Hello all!

We're postgraduate students at NTUA, Greece and we've been assigned a project concerning an evacuation planning of an underground disposal area of dangerous materials in case of fire. A software simple but adequate to do the job would be welcome.

We await for your suggestions!

Hi ntua, welcome!

Am I right to assume that the evacuation model would need to be able to incorporate hazard data (e.g. data from fire/gas modelling software) to represent the effects of the dangerous materials burning? If so then there are a number of evacuation models which can do this e.g. STEPS, EXODUS, Pathfinder, FDS+Evac (and many more).

Do you have any specific requirements regarding the human behaviour i.e. what types of behaviour would you like to be represented?

Is funding an issue? The reason I ask is because if you don't have access to funds then there are a few free evacuation models which might be worth considering.

There is some literature mentioned at the top of the Models page which gives advice/reviews of some of the main evacuation models.

As you probably can imagine, funding is an issue. We are doing that project as part of a lesson of our last semester, so we need an educational (free) version of the software we will be using.

As fas as I am concerned, there's no specific need for analysing the hazards in the program, cause our main goal is to specify the exit points, and the time people need to reach them. Luckily, the structure we're analysing is occupied by 5-6 employees only, so there's no need for such a complex software as those used in heavily occupied subway stations etc.

I see. Maybe the best thing to do would be to try some of the freely available evacuation models to see if they have enough functionality for what you want. Some free available evacuation models include: FDS+Evac, TSEA (30 day free trial), MassEgress (online simplified version available: http://eig.stanford.edu/pengao/Egress/ )...and there are a number of others also.

I think also there are some more sophisticated evacuation models which have discounted prices for academic use but you would have to contact the model developers themselves.

Having said this, the situations which you have described seem quite simple and you can work out things like approximate maximum travel time etc using hand based calculations (there would perhaps be minimal interaction between occupants if there are only 5-6 people involved) using standard walking speeds etc.

Dear ntua,
I am just working on the comparison analysis of Evacuation Models for providing safety in underground spaces. So, I will be very happy to give you some suggestions in the choice of the model. Maybe FDS+Evac will be the solution for your purposes because it is free and it permits to easily create your scenarios. Well, of course there are many other good models, and the choice of the best model for modeling underground depends on the scenario you have to simulate..
If you need help in using FDS+Evac and in modeling scenarios, do not esitate to contact me! I am always happy to share information with people working in this topic.

Best regards,

Enrico

Dear gents,

I am using for the moment the software called LEGION. Do you know it? I think it is the best evacuation model.

Thamkis,

-------------------------------------------------------
Professor Abilolado Babakovisky

Abilolado Babakovisky, why do you think LEGION is the 'best'?

Dear Mikal,

This website is very good:

http://www.legion.com/case-studies

Goud exampples of projects using LEGION model. Please read the website.

Thamkis,

------------------------------------------------------------
Professor Abilolado Babakovisky

Which parts of the model or the website make you think it is 'the best'?

Dear Mikal,

The visuallization; the intthegration between people and vehicles is also good.

What is your experience in evacuation modells? What are the good models for you? Do you work with this models in your company?

thamkis,

-------------------------------------------------------
Professor Abilolado Babakovisky

Hi guys,

Just read the discussion.

Regarding the incorporation of fire hazards into the People Movement (PM) models, as far as I know (based on my experience in using some of the PM models); some of these models allow the user to do it manually or to import from a fire simulation model. For instance:

- STEPS (data from FDS);
- EXODUS (data from CFAST and smartfire);
- LEGION (data from FDS).

These 3 models I am familiar with. I have also used SIMULEX a long time ago, but I don’t think it does allow anymore fire hazard inputs, correct me if I am wrong. I just used pathfinder once, during the beta test; so I am not sure neither if allows the user to insert fire hazard. I should probably read the review provided in the following website :) : http://www.firemodelsurvey.com/EgressModels.html
I personally think that it is good when the PM model allows the user to insert data generated from FDS.

kind regards,

Rodrigo

p.s.: for me, PM models = evacuation models = egress models = pedestrian models
I have written few papers using this terminology though.

Hi everybody,
I think that to define the best model is not easy due to the fact that depends on the purpose you have. For underground structure I think that fire data importing is a fundamental feature. The second point is that you need to know very well a lot of models!

For Rodrigo,
Pathfinder at the moment does not allow to import any fire data.

enricoronchi, I agree.

I think it is the wrong question to be asking :"What is the best evacuation planning software?" or indeed to be saying "The best model is X" (typically representing/requiring a huge over simplification). Many models (and model developers for that matter!) have particular strengths in different areas and understanding. In addition to the 'fitness for purpose' argument, you could also argue how certain types of behaviour are implemented influences the quality of the model. Alas it seems to be a one aspect which appears to be overlooked by many users i.e. how is the behaviour implemented within the model. I think consulting model developers rather than looking on third party websites is probably always the best way to find out how a model works and what features it currently offers...if they are willing to share the information that is :) . If not then Christian Rogsch presented an interesting paper looking at how different types of behaviour are implemented at the last PED 2010 conference with the following reference:

"To See Behind the Curtain – A Methodical Approach to Identify Calculation Methods of Closed Source Evacuation Software Tools", Christian Rogsch and Wolfram Klingsch, PED 2010

(see here http://www.evacmod.net/?q=node/2130)

Thanks Enrico. I am planning to buy pathfinder, and I use FDS; so it would be nice if pathfinder would allow inserting fire hazards though. But never mind.

regards,

Rodrigo

For Michael,

I agree with you that users have to consult model developers to understand how a model works, but the only way to exactly understand what you can do with a model is of course trying it! Unfortunately not everybody have an academic point of view and they can not spend too much time testing models just like I can do working for an University.
Regarding how behaviours are implemented in a model I think that the first point is understand how deep is the reproduction of the behaviours (i.e. which human behaviour features you can directly or manually reproduce such as herding behaviours, smoke triggering evacuation, etc.) and then, how they are inserted within the model.
For example, I will just prensent a paper at INTERFLAM2010 regarding the way egress models reproduce human behaviours in underground environments.

For Rodrigo,
Maybe this topic that I started in the Pathfinder forum some weeks ago can be useful to understand the future development of Pathfinder regarding fire:
http://www.thunderheadeng.net/forum/index.php/topic,296.0.html

Regards,
Enrico

enricoronchi,

Since I'm going to Interflam2010 I look forward to seeing your talk!

With regards to your points that you mentioned, I agree that it is important to identify what behaviours can be represented/reproduced within a model (in addition to clearly stating ALL assumptions and configuration aspects used in the behaviour). Ways of representing human behaviour within models is becoming more and more complex and will require more skilled users in the future. I think if a person doesn't have time to learn how any model works or a model developer is unwilling to make it known, then they definitely should not be using the model to assess a structures safety. A lack of understanding as to how behaviour within a model works can translate into misuse or a 'blind faith' mentality.

This was broadly discussed at the previous PED2010 conference where it was mentioned that building regulators were being required to asses the validity of evacuation model results (produced by engineers) without having any knowledge of how the evacuation model worked. It was also mentioned at the last Human Behaviour in Fires Conference by one regulator in New York and he said that he simply did not have time to review how a model worked before assessing the model results validity.

The idea of a regulatory board which could accredit evacuation model users was proposed at PED2010 though I don't know how much progress has been made with this.

Rodrigo,
I was just talking with one of the Pathfinder developer and he said that they are currently working in display FDS results in Pathfinder 2010. It could be a very interesting upgrade.

Thanks Enrico for this information. Good to know that indeed.

cheers,

Rodrigo