you just break down what rolls Ry are doing with
Their engines and people say oh well you know my system’s way more complex than an aircraft engine I’d challenge that right an aircraft engine has 18,000 individual components that all have to work without fail ever it delivers the power so you often just see that when you sat on the
Airplane that’s what happens when you take the skin away it delivers the power of 68 Formula 1 cars right it’s an incredible amount of power the load on one fan blade alone is is the weight of nine London buses 90 tons on the end of one of those fan blades when it’s spinning
And it’s sucking in 1.3 tons of air every second when it’s taking off so that’s a complex system to try and make work so Rolls-Royce and and others like them have been and I think you used the word digital twin been around for a long time may have been called a digital twin
But that closed loop between physical and cyber has been around for a while so they’ve got the program the intellig intelligent engine they’re creating billions of data points on every flight about their engines they have an engine Health monitoring system so it sends alerts bang there’s something going
Wrong here but they also have Smart analytics and their smart Discovery platform which starts to get into the hey what what is actually going on why is this happening and they do fully Integrated Systems level simulation not just point solution just look at the structure just look at the fluid they’re
Looking at everything in a highly integrated sense they’ve got a digital twin program they’re starting to connect it all together and they’ve already this that was a 2018 quote so they’re years in to investing in artificial intelligence to help them improve their operational availability so I would argue that physics based simulation has
Got a lot to teach sorry um Advanced product design development has got a lot to teach what we’re trying to do in place-based and analytics and I’ll get to that in a minute but here’s a here’s a quiz question for you everybody likes a drink and a quiz on a Monday night
Right so just pretend you’re in the pub so between x% and y% of digital transformation project fail and it depends whether you listen to McKenzie Forbes or Bane right who thinks it’s between 30 and 50 one the Lone Wolf who thinks it’s between 45 and 75 maybe half who thinks it’s between 70
And 95 yeah may maybe the other half maybe just shading it well you’re right between 70 and 95% of digital transformation including digital twin initiatives in these big companies and and other companies fail Forbes puts it somewhere in the middle about 84% so the odds are stacked against
Us you can start to read you know why do these things why do they why do they not work right there there are a lot of reasons three that stand out lack of alignment to business outcomes there a focus on the technology and not the outcome and we’re here talking about
Digital twins today who wants a digital twin nobody think of the analogy between the drill and the hall if you’re talking to the customer you’re talking to the business they don’t want a drill they don’t want a digital twin they want what a digital twin delivers but we all fall in love with
The drill and we need to fall in love with the hole in the wall because that’s what’s going to hang the P that’s where the pict is going and that’s what’s going to hang up so we need to focus on that lose the shiny toy syndrome and the
The other focuses on training and skills we have a massive digital shortage in the UK and it’s a global phenomena there are something like 250,000 open dat science positions in the UK right now so we’ve got a huge shortage of skills so we need to make these Technologies highly accessible as
Well so if I flip now and I said I was going to talk about place-based digital twins so things that happen in regions in locals and so forth I don’t if anybody is in that kind of play-based digital twin space but this is where where we focus as a company and the
Applications that we’re doing with specific focus on Net Zero and climate resilience applications so trying to apply this technology to either accelerate our journey to Net Zero or help us deal with the here and now in terms of of climate resilience and I would argue that at
Best oops sorry I would argue that at best we’re here on this journey data is highly fragmented um lots of it’s just still stuck in spreadsheets on Joe’s laptop down the corridor in the local Authority there’s there’s moves to make open data modeling and simulation is
Unid diens and very siloed so you might do a traffic simulation you might do a flood simulation but you don’t do the two together when in reality when you have a flood you need to understand where the vulnerable people are where the vulnerable assets are and so forth
You only get that when you do both right there’s a heavy Reliance on third- party expensive Consultants apologies to any sponsors who may be here who are third- Party expensive Consultants but there’s there’s there’s a heavy Reliance on them um strategy fragmented you know Rolls-Royce are designing the best jet
Engine nobody’s out here designing the best climate resilient City or the best Net Zero city they’re building a climate plan a traffic plan and an energy plan and they’re not talking to each other about the three so we’ve got a long way to go to build that systems View and
There’s really a lack of clarity on value creation and use cases so I would say at best we’re there some some towns and regions that we talk to are certainly here they’re a lot down here so we’ve got a lot to do so what we doing what we’re trying to do we tackle
Some of the biggest challenges that we see in that adoption so we position ourselves as decision intelligence provider means we using data Advanced Data science to try and try and solve that problem and um and Tackle climate resilience and Net Zero applications we were founded in 2019 we’re a spin out of
The University of Leeds we’ve had over5 million pounds of VCH Capital funding um since since founding we got 3 million just last year um so we’ve got some really good C Tech we’ve got a product out called Compass engine and we’re we’re a founding member of the the
Digital twin Consortium um which is an object management group led industry body out of out of the us and we use some core technology called graph technology um under the hood and again I’m going to I’m going to show some gner stuff here to try and give it some
Independent credibility again it’s a bit of an ey chart but this is their emerging Technologies and Trends impact radar it’s worth taking a look at that every year when they publish it and they talk about some of the key things that that are driving either the productivity
Revolution or the smart world the areas that we live in there’s there’s things on cyber security and privacy as well they talk about some of the key Technologies you see digital twins is right in there as getting close to maturity but one of the critical enabling Technologies is something
Called knowledge graphs and graph technology and that’s that’s the foundation of what we do with both data and with models and simulations that’s what enables us to hook everything together and connect it together if you’re not familiar with knowledge graphs and please don’t test me I’m not on the technical side of the
Organization the analogy is like going from an encyclopedia to Wikipedia so if you think about the way we organize and manage data and models and simulations today they’re like pages in an encyclopedia they’re quite unidimensional each page has a lot of information but page five and Page 125
Don’t know anything about each other even though they may have context and they may be talking about something similar so you as an individual have to Wind Through the contents page and try and make sense of everything that’s going on compare that to wik IIA you ask
A question and it tells you the answer to that question and everything that’s related to it gets connected and surfaced for you and it’s just hyperlinks we’re doing the same for scientific Technical and Engineering data sets and that means you can get to decision making way faster than you can
With with traditional models digital twins this is a government publication um improving infrastructure performance 2050 puts data and digital twins right at the middle of connecting SGS with the stuff that we need to do on the ground so this is the this this is the area this is the critical piece of technology
We need to develop and we’re starting to do that and we’ve got a whole bunch of use cases I’m not going to bore you with all the all these videos I’ll just give you a few examples of the work that we’re doing with national government Department of Transport local
Authorities um this example here is looking at where to place low emission zones most effectively and just rapidly optioneering through that this is a a project around housing regeneration where you’re not just looking at it from a carbon perspective of say hey insulating these housing better but how
Do you that and get people out of fuel poverty at the same time so it’s a socioeconomic and carbon calculation that’s coupled together it’s not unidimensional and you get a different answer when you you merge things together and and the this is a project we’ve got going on with Westminster
Right now which Westminster Council which is looking at school streets so they close the roads between 7:30 and 9:30 and between 2 and 4 where do you put those displaced traffic if that traffic starts going down streets where you’ve got houses with low EPC rating well well those buildings those people
Who live there are going to be adversely affected by noise and the emissions that you’re diverting so is it a just transition or is it just moving a problem from one place to another and then most pertinently is is something like this which is where we get the coupling of different simulations where
If we have a flood and it’s very relevant right now with the the horrendous things that happen in Chesterfield and and around where we actually started to look at if you if you do connect a flood simulation and flood data with traffic simulation how can you very quickly option here and
Analyze how you should respond to that um and this is just how we’re trying to deliver that capability you’ll see there’s no coding here you don’t really know you’re using modeling and simulation you’re using tools in the background that will give you quick directional insight into the types of
Responses that you can develop over time and we’ve done that by pre-packaging data prep packaging analytics and simulation are giving you the ability to rapidly analyze and here you you start to see rag networks of the worst impacted properties the worst impacted people and the worst impacted roads and
That’s coupling traffic and flood simulation so that’s it probably more than 10 minutes Chris I apologize but follow us on LinkedIn follow us on the website and if you’ve got any questions you can happily contact me so thank [Applause] you and