7/25/2010 Bronx 4th Alarm Box 4795

3511 said:
It is not my intent in raising this issue to seek a personal vendetta against anyone. If the individual of whom you "elude" to is who I think it is, I disagree. He has flatly denied any input to corrupt the FDNY response system. I know the man, he is upstanding, and I take him at his word.

No vendettas here either.  Just stating what I know and have known for many many years.  I too know the fella, he ran a good coffee mess of MT, and he really enjoyed his Friday Nights in the Bronx back in the day. Maybe its just coinsidence then...... 

My focus in on a system that appears to be flawed.

Agreed

I guess in the grand scheme of things in this world, who responds whatever due to alarm boxes in the Borough of the Bronx is not that important. But maybe not.
Being how things run in the "Gentleman's Fire Department" it is important

Maybe the citizens of NYC, the TAXPAYERS, are entitled to the most efficient system that will save their lives.
It also saves the lives of the members who work there.
Maybe, in this time of financial trouble, units could be closed based upon their workload. Both the citizens and the FDNY would be shortchanged if units were closed because of an inefficient (or biased) system that determines that workload.
Not to sure I follow you here, Geographic Locations are more important then workload in my mind.

If minutes really do count in units arriving to a fire, then this issue has some importance.
Being that fire doubles in size and intensity every 30 seconds; I would say yes.

Perhaps some of the other participants of this forum can contribute examples of where, in other neighborhoods, and how, the response system is flawed.

Lower Manhattan comes to mind.  With all the closures since the inception of the FD there have been numerous realighnments for the response cards not to mention that the redirection and composition of streets, complete disappearance of streets due to Zeilenbau and gentrification.
 
Interesting replies guys. Thanks.

445917,

I have heard those stories. I don't have any first hand knowledge and so will not join in the accusations. I will take a man's word.

Are those the truck companies you worked at or buffed?

Bomberos,

Great points you made, particularly that firefighters' lives are also at stake.

As to workload vs location, I would think they would both be a factor in determining closures. See events in City Island. If it were one OR the the other, there would either be no company there at all (workload), or multiple sections year round (isolated location) rather than threatened closure.

To both of you, of course previous closings and new locations of firehouses changes the response assignments. No problem there.

I see a "system", however, that appears somewhat random in assigning companies beyond the obvious 1st and 2d due at a box. Sometimes it seems to heavily weight distance, other times accessibility. And then it seems to measure as the crow flies (and fire engines ain't crows) or consider one-way streets, which units don't necessarily adhere to in an emergency.

In fact, I don't see  much of a system at all.






 
So basically if I'm following correctly, someone is playing favorites?
 
I am not implying that either. I have been in the Communications Offices and the dispatchers certainly don't. The units just pop up on their screen. They have input when things get real busy but then they are just trying to get the closest available to the fire. Very professional. And none of the chief officers I know claim to have any input into the assignments.

I guess I am suggesting that the computer program itself is flawed, not weighting the appropriate factors to determine who should respond. Or maybe whoever inputs the criteria in the program is not doing it correctly, if there is such a person. Garbage in, garbage out, ya' know.

Anybody know how that program is administered?
 
I remember reading the Rand study that was done in the 60's.  A lot of technical stuff,  however one point I recall reading was that the 3rd due engine assigned on a Box was not necessarily the 3rd closest engine. The rational was to not strip an area of fire protection ( the 3 closest engines to a box) but  send the 1st and 2nd closest engines, the 3rd would come from a greater distance. This would leave an engine near the area of the first box available for a subsequence alarm. They developed some long-assed algorithm to figure a response.
 
The third closest Engine not responding was known as the "Reserve Engine" the idea at that time was not to completely strip an area...(even though relocations were still done)........i know of one particular BKLYN  Engine who was in reserve on all the boxes in a several block area that was being vacated & demolished to make way for several Hi Rise Project type bldgs..........there was a tremendous amount of jobs there during about a year that it took to vacate the people & business s & level the area......at the time demo companies got premium pay if working in a fire damaged bldg.......so do the math.......this particular Engine could look out the top floor rear window of the FH & see the smoke ........even if it was a multiple they did not go.
 
The way Starfire is set up, it's line of sight distance in 1,000 foot increments from a box location. It does not take into account twists and turns or geography. Is it flawed?? At face value, no. Can it be tweaked?? Absolutely. Should it be tweaked?? Absolutely.

A new fella was just assigned to Alarm Assignments, but from what I hear, the first order of business is street files. And honestly, that's more important. Can't get a box number if the street isn't in the database.
 
What you have to realize is that FDNY Starfire assignments are based on the fire units - direct line / air miles - to a given location. For this incident, was the correct box used? The box used was at Field Pl, & not 183 St, & the assignment is changed.  When 75 moved, were their assignment ever changed?

First the assignment for East 183 St & Grand Concourse was 75-48-42-46-79-43-88-92.
Next, the assignment for Field Pl & Grand Concourse was      75-48-42-43-79-46-88-81.
City sources say that there are the nearest units                75-48-88-42-43-46-79-81-95-45-92-93-90-50
all 2 miles away or less. (most likely air miles)
Now remember that 88 is restricted getting there because of the cliff between the Grand Concourse & Webster Ave.
Based on city info the mileage for 75 is 0.26; 48-0.33; 88-0.77; 42-0.88; 43-0.93; 46-0.98; 79-1.1;
81-1.32; 95-1.39; 45-1.42; 92-1.77; 93-1.77; 90-1.92; 50-2 miles.

Pre L-56 & L-59 move into the West Bronx, I do not remember L-38 assigned with L-33 on any box along the Grand Concourse. L-37 came as far south as East 183 St. for boxes on Ryer or Valentine Ave.  L-27 went from E-45 quarters to the Harlem Rv. on the box & as far north as Davidson Ave & West 184 St.  L-36 was 2nd due at University Ave. & West 183 St.

Also the first 9 or 10 engines are the closest units. There are fire companies over the years that have no normal multiple alarm assignments or would only respond up to lets say a second or third alarm before they are omitted or bypassed. The next group of units were used to relocate into first alarm units, before the 4th alarm units were assigned. Remember that the number of engines on each alarm level has also increased over the years. I hope this helps you to understand FDNY's alarm assignments.
 
Atlas and 150,

Thanks for joining in this discussion.

150, I believe you are a dispatcher so your views are valuable.

Atlas, I don't know your background but you, too, seem to have some valuable stuff.

The idea here is to understand how this "system" works.

Before I break down the info you have sent let me ask you both to please define

"direct line / air miles" ...and...

" line of sight distance in 1,000 foot increments from a box location"

Thanks.




 
Sounds to me that Starfire uses linear programming analysis techniques based on certain algorithms that Rand developed which are generally fine in concept in these type of applications.  However, were the first alarm assignments through at least the second alarm assignments checked against the previous assignments pre-Starfire, to see if the assignments made sense particularly where  there could be overriding concerns such as terrain etc.  It would seem this type of fine tuning should be done manually by experienced FDNY personnel with institutional knowledge such as dispatchers.  I recognize that a variety of scenarios would be difficult to evaluate - however a basic assumption that nearby boxes in the area are not out could be assumed to test the reasonableness of the assignments developed by the Starfire program.  This would be one of the type of techniques used to tweak the program and improve its overall effectiveness. Logistically this takes time to do particularly when one considers the overall number of boxes in the City. However it seems that many well-informed people seem to know that these problems tend to be clustered in certain areas. Logic dictates that those areas be addressed first - like anything else in life you set priorities etc.

This is pretty much accepted practice in industries using these type of computer programs.  A similar application that I am very familiar with is the timing and coordination of traffic signals where a variety of computerized traffic engineering software programs have evolved over the years.  The results of the newer
timing programs are improving all the time but they still require fine tuning by "human hands" who know the operational flows on the street - for example certain blocks have a lot of so called "marginal friction" due to heavy volumes of parking activity/maneuvers which can impact vehicle through speeds etc. 

 
What some of you guys might not know was that FDNY response policies were never standardized borough or citywide. Different area had different responses. Example would be the northern regions of the Bronx. The first alarm only called for two engines & two ladders. This was in affect from Riverdale to Classon Point.  The second alarm saw the response of four engines & one ladder. Besides there were no 'fast trucks'.  This policy was also found in remote areas (bedroom communities) of the three other out lyings boroughs. 

Rand came to FDNY in the late 70's & early 80's studying what was going on within FDNY. When they released there first response plan there was a lot of problems with it. One that stood out was that Ladder 39 no longer responded to Jerome Ave & East 233 St. which was a few blocks from their former station. Engine 52 & Ladder 52 were first due on the first alarm to that location. The main problem was how did the 'twins' get around Van Courtland Park. This was based on 'blocks'  that were placed into the computer that generated the first set of computer responses. Based on these blocks & other considerations, Rand developed the idea that associated a cost factor with responses & required coverage of a company. Example - if Engine 70 was to respond to the Pelham Bay community on the second or third alarm what would it cost to protect their isolated first due area. In this case, Engine 70 has no other boxes to respond to other then first due on the first alarm. If the cost level was too high, it was better to bypass that unit. Good idea!  This played a major part for the companies that are along the Queens - Nassau border line too.

However some of there ideas was quickly placed into operation by FDNY & one was in the busy areas to hold the last due engine on the first alarm in reserve for another call. Rand also got FDNY to standardize alarm levels. There were boxes in The Bronx that had 5 or 6 engine companies assigned on the upper multiple alarm level based on what the hazard was in the area of that box. The levels vary box by box. Manhattan all way has a high number of units responding into the lower area of the borough.

Rand also produced the programs where there was a reduced response to boxes based on its local & previous incident history. Some of these boxes saw the response level on the first alarm drop to 1 & 1 or 2 & 1 only if it was a pulled fire box. No battalion chief was assigned. This was done to increase the availability of units borough wide. Remember that pre-war years The Bronx only had 7 battalions!

Years back when some of the engine companies had hose wagons assigned to them, the public always counted them as a separate company.

Pre-Rand all FDNY assignment cards were figured out by hand. The Alarm Assignment Unit had a large full time crew that measured all the assignments. As the crew changed the new unit members never knew why certain companies had restrictions placed on them. Engine 68 only responded on the first alarm due to their area, hills & dead-ended streets, but Ladder 49 responded without any restrictions. Engine 72 today only responds first, second, or third due on the first alarm because of the Satellite Unit. On the other hand, over the bridge in Queens Engine 273 was restricted because Engine 272 (also in Flushing) was eliminated. 

Over the years Engine 295 & 297 were also restricted because of there area, but also for the elimination of 
Engine 296.

Years back there was also a term used by FDNY & that was a 'KEY COMPANY'. If that company was out, the job wanted them covered. Today unless there was a change in the last few years, responses are based on a response neighbor, two of the same type of units. Years back prearranged relocations were printed on the assignment cards & automatically implemented by the alarm level disregarding other ongoing incidents. Years back, they only planned for one incident happening at a time. Sometimes moving two or three companies to cover one unit. Example Engine B went to Engine E who relocated to Engine M who was fighting a fire. 

Now back to The Bronx, I saw over the years assignments change because of a company thinking that they could beat another company into a box. Air Miles might work for helicopters, but does not work for fire apparatus. Computer generated assignments must take traffic flow into consideration. Highway responses can not & should not be considered. Street responses only! Highways are not a dependable route. Who should be second due on City Island?

I hate to say this, but one think that you should realize that FDNY is based on Old Fashion traditions. The same hold true for Alarm Assignments. The original Bronx was south of Kingsbridge Rd & East Fordham Rd to the Bronx River. Then the Bronx River south to the East River. The East Bronx & North Bronx were areas annex from Westchester Counties. The old assignment cards for this area had major discrepancies on them.

Two adjacent units normally did not respond on the same upper alarm levels because it would leave too large of an area unprotected for too long of a time. Some of the response policy changes has eliminated  these problems.

I been around for a good number of years having buffed the job & also being a member of FDNY. What I write about is things that I know about. Not saying that all was 100% perfect, but telling you the way it was & maybe why.  If I was to tell you that Engines 41, 42, 43, 48, 50, 61, 62, 64, 71, 79, 81, 83, 90, & 97  did most of the Bronx relocating. This was pre-Engine 66 & 72. Also not counted was Engine 85.

Please enjoy the information. You see over my years other people took the time out to explain to me why!
 
I would like to address one issue that I failed to do on my previous reply:

[i]This is pretty much accepted practice in industries using these type of computer programs.  A similar application that I am very familiar with is the timing and coordination of traffic signals where a variety of computerized traffic engineering software programs have evolved over the years.  The results of the newer
timing programs are improving all the time but they still require fine tuning by "human hands" who know the operational flows on the street - for example certain blocks have a lot of so called "marginal friction" due to heavy volumes of parking activity/maneuvers which can impact vehicle through speeds etc. [/i]


I agree with the comments made. Stop & think about the time the 52's would save if there was a traffic control program near their fire house. If they could travel south on the northbound service road to Manhattan College Pkwy, they might be able to save a minute to 90 second on runs south of their quarters. We do not have a traffic light control program here in NYC. We can use it there & also in Manhattan so that fire companies like Engine 65 for example could drive to 5th Ave & head south. Or Engine 46 & ladder 27 can reach  Park Ave to travel north bound or for one or two blocks along Washington Ave.

But instead the city is spending money to make it harder for emergency units to respond. 
 
Also ENG*324 only has First alarm assignments due to Sattelite *4 ....i know you mentioned ENG*72 having the same assignment. I would imagine all Sat/Eng s would be the same.
 
Yes it is for all of the Maxie-Water units. I used 72 Engine only as an example.  There are also restrictions on using certain companies to relocate & mandates to cover other companies. These list also includes ladder companies. 
 
My heads spinning!  ;)

Thanks for the wealth of information on the subject brothers. I just need to read it a couple 100 more times to take it all in!
 
Atlas,

So " direct line/air miles" means "as the crow flies", or helicopter as you put it?

Is that how STARFIRE determines unit distance from a box?

150 has stated that it is based on "line of site disance in 1000 foot increments from the box". I am not sure what that means.

My question had a simple intent. I asked you both to define your terms first so we can address the "system". What are the parameters we are working with?

Let's be careful here. This is a good discussion, to my mind, anyway, and the replies seem to indicate that others are interested also. But we are trying to eat an elephant, and the only way to do that is one bite at at time.

No need to get into the RAND system just yet, unless it is still untilized in STARFIRE. Nor the older way of determining box assignments just yet. Let's establish first what we have now and then maybe do some comparison.

 
When I say line of sight, I mean straight line distance. You could call it as the crow flies, air miles, it was taught to me that it's straight line from the box in 1000 foot increments.
 
150, thanks for the reply.

"Line of site, in xx increments" is similar to terminology used in surveying techniques. It renders an image of surveyors out on the streets with their theodolites measuring distances from points A to B, then B to C, etc., when line of site is interrupted.(I have done a lot of surveying in my life.) It could be one factor, an accurate one, to determine response patterns. That's why I was asking for a definition.

Atlas, your statement...

"Based on city info the mileage for 75 is 0.26; 48-0.33; 88-0.77; 42-0.88; 43-0.93; 46-0.98; 79-1.1; 81-1.32; 95-1.39; 45-1.42; 92-1.77; 93-1.77; 90-1.92; 50-2 miles."

...sounds official. What is the source of "city info"? And to which location, 183d or Field Place and the Concourse?


 
3511 - I assume you ran the gun (theodolite) and didn't hold the Philadelphia rod - just kidding -  did a little surveying over the years myself - actually did some on Manhattan College Pkwy as part of my surveying requirements at Manhattan College - great place with many fond memories!! Surveying is a little different now with digital equipment and lasers etc.

I like this thread and am interested like yourself in flushing out some more information on the Starfire system.
 
efd,

Did 'em both...and cut line through the Everglades with a machete...talk about line of sight! And then added it all up to a "close".

Glad this is of interest to you, too. No agenda here, just trying to figure out the "STARFIRE system".

Hopefully we can glean some more info. Stay tuned.
 
Back
Top