Hello everybody. Thank you for joining us for this session.
I'm John Anthony, uh, past president of British APO, um,
so I hope you're all having a good time today. It feels good,
so, uh, let's hope it continues today, tonight, tomorrow, and carries on in the way it does.
Uh, we've got no pianos, no music, no singing for this session,
as we did first thing this morning.
But this session's all about cross-agency cooperation and learning from what we do.
Um, this was going to flow into a Jessup session after these two,
but that's not gonna happen, so.
No one from Jessup could make it, so we're just gonna flow on through these two sessions,
so we've got a little bit more time if we need it, but our first session is very much based on
uh a police experience, so we've got um, Ian Williams, a former police officer now with
Motorola Solutions, and we've got Dave Hannan, who is a serving Chief Inspector from
Lancashire Police, and we're gonna hear about, um, really,
your experiences as a silver commander and dealing with the information flows that come to
you, or perhaps don't always come how you'd like them to.
So, gentlemen, if you'd like to take the floor, thank you.
There will be some time for questions after as well.
Thank you. Uh, thank you, John.
Um, yeah, so I, Ian Williams, I'm the Motorola side of things,
so I was a, um, a police officer until about 5 years ago,
and I very much used to do the job that Dave does now.
So I was that public order commander, that firearms commander.
And I also ran the operational side of digital policing for a force,
and I'm very pleased that Dave's come with us because Lancashire Police are a really valued
partner of ours. They've got a number of solutions from us,
so we'll tell you a little bit more about that, how Dave came to talk to us about what we're
going to talk about and some of the problems they've got.
And how we're going to hopefully put a proof of concept together to solve some of those
problems. So Dave, good afternoon, everybody.
So I'm Dave Hannan. I'm a silver commander at Lancashire Police,
and the reason why the Silver commander bit is quite important is that essentially is the
challenge that we've set Motorola and one of the things we're looking to achieve.
Ultimately what that means is I'm responsible for managing big incidents,
events, and sporting events. So I will do anything at football matches up to
a high level risk, um, events that I work at West Division in Lancashire,
so we have Blackpool, so we have the air show that has about 350,000 people come to the air
show, and I was involved last year during the incident post um.
Southport. So you've got a varied vast array of different
incidents that we deal with from planned to spontaneous incidents to quite large scale,
and what I've discussed with Motorola is the challenges that I have as a silver commander in
relation to the information that I get because when you go through it,
and we'll go through some of these now.
Can we go straight into it, we'll go is um I don't necessarily get all the information,
so I am very much reliant on a frontline officer telling me what's going on.
So quite often I'm nowhere near the scene as a silver commander.
That is a tactical operation. We have a bronze commander.
They are down on the ground and they will do all of the tactical work in terms of managing
the resources. I then set the parameters.
If I don't know what's going on, it's very difficult for me to actually set what it is
that I want and what I need.
Um, so I am very much reliant quite often on the officers telling me.
Sometimes we do have some CCTV, we might be able to get some drone footage,
but an awful lot of it is them telling us.
And then I'm still relying on them being where I expect them to be,
because quite often I'll tell them that they need to be at point A.
And they say yes, we're at point A.
Do they, are they there? I don't know, cause I actually haven't got a
way of checking it out.
So the problem with we've set them out here is, um, planning for events in terms of how is that
planned and how do we work it through in terms of manage our resources,
making those informed decisions.
So in policing we use the NDM, the national decision making model,
and that's always about what information you get in.
So the more information I've got, the more I can then put into my decision making,
and I can use those resources correctly.
Anticipating those potential risks respond effectively to changing circumstances.
So again, how do I know what that's from, because if I've got it from the front line cops,
it's something happening elsewhere that I don't know about because I'm blind to that.
So where is that information coming in? Is it coming in through telephone calls or any
other form of information, which will then allow us to enhance safety,
mitigate risks, and improve overall performance in high pressure environments like public order.
Very efficient, that last line, um.
I'm gonna put that one on there. So when we're looking at the events,
we go through planning meetings. That's for pre-planned.
Obviously, as you know, the spontaneous incidents, we don't have the foresight to be
able to turn around and say we need X amount of resources.
If the panic button's pressed because something's kicked off,
we take the resources that we've got available to us.
But if the event is something that we know about, for example,
a football match, we can plan it and we go through various planning meetings where we will
sort out what resources we want based on the information and risk.
Then as a silver commander, I'm based, usually either at a divisional headquarters if I'm at
Blackpool or at headquarters uh in Hutton, and then I get the information feeds in.
What this is trying to depict here is the different information feeds I've got.
So I can get some CCTV access.
I don't have any control of that CCTV access, so I only get the feed that I'm given from that
local authority. So they will then feed in.
If I know that there are other cameras, if I'm got the foresight of a pre-planned operation,
I can put an officer working in that control room or they've got a radio to try and get them
to give me that feed. Otherwise I'm reliant on what they show me on
that screen. I've got no control.
Um, I've got some access to football clubs, again, same theory,
it's their CCV system, so it's what they give us access to,
so I can only see what they send to us.
And then other feeds that we sometimes have access to is,
um, drones, uh, helicopter and body one video.
At the moment we haven't got access to body one video live streaming.
It is something that we will come to and in car footage.
So these are things that I don't have access to at this moment in time.
The drone works, um, sporadically, um.
So as you can imagine, all I've got is cops say so, some CCTV that I've got no control over,
um, and possibly a drone if it's available to me, and that is my situational awareness,
and I'm nowhere near that incident, so I've got to rely on my cops.
We're in 2025 and I'm I'm a big believer that we can do this better.
There's technology available to us and I know there's more that we can do.
So I start these conversations with Motorola and with Ian around how can you make this
better? What can you do?
How can we make it a little bit more joined up so that as a silver commander,
I've got all the information that I need to make those informed decisions.
That moved Ah, there we go.
This is an interesting one we've discussed around as well.
So spotters and fit, this comes on to some of the other um options we've got later on down
the line, is if I'm doing a football match, we will brief the resources and we'll tell them,
these people are banned from the football match. They've got football banning orders,
they are not allowed in that football ground.
So the cops get shown that, they get their images and off they go on their jolly ways,
and they completely forget that image, they completely forget who that person is,
and Again, we are reliant on those few officers about identifying those cops.
Again, when 2025 is not a better way that we can stop people that shouldn't be going into a
football match, going into that football match, other than relying on a couple of people who
are our spotters. So there may be options moving forward.
Postma and that's the one afterwards. So the post-match investigations which we will
come to, that one is more important, I would suggest for,
for example, after the Southport incident.
So in Lancashire, we did have a significant period of disorder in Blackpool.
There was things thrown at our officers. There was public order.
public disorder, so we went into a period after that of a significant police operation to find
that evidence, and that was all manual trawls. We literally had officers sat for they've
worked out probably about 200 hours or something like that manually going through CCTV
to try and find the offenders for certain things.
Again, that is a lot of officer time looking at footage that can potentially be done a
different way. So we've got the planning into the operational
phase, into the um post-matcher or post incident investigation.
that one goes. So summary of the problems for as a silver
commander, um, so I get the organised social learning post operation is a paper exercise.
You'll be surprised quite a lot of it is a paper exercise.
Uh single live stream, CCTV from local authorities, no access to the public CCTV or
limited, um, it varies on clubs, no operational analytics, um,
bronze units, poor situational awareness. Now this is an interesting one with the bronze
units. I sometimes get CCTV footage, but then the
bronze commander doesn't get it. So the bronze commander can see what they can
see. I can see something else, and then we get a
terminology that we have in policing called um brilling, which is where the silver is actually
then starting to direct the resources that the bronze should be doing cause the silver can see
something that the bronze can't see.
So we're not seeing the same thing.
um, so why can't they see what I can see and why can't they tell me?
So there's challenges there, paper log decisions again we're doing it all on paper,
um, and validation on identity in the field and then post investigation,
it's very time consuming, so.
That, in summary, very quickly is the challenge that I set,
Motorola, can we do this better because it's got to be done better in 2025.
And I think Ian's gonna try and show you what we could potentially do.
Yeah, thank you. So.
Just coming back to, we're not talking about one single solution here,
we're talking about a number of solutions working together that are going to solve some
of the problems that Dave's highlighted.
So looking looking down those there, we've got resilience and planning software that allow us
to do that pre-planning.
Some of it's a little bit about the operational running from a command level,
but it's also about the post operation debrief and having all that auditable so that you can
actually learn from.
Your operations and feed it into future operations.
And then we've got cloud NVRs, which is basically, and I'll come on to that,
but it enables you to look at multi-stream CCTV, which is one of those problems,
but also allows you to put analytics on top so you can actually start to find people and
vehicles, etc. Um So multi layered mapping, I mean we've got a
product manager here who's actually busy at the moment building out more intrusive mapping
on multiple levels to allow us to not just look at resources and events,
but all sorts of information that either the force has got or that's publicly available to
enable those bronze and silver commanders to do that job better.
Retrospective facial recognition, but this also goes into operator
initiated facial recognition.
Some of this is through the Home Office, so it's national facial recognition in terms of
identifying people from the national custody images, but some of it is our own,
which allows you to introduce your own watch lists and vulnerability as well there,
so you can start to really nail down into that offender management,
football banning orders, that sort of thing.
Then lastly, it comes back to where we started, which is that the post event appearance search.
So this isn't live during an operation where you're trying to find people and places and
vehicles that are moving around in the operation.
This is when you're coming to wash up and you've got hundreds of hours of footage and you
need to trawl through it to actually find those suspects and those people,
and it's about making sure that Dave's not spending.
Hundreds of hours trawling through that footage to find the people that they're looking for.
So those are the solutions we're looking to put in.
What's bordered there is the stuff that we're talking about for proof of concept at the
moment in terms of how we feed in an existing operation that you've already
done, and how we can really mapping the savings about what you're getting from that,
because we know at the moment what it takes to do that operation,
to go through that process of putting hundreds of officers hours into finding those people.
What we're looking at is applying this technology to rerun that historic situation
to see what the savings would be, so that's all work in progress and hopefully we can report
back on that in the weeks to come.
The multilayered mapping and the resilience with those outside conversations,
not in scope at the moment, but they hopefully will be in the future.
So what does that look like? So the first thing is that software that allows
us to do that pre planning and it allows us to put in those trusted plans that at the moment
are sitting there on paper that keep getting brought out every time there's a new operation.
It's about putting those into software so that you can have decision trees and working towards
making deploying the right resources, but leaving an audit trail of what you've done as
well. And again, you've seen element of this before.
This is the silver command aspect, so it's really about joining up all of that CCTV from
the various sources, whether it's full own systems of bodyw video,
IA, NAS, etc.
or whether it's from the clubs that various clubs and partners you're dealing with,
or the councils, and it's bringing all that into one system.
So there's two key things here. One is that Instead of having a single feed
from each council that you have to then shout up on the radio to get that feed in and move
the camera, the police will have control of that, so they'll be able to look at all of the
partners' cameras and have a better and quicker way of finding that information.
The other side to that is because we're putting in those NVRs in the partners
and in the police, apart from being able to see multiple live feeds,
you'll also be able to apply analytics to that, so you'll be able to find people and vehicles
quickly in almost real time in terms of the cameras that are available to you.
And then we've got the mapping side of stuff which Dave talked quite a lot about knowing
where his resources are all the time, um.
And then, as we come down to the bronze commanders, so these are the people out in the
field actually supposed to be deploying people and solving the problems rather than from that
strategic silver level, they also are going to benefit from better situational awareness
because they will have access to some of these tools as well.
By putting in the facial recognition there.
retrospective facial recognition is normally used for,
you've got an image of someone who's committed a crime,
you don't know who they are, and you're going to go off and find them.
We have the ability to apply that during a live operation, so I'm not talking about live facial
recognition here, I'm talking about bringing live feeds in to a retrospective facial
recognition system that allows somebody in the control room or a bronze commander.
To pick out particular faces in that video and search on them.
So in other words, it's not live facial recognition, but it's very near live facial
recognition, and it allows you to actually look at the specific people you're looking at rather
than that collateral intrusion of everybody in the video.
So an example would be Dave's or Dave's bronze.
We've got 50 risk supporters stopped at the turnstiles, very quickly we could get someone
from his forward intelligence teams to be filming that into the control room,
and someone in the control room could be picking out individuals that they want to
identify and focus in and pull them out and arrest them as necessary.
Following on from that, the officers in the field, so those officers deployed,
not just your FIT teams and your spotters, but also your wider frontline officers are
going to benefit from officer initiated facial recognition.
So that's been able to do it on a mobile device.
So Lancashire use Fronto, which is our mobile solution.
We are building into that the ability to search the The Home Office biometrics
system of custody images, national custody images, but also the proof of concept taken in
their own version of facial recognition, which will allow them to put local watch lists on for
missing persons, for football banning orders and that sort of thing.
So those officers will very quickly, by taking a photograph,
be able to identify people who are known to the police because they've been arrested,
but also anyone who's of interest through intelligence, etc.
And that's just put an arrow in the top, because at the end of the operation,
you obviously want to do a debrief, and because you've collated all of this on a digital system,
it makes that system of debriefing a lot easier, but it also makes that corporate knowledge
easier to access for future operations.
And lastly, we're talking about the post-match investigations.
So there's two elements to this. One is that,
again, back to that facial recognition, but before we get to that is the the AI object
search. So if you can imagine that you've got,
say, 70 hours of footage and you've got several offences and you've identified that that person
with that red top has committed that offence, you want to find out where else they appear in
all that footage you've got.
At the moment, you're having to sit down 10 detectives to go and watch hundreds of hours of
footage. They do that, and they're going to miss stuff.
So the AI there allows you to actually very quickly find similar incidents of that person,
that vehicle, etc. so that you can then go back and very quickly
in the short space of time, find what you're looking for,
find their associates, etc. etc.
And obviously working with facial recognition, once you've identified them through appearance
search, you can actually identify them if they're on your databases very quickly,
so it should make that post investigation for something like Southport a lot quicker.
Um, and Dave, I don't know, do you want to talk through that?
So this is, so these are the sort of, um, outcomes really that we want to,
um, see if we can achieve.
So what I'm hoping for at the end of this, so as I said,
this is a proof of concept, none of this is live yet.
We're not going through anything. We are working with Motorola around what
options, how it's doable, as with every IT systems I learned,
it's never as simple as just switching something on.
Um, but at the end of this, I'm hoping that we've got far better audited planning,
including risk assessments, briefing, and implementation.
Um, that'll just make us a little bit sort of cleaner in terms of what we've done,
how we can learn from it, and then we can be a bit more of an intelligent organisation to make
the relevant changes based on anything we do to get right or wrong.
Um, full situational awareness of events and resources.
I mean, that's a really big one for me.
Um, I talk around the resources and the mapping, I laugh and joke about this one,
but I generally I've got an event coming up in a few weeks and I've got somebody called a
plotter. I've got a person that is going to sit on a
board and tell me where the resources are.
It's ridiculous.
Why have I got someone literally doing that when I should have the technology that shows me
where they all are. That's where we need to be and that's where
we're going. So if I can get that situation,
and it is better because at the same time he's telling me that they are here and guarantee
they're in McDonald's over here, but I might as well just put the camera from McDonald's.
That's where they'll probably be real time identification of subjects and vehicles,
um, so if we've got the intelligence information, we can start to track them,
um, using some of the technology that's available to us which will allow the.
Identify where I went to them, what I'm putting resources there,
but again, play that out, if I know who they are, I can send the right resource cos they're
the nearest ones. So instead of sending someone from anywhere to
go, I'm gonna send the one that's there that's gonna go over there cos it's quickest.
Real time alerting to common foreseeable events leading to a faster response.
This is a really big one for something around missing from homes that we can do,
um, or a wanted offender. If an offender's got a certain location,
I can then alert everybody, maybe with a photograph, send it to all the people that are
in the area and say that's the offender, because if you're going to walk past him,
that's the one football events, section 34s, which are notices.
We can ban people from a certain area. Again, I can then send that image to everybody
so that they know this person is banned from this area because of the behaviour that he's
done. At this moment in time, if I give somebody a
Section 34, he can pretty much probably walk past Ian in that area and I will never know
that they've been given that Section 34.
So what technology can we use to make sure that everyone is aware of that availability?
Audited decisions by commanders, so that's just making sure that everything we're doing is
recorded. We do do it at the moment, but it's paper,
so it's better ways of doing it, and obviously auditing the debrief.
So the major investigations. It's something that I'm really keen to look at
the proof of concept. I'm really intrigued, but obviously out of that
Southport investigation we did do.
Hours and hours and hours and obviously we can run that through the system and they can come
back and go, we found everything that you found within X amount of time and guess what,
you found all this that you actually might have missed, then that's going to be really quite
interesting and that'll be a great user case for us moving forward.
Um, and obviously then it'll obviously identify things that we've probably missed,
um, and then that will obviously efficiencies and savings,
reducing the error rates, which saves us money, which is always a great one to try and get
things through. So that is what I'm hoping for,
and that's the challenge I've set in.
And, and that's it from us, really, um, I'm sure.
Uh, you've probably got some questions. If it's anything too technical,
don't ask us because we're both cops or ex-cops, so we don't know anything about technology.
We just know about the outcomes.
OK, OK, thanks very much to you both, uh, to Dave and Ian for that.
Uh, do we have any questions? If so, I shall bring this microphone to you.
Bet it for you. Hi, um, my name's Kendall, I'm from the Home
Office. You said you searched the Home Office biometric
systems. Isn't it possible, um, to implement a system
where that immediately flags to you, somebody who's on a football banning order?
So something like a banning order, if it's on the PNC,
yes, it will come through the Home Office, absolutely.
There are other things that the force will, so just to be clear on the Home Office side of
stuff, all forces are going to get access if they wish to,
they're going to get access to the Home Office biometrics OIFR system,
and we're working with another force in order to get that live hopefully quite soon,
and then all of our customers will benefit from that.
In terms of actually football banning orders, I'd agree with you if it's on the PNC,
they'll flag up and they'll get a hit from that.
I think there are other areas of policing where you're not going to get that missing persons is
a key one because unless someone's a missing person and they are also known to the police
because they've been arrested, they're not going to be on that system.
So having a separate system.
Um, a separate, um, OIFR system will allow you to put other local watch lists in that aren't
necessarily available on the uh on the Home Office site.
Does that answer the question?
Give me your back now. Yeah, I'm, I'm just saying that I think a lot
of people through migration, passport control, there's a lot of biometric information.
You just go through Houston, you know, there's about 40 or 50 high resolution cameras there.
Going down the line in the future and what will be possible will be huge.
It's obviously, it's all gonna be staged approach and when we can get access to it,
uh, and all the legalities behind it, which is why we're doing everything as slowly or quickly
as we can within the regulations that we can do it.
So OK, thank you. I think it always amazes me that in this
technical digital world that we live in, when we look back at what we're doing in emergency
services, it is still such an analogue world that we're in.
Um, I guess your next steps is to come up with a road plan,
business case, etc.
to start it sounds like you're quite early stages.
An awful lot, well, it's the business case and the business plan's pretty much already in
place. We've got something in.
In play and that's what we're saying we're talking about the proof of concept,
so a lot of this is actually ongoing in the background.
Some of it is closer to be ready to be um launched, some of it is a little bit further
ahead, some of it is just waiting for agreements to sign off,
so there's several different facets when we're just picking them off back next year for an
update then. Yeah, I think so.
If we can drag him back here. OK.
Any other questions?
If not, we'll show our appreciation to you both. Thank you,
David. Thank you.