Bestand wählen

Visualizing Fire Department Responses with CartoDB

Zitierlink des Filmsegments
Embed Code

Automatisierte Medienanalyse

Erkannte Entitäten
the n response this with that
what kind B is anybody in the room worked with Kennedy b before OK to small agreed to a large degree of playing around that OK great
on to just 2 things I want to talk about a 2005 but there was a vegetation fire in southern California just north of San Diego 17 fire engines 3 ambulances and 10 other support vehicles this fire lasted for nearly a week and caused by 18 million dollars worth the damage this was just back in May and the incident that was determined to be a carelessness of cigarette butts thrown in the dry grass lights up and and next thing you know there's some 12 homes destroyed 18 million dollars in property damage In 2003 March of a medical incidents of a significant size required 3 ambulances responding and therefore fatalities so I mention this
just to remind people of all the things we're here and we're discussing technology in discussing technology and demonstrating always great things in this particular case and in ultimately all these cases but it's not about the data it's not about the databases In this case the fire department about people's lives and it's about property damage at about a government's ability to serve and protect its citizens so GIS in this case open-source tools from allowing them to do that really just as a reminder our purpose here is not the time to foster technology it's to foster solutions for these focus but so visualizing information why do we wanna look at it in a more interesting way than just on spreadsheets are charts and will stories are powerful moment people wanna hear your stories people are not static beings in other words they don't do well looking at most of them don't do well looking at columns and rows of numbers of summaries of data they do better with the visual of a visual exploration of to the answer the tools like already can make it so that these decision makers fire chief city council citizens and can understand in a more intuitive and quicker way on what they're dealing with if the taxpayers what they're paying for if the fire chiefs and what their staff are supposed to be doing this is quick background and when it comes to some urban fire response in particular what we're talking about ultimately is response timeline so the word someone picks up the phone to report emergency so that phone comes in and to the emergency that indicates the USS in the case of the United States the number is 9 1 1 to dial an emergency in that call comes in the operator dispatches that call to a fire department come that fire department then the goes enRoute in other words they leave the station and they go to the incident they arrive on scene and at some point that incident is come so complete and finished and that units that fire a trucker that ambulance again becomes available so these 3 time differentials that we're talking about duration turn out how long did it take for the truck for and receive the call to when he had left the station how long did that take duration travel how long was it on the road and we're modeling travel times that and is the key metric there when they are deciding when to build new fire stations or where to relocate fire stations duration unseen how long did that event take but true ultimately metrics that come out of this duration response In other words from when the call came into when they arrived on scene how long did that take any spare departments have specific metrics that they need to meet in order to In order to be viewed as having achieved their goals and duration committed of this is more relevant to specific units status specific fire-engine the specific ambulances how long with these the units committed to responding to this incident when they're dealing with 1 incident not available to respond to another incident and so from these 2 things we can derive a large number of other metrics resource drawdown is a metric on that's often use and again this is to indicate how quickly over time are the resources available to us drawing down and at some point you reach a critical point where you are unable to respond to new incidents and so you establish relationships with neighbouring cities mutual aid so that they can help you when you resources are depleted the I'm doing another presentation on which will start late I'm on friday which goes a bit more into these metrics from of a mechanic standpoint and how we deal with these in the post GIS database so if you're interested in that that'll be Friday morning I'm so modeled response versus actual response that's ultimately what we're looking at Curry DB here help us compare is the when we take information road networks with speed limits surface types intersections of surfaced um width of lanes we use that information to essentially model travel time how long should it take to get from a fire station to any particular incident and we got actual response data which is the information that we deal with more often which is we have the timestamps from emergency 9 1 1 when that call came in when that call was received on site and and and and how long it actually took that unit to respond of so we are comparing those 2 things and what we're looking at right now are different ways that we can express and visual aside visualize any information with Cairo DB so I think I'm a skip over a little bit more this information again I can talk about some the modeling data and if you're interested on Friday the and so I
just like to get to the demonstration and hopefully the internet
gods are the on my side today and if they're not in this
presentation will be in very very short but
it and and and it if
it and it and
what it and
so we'll just take a look
again quickly here at some of the data that we're dealing with and what we're working with here is the city of Carlsbad California which is located about just north of San Diego from the city boundary is up here this fire department Carlsbad that fire department has 5 of fire stations a located throughout the city and will all be looking at just a couple of these and the information that we're dealing with when it comes to the modeling and is really two-fold 1 is the station about times so far I zoom into a particular station what we have here for colored
by all those don't show quite as good on this display here from logic from lighter to darker In other words the amount of time it takes to get from the station to the other areas that it served the information we're looking at here is from 1 to 8 minutes so anywhere from so from less than 1 minute to minutes of response so this information has been modeled for each of the stations and you can kind of see a lighter areas around each station location that indicates and really where they're coming from if we were to bring up a terrain map this part of the country in particular has got on as you get away from the coast has a lot of mountainous areas and that as you would guess begins to severely impact travel time vs. a a flatter on the area the other information working with is is travel times and so here's information for station number 1 again we're looking at these colors wrapping outward from anywhere from 1 minutes for less than 1 minute up to but how 8 minutes so we've got these ranges to green being 1 minute and this father out area being 8 minutes so we got this information for all 5 of the stations here so we just grabbed to
release 2 of these stations we're looking at stations number 3 number 5 0 what we see here are again the areas that they serve some other words these are their 1st response areas where if a call comes in any of these locations in red units were stationed number 5 would be the 1st ones that are set to respond to that if we turn on their on travel time rings we can see there's an area of overlap so in other words by design the stations can aid each other in responding to these incidents that our maybe at the outliers so station 3 is busy responding to incidents closer to itself on then station 5 even though might be in a station 3 area station 5 can respond to those the so we've got here is other red area is on those areas from both of those responses where it was going to take upwards of 8 minutes to respond so In the event the full turn those incidents on that occurred within those areas no matter which station
response it's going to take them upwards of 8 minutes to get to that location so these are the most critical period were really trying to so minimize of the fact that it might take longer to get there and managers interesting we're seeing here is that notion of actual responses verses model responses so here in red are that model data based on the road network information and expected travel times the black dots are actual incidents where it took up to 8 minutes to respond to those incidents in any of those areas so you can see already we know in every which allows assume your model is just that it's not a it's a model it's not reality so we've got is here where we were and so it took us longer to respond to those areas and what's not clear about the symbology is that there's actually some of the serious incidents where the took as far less than 8 minutes to get into those areas even though the model said that it should take a lot longer so the model is wrong maybe some of our input data is wrong and but just by presenting something like this to the fire chief verses a list of units in a list of responses which were less or greater than on he commands and we look at this and some like a fire chief was familiar with the territory with their terrain he may know these areas have Rosa under construction or perhaps these roads OMB are in need of repair and that was not indicated in the model and so this is our information again visually that they can use to help make better
decisions 1 last thing so that I'll show you and I think we should be on track to take some
questions the but is by simulating the actual travel time so where we're looking at here this is where I am myself ran into some struggles on complexities with our cargo Debian and maybe not necessarily currently be itself but more post GIS and is taking information and translating into a way that we can visualize it visualize the animation over time so if we look at this this is just a simple line that I drew obviously this is not an actual Rhode and but we have what we would have in theory is a point at 1 and let's say it's this and the line here where the incident actually occurred and we've got a point appear at the top which is where the unit was when it set out to reach that incident now that location of that of fire engine that or that ambulance maybe in the station in maybe not in the station as we we're if we refer back to earlier slide or we talked about the different timestamps where vehicle becomes available after an incident is finished it could be that that ambulances route back to the fire station and has its traveling Mary receives a call the go to another incident and so it's not in the fire station itself it's at a different location and we have that information we have the latitude longitude of all the units where they were when they were dispatched to go to an event so 1 of the things we're trying to play with this taking this information I and in this case what we've done is we've taken those 2 points 1 it's not too clear here let's do this my well taken is that starting point and ending point we generated or out from a to B but that about gives us is simply align with the number of vertices and if you use Carter dB at all they have a a tool called to work which allows you to create time-series animations and visualizations of the data that Turkey expects to have point data to work with from and so if you were to just take this simple graphic that I had shown before which was aligned made out of 4 points and you turn on animated you're going to get 4 dots on the map that go from point 1 2 . 2 2 . 3 2 . 4 it's just going to jump from 1 to the other which doesn't really give you a sense of the the of of the thing in motion so what we've done here is taken Outline interpolated points during interpolated
points the path up over time and essentially given a timestamp to each of those points so that we can use this to work tool to presenters animation now when we've done here as I was saying is just a proof of concept that that is the information we use is the gross very high level OMB and there's a lot of complexity that go into that so for example we could take this animation and compare it to on the drive time the rings that we're looking at earlier and we might put timestamps on this data so that when this all thing in motion hits the border of say the 5 minute mark so that we ensure that it's actually going to hit that point along the way another way we might do it is actually look at the road segments and the speed limits because in reality it's not going to be an even travel time from point a to point B you're going to get out and from a high-capacity freeway you're going to be able to trust travel faster you get under local road you're gonna slowdown so this from isn't would really happened but we're just starting to play with some of this kind of stuff to get a sense of of how we might do it on like I said we ran into some some issues with post JS encountered Evian just some of the limitations of those tools so that I think I gave made the timeline still here on so that everything is running this 1 it is thank some folks of fire stats who we work with the work on modeling this data you have Geographics the Carlsbad Fire Department Carter be in their community OCU knows these guys the Corradini plug-in that they made for q j us is but indispensable offer us even the crashes after every time I upload data but it still makes it so I can just use q GIS do some stuff and they're pushing up to call to Carter and then every network analyst and like it or not this is still on and do the best tool out there in my opinion for doing deep level network analysis of of travel times pt routing is great but there's a whole lot of work that goes in this setting it up right arm and it all comes down to the data of course so with that thank you again for your patience while I got things going um you know if there are any questions at all yes the in the the the I and forwarded by man is all the sense that by the is under fire department because in some and elected by whole thing on he it's the ambulance now for the system all the yet yet typically come typically any disaster no matter the scale the initial 1st responders will also will always be the local fire jurisdictions local fire local police so they will be the ones that go there 1st depending on the scope of things will have a state level or federal level emergency services that would then come in and well and then say take over as being in charge and so for the sample for that make the sense that the new have like no London kind the DB so that that use of will for a simple it's I and made uh but area or dangerous make breed of something the you should so so on and we working with the Geographics there the folks who did the transportation modeling on varied depending on what they're trying to achieve yeah they could essentially say in an earthquake we know these particular roads may become unserviceable on and so the responses are going to have to take some of other rout from they can go in and simulate and say well let's pretend that this bridge is falling down and it's impossible and 1 other thing I just point out on it's become more and more common in the United States for ambulance services to be actually run out of the fire department and so I myself at a small a car accident several weeks ago on it was myself and 1 of the gentlemen nobody was hurt a police car showed up at a substantially large size fire-engine showed up an ambulance showed up but in the end I paid 130 5 dollar fine for the failing to stop at a stop sign on but it was amazing to me that 7 people and I don't know a hundred thousand dollars worth of equipment showed up from and within 7 minutes left on so it's interesting that for us often ambulance service comes out of the fire departments interesting to note that the that that thinking the alright thank you it
Bridge <Kommunikationstechnik>
Vorzeichen <Mathematik>
Metropolitan area network
Kategorie <Mathematik>
Gebäude <Mathematik>
Kritischer Punkt
Ausreißer <Statistik>
Dienst <Informatik>
Grundsätze ordnungsmäßiger Datenverarbeitung
Ordnung <Mathematik>
SCI <Informatik>
Mathematische Logik
Demoszene <Programmierung>
Open Source
Spannweite <Stochastik>
Weg <Topologie>
Endogene Variable
Open Source
Plug in
Statistische Analyse
Demoszene <Programmierung>
Wort <Informatik>
Komplex <Algebra>
Einheit <Mathematik>
Nichtlinearer Operator
Konstruktor <Informatik>
Zentrische Streckung
Explosion <Stochastik>
Kombinatorische Gruppentheorie
Physikalische Theorie
Inverser Limes
Leistung <Physik>
Physikalisches System
Endogene Variable
Mapping <Computergraphik>
GRASS <Programm>


Formale Metadaten

Titel Visualizing Fire Department Responses with CartoDB
Serientitel FOSS4G Seoul 2015
Autor Wickman, Paul
Lizenz CC-Namensnennung - keine kommerzielle Nutzung - Weitergabe unter gleichen Bedingungen 3.0 Deutschland:
Sie dürfen das Werk bzw. den Inhalt zu jedem legalen und nicht-kommerziellen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen und das Werk bzw. diesen Inhalt auch in veränderter Form nur unter den Bedingungen dieser Lizenz weitergeben.
DOI 10.5446/32084
Herausgeber FOSS4G, Open Source Geospatial Foundation (OSGeo)
Erscheinungsjahr 2015
Sprache Englisch
Produzent FOSS4G KOREA
Produktionsjahr 2015
Produktionsort Seoul, South Korea

Inhaltliche Metadaten

Fachgebiet Informatik
Abstract Local government fire departments need to demonstrate their performance and efficiency. In this session we will show how CartoDB and Torque are being used to visualize fire department responses to emergency events throughout the city allowing city officials to better understand how they are performing. We will also briefly discuss why routing based on Open Street Maps is not yet sufficient enough to be used for this analysis. Effective Response Force (ERF) is one method that fire departments use to measure their level of success. An ERF is a set of specific resources required to perform a particular task within a set amount of time. For example, the Effective Response Force for a residential building fire, which is less than 200 square meters in size, needs to be four fire engines, one ambulance and a fire chief. These resources may be coming from different fire stations; they may be coming directly from other emergency events. They may even come from neighboring cities. Using CartoDB and Torque we can visualize several things; the expected travel routes each of these resources may have taken, compare these routes to expected drive-times based on GIS road network analysis and also show the order in which each of these resources arrived at the destination.

Ähnliche Filme