• Welcome to The Building Code Forum

    Your premier resource for building code knowledge.

    This forum remains free to the public thanks to the generous support of our Sawhorse Members and Corporate Sponsors. Their contributions help keep this community thriving and accessible.

    Want enhanced access to expert discussions and exclusive features? Learn more about the benefits here.

    Ready to upgrade? Log in and upgrade now.

Anyone used Egress Simulation modeling software on an assembly project?

Y

Yikes

SAWHORSE
Joined
Nov 2, 2009
Messages
3,951
Location
Southern California
I was wondering if anyone here has either utilized, or plan-checked a project that utilized performance-based crowd simulation / egress modeling software such as Pathfinder by Thunderhead? I saw an egress simulation on a recent video that analyzed the Station nightclub fire, and it extensively utilized crowd and fire modeling to show the failure.

At around the 10:20 mark, it discussed upcoming proposals to require assembly spaces to use performance-based crowd response modeling software as part of code compliance verification. Have you seen this simulation used on any projects you've come across? If so, at any time was the performance-based analysis part of a proposal for something less than required by prescriptive code?

 
Last edited:
I haven't heard of it. Unless its use is specifically allowed by the code as an alternate to the well-known prescriptive parameters, I have no interest.

The station club is no mystery. The secondary exits were locked, and some were concealed from view.
 
Yes.

I have used Pathfinder in combination with FDS quite a few times, whenever I had to deviate from prescriptive Code.


pathfinder.jpg




fds.jpg
 
Yes, there is now a big sign on the building, very interesting project peformance based design with very robust fire sprinkler in lew of passive protection in a high rise, smoke development layer for exiting distance
 
Tim Mailloux said:
I am more of a crayon guy.........120 colors and a built in sharpener.
Click to expand...
I just chew mine sharp...

I guess I would want to know just how the software works and if it has ever been real world verified.....Just a little leery on the fake world telling me how it would work in the real world...
 
In peformance based design, allowance in Massachusetts it is required the applicant pay for a 3rd party engineering firm to evaluate the PBD and the building, fire officials and the involved parties present the project to the appeals board for their approval to use the alternative methods.

We may also require test and reports an no cost to or community to support alternative meas and methods
 
TheCommish said:
In peformance based design, allowance in Massachusetts it is required the applicant pay for a 3rd party engineering firm to evaluate the PBD and the building, fire officials and the involved parties present the project to the appeals board for their approval to use the alternative methods.

We may also require test and reports an no cost to or community to support alternative meas and methods
Click to expand...

You can send an alternative material to a laboratory and burn a sample in a test furnace. How do you do a laboratory test of trapping 100 people in a burning building with inadequate exiting?
 
Yankee Chronicler said:
You can send an alternative material to a laboratory and burn a sample in a test furnace. How do you do a laboratory test of trapping 100 people in a burning building with inadequate exiting?
Click to expand...
Apparently, they’ve based their model on behavioral studies of how people respond to queues, circulation paths, etc. Here’s a study comparing their model to NFPA for exit time off a transit platform:

Comparing NFPA 130, SFPE, and Pathfinder

Evacuation simulation of a center-platform station using Pathfinder, then comparing results with NFPA 130 and SFPE calculations.
support.thunderheadeng.com support.thunderheadeng.com
Their model shows that people will first head to the main exit, then as they see a queue form, they will decide to double back and find another exit. Time to complete clearing of the platform was 80 seconds longer on Pathfinder modeling vs. NFPA.
 
I have reviewed alternat solutions using these studies. A couple comments:

1. Humans are only unpredictable until there is a major life safety threat. Once a major life safety threat becomes known to the occupant, they behave in predictable ways.
2. In my experience, design professionals apply a much greater margin of error to these types approaches (A traditional error rate of 1% is increased to 10-20%).
3. When reviewed objectively, a lot of these approaches are "common sense". For example, one was used in a building with a walking track to eliminate one of the exit enclosures. Because occupant queuing was minimal at the exit, there were not the traditional delays that code expects for occupants getting from the floor area into the exit enclosure.

Like most tools, it is possible, but must be used by a qualified professional who knows how to use it.
 
tmurray said:
I have reviewed alternat solutions using these studies. A couple comments:

1. Humans are only unpredictable until there is a major life safety threat. Once a major life safety threat becomes known to the occupant, they behave in predictable ways.
2. In my experience, design professionals apply a much greater margin of error to these types approaches (A traditional error rate of 1% is increased to 10-20%).
3. When reviewed objectively, a lot of these approaches are "common sense". For example, one was used in a building with a walking track to eliminate one of the exit enclosures. Because occupant queuing was minimal at the exit, there were not the traditional delays that code expects for occupants getting from the floor area into the exit enclosure.

Like most tools, it is possible, but must be used by a qualified professional who knows how to use it.
Click to expand...

So who is a "qualified professional," and how should/must such software be used?

The egress size and capacity requirements in the codes have been developed pragmatically over many decades of real world experience. They work -- as long as people don't chain the exit doors closed, like at the Station Club or the Tyson chicken processing plant. The prescriptive requirements in the codes are easy to understand, easy to design to, and easy to review. So why do we need fancy software modeling?

Answer: because developers don't want to have to spend money on exits, so they're constantly looking for any excuse to omit a door or a stairway, or to spread them farther apart. Follow the money.

I haven't played with egress modeling software but, a number of years ago, I worked with smoke propagation modeling software. I had several discussions with a couple of more intelligent fire marshals about it. They were aware of it, and they didn't like it. They felt that it was better at predicting the behavior of "cold smoke" than smoke from a real fire. If you're not familiar with cold smoke, that's what you get when you set off a smoke grenade or smoke canister for testing a smoke evacuation system. The fire marshals agreed that testing with cold smoke is better than nothing, but they also agreed that cold smoke doesn't behave like real smoke in a real fire.

Let's say I'm not a fan of using software modeling to replace what we know works.
 
Yankee Chronicler said:
So who is a "qualified professional," and how should/must such software be used?
Click to expand...
The ones I have seen are from engineers with a masters degree in fire protection engineering. Largely they will include what software they have selected and why. How they are using it, any assumptions made, and if there are any modifications to the software/model.

On the statement of money, yeah sure. I don't really care if it saves them money or not. I care about the performance.
 
Yankee Chronicler said:
So you're saying that panicky is predictable? I guess panic is predictable, but the behavior in panic mode isn't very predictable, except in very general terms. :cool:
Click to expand...
If you watch a video of a sudden incident that threatens people's lives, everyone does the same thing at first. They freeze. The amount of time they stay frozen is based on how fast they process the threat and how they deal with it. It is important to point out that this threat is not being processed in the conscious mind, it is subconscious.

The next thing they will do is one of the following; stay frozen (imminent death/major injury/trauma, shut down of cognitive process to protect the consciousness), faun (convince the threat you are friendly - probably not the best idea in a fire), flee, fight.

That's it.

The only real option there is flee. So how do they flee? Well, we know ~90% of people head back towards the way they came in. We know as queues get longer, people's conscious brain start to take over and they start to evaluate if there are better options.

All the exits in the world will not save the person who stays frozen. It won't help the one trying to make friends with the fire. It won't help the guy trying to fight the fire.

When looking into how the brain processes information ahead of making a decision and how those decisions are influenced, I would recommend reading "Thinking Fast and Slow" by Dr. Daniel Kahneman. He won a Nobel prize for the underlying theories discussed in the book.
 
Yikes said:
Their model shows that people will first head to the main exit, then as they see a queue form, they will decide to double back and find another exit. Time to complete clearing of the platform was 80 seconds longer on Pathfinder modeling vs. NFPA.
Click to expand...
Part of the reason we bump it to 2/3...

1030.2 Assembly Main Exit​

Pursuant to section 29-381a of the Connecticut General Statutes, in a building, room or space used for assembly purposes and provided with a single main entrance/exit, the main exit shall be of sufficient width to accommodate not less than two-thirds of the occupant load, but such width shall not be less than the total required width of all means of egress leading to the exit. This section applies to Group A occupancies that are newly constructed, have an increase in the number of occupants by addition or alteration, or are created by change of occupancy. Where the building is classified as a Group A occupancy, the main exit shall front on at least one street or an unoccupied space of not less than 10 feet (3048 mm) in width that adjoins a street or public way. In a building, room or space used for assembly purposes where there is no well-defined main entrance/exit or where multiple main entrances/exits are provided, exits shall be permitted to be distributed around the perimeter of the building provided that the total width of egress is not less than 100 percent of the required width.
 
steveray said:
Part of the reason we bump it to 2/3...

1030.2 Assembly Main Exit​

Pursuant to section 29-381a of the Connecticut General Statutes, in a building, room or space used for assembly purposes and provided with a single main entrance/exit, the main exit shall be of sufficient width to accommodate not less than two-thirds of the occupant load, but such width shall not be less than the total required width of all means of egress leading to the exit. This section applies to Group A occupancies that are newly constructed, have an increase in the number of occupants by addition or alteration, or are created by change of occupancy. Where the building is classified as a Group A occupancy, the main exit shall front on at least one street or an unoccupied space of not less than 10 feet (3048 mm) in width that adjoins a street or public way. In a building, room or space used for assembly purposes where there is no well-defined main entrance/exit or where multiple main entrances/exits are provided, exits shall be permitted to be distributed around the perimeter of the building provided that the total width of egress is not less than 100 percent of the required width.
Click to expand...

Even under the ICC version of the 2021 IBC, the main entrance/exit in a place of assembly has to provide capacity for at least half the occupant load, irrespective of how many other exits are provided or their capacity. In assembly venues that have a main entrance/exit, this generally means that the total egress capacity will exceed 100% of the occupant load, sometimes significantly.

1030.2 Assembly main exit. A building, room or space used
for assembly purposes that has an occupant load of greater
than 300 and is provided with a main exit, that main exit shall
be of sufficient capacity to accommodate not less than one-half
of the occupant load, but such capacity shall be not less than
the total required capacity of all means of egress leading to the
exit. Where the building is classified as a Group A occupancy,
the main exit shall front on not less than one street or an unoccupied
space of not less than 10 feet (3048 mm) in width that
adjoins a street or public way. In a building, room or space
used for assembly purposes where there is not a well-defined
main exit or where multiple main exits are provided, exits shall
be permitted to be distributed around the perimeter of the
building provided that the total capacity of egress is not less
than 100 percent of the required capacity.

1030.3 Assembly other exits. In addition to having access to a
main exit, each level in a building used for assembly purposes
having an occupant load greater than 300 and provided with a
main exit, shall be provided with additional means of egress
that shall provide an egress capacity for not less than one-half
of the total occupant load served by that level and shall
comply with Section 1007.1. In a building used for assembly
purposes where there is not a well-defined main exit or where
multiple main exits are provided, exits for each level shall be
permitted to be distributed around the perimeter of the building,
provided that the total width of egress is not less than 100
percent of the required width.
Click to expand...
 
tmurray said:
If you watch a video of a sudden incident that threatens people's lives, everyone does the same thing at first. They freeze. The amount of time they stay frozen is based on how fast they process the threat and how they deal with it. It is important to point out that this threat is not being processed in the conscious mind, it is subconscious.

The next thing they will do is one of the following; stay frozen (imminent death/major injury/trauma, shut down of cognitive process to protect the consciousness), faun (convince the threat you are friendly - probably not the best idea in a fire), flee, fight.

That's it.

The only real option there is flee. So how do they flee? Well, we know ~90% of people head back towards the way they came in. We know as queues get longer, people's conscious brain start to take over and they start to evaluate if there are better options.

All the exits in the world will not save the person who stays frozen. It won't help the one trying to make friends with the fire. It won't help the guy trying to fight the fire.

When looking into how the brain processes information ahead of making a decision and how those decisions are influenced, I would recommend reading "Thinking Fast and Slow" by Dr. Daniel Kahneman. He won a Nobel prize for the underlying theories discussed in the book.
Click to expand...
Some of us actually get the change to study real life public stimulus response.

If any of you have access to research-gate , a paper on emergency communications based on the behavior of unsuspected Group A building occupants is available here:


In my experience, prescriptive Life Safety design parameters are frequently too restrictive for any given commercial building.

For example, travel distance limits are based on conservative assumptions. If the actual ceiling height is over the assumed height by Code (all other parameters remaining the same) smoke will probably take longer to drop down to occupants' heads, so they have more time to evacuate. The travel distance could be increased over the prescriptive limits in that case.

On the other hand, people movement has a lot to do with their profile. A family will move at the pace of its slowest member.

It is things like that which drive the results of the model, so it is critical to have someone experienced in the team setting it up.

That's why performance based design must be carried out by someone acceptable to the AHJ.

An engineer with 4 years of experience (barely meeting the PE licensing requirements) who has a Master's degree on facade fire protection, would not cut it for me, even if they have purchased the most expensive software application.
 
I find it all fascinating. A look at their annual conference presentations infers a clot of factors maybe getting modeled into the simulation.
www.femtc.com

FEMTC 2024

The 7th Fire and Evacuation Modeling Technical Conference was a hybrid in-person and live-stream conference.
www.femtc.com
 
You must log in or register to reply here.
Back
Top