Probability of Detection (P.O.D.)

[Catu]

I've found it in several articles, have you heard of it?

It is supposed to be something that a mathematical formula to determine the probability of the dog to do a detection giving different factors.

Do you know something about it or do you have a link?

[k9sarneko]

I know of a mathematical formula for POD but cant remember exactly were you can find it. I was taught it at a seminar but since I have never used it I have forgotten how to do it. I believe it is calculated out from PLS, using variables for weather conditions, terrain ect. We are asked to provide to our POD's to the IC all the time on any area we are assigned. Given the time of day, terrain, weather conditions, age and mental and physical condition of the subject, each handler is asked to evaluate a POD. This is not as cut and dry as it sounds. POD is too high and the handler may commit too strongly to an area and not trust their dog to clear the area and move on, thereby losing valuable time and a valuable resourse of a fresher dog. POD too low and the handler may feel that they are "wasting" valuable time in the area and either fail to cover the entire assignment or not complete their assignment in a professional and thorough manner.

POD's should be thoroughly, carefully thought out keeping mind that any area assigned is not a waste of time, there may be a low POD to your area assigned but I have seen victims do some weird and wacky things so they must be all be checked properly.

[k9sarneko]

I wanted to add that I have had IC's who use the mathematical equation and others that have used their geographical knowledge of the area to be searched and years of experience to assign areas.

I do know that with water searches the coast guard will use mathmatcal equations that incorporate tidal flows ect to calculate areas that have a higher POD to place our HRD dogs. Sometimes even then your best resource is the local fishermen who know all the frequent catchment areas.

[jocoyn]

They use it around here a good bit -it is really a search management tool - it comes in most useful on searches with multiple operational period, where you are shifting to grid and air scent searches using search sectors.

I found decent article on it - I have an incident commanders pocket handbook and it has a lot of useful info for determining POD etc. NASAR has a book, [and course] Managing the Lost Person Incindent that really goes into it. I actually thought it was one of the better courses I have taken.

Probability of Success = Probability of Detection [ie probability you would find the subject if they were in the area] x Probability of Area [ie probability the subject is actually in the area]

A lot of dog teams have a hard time with POD for a dog - there are calculations like the stuff Graham-Hatch did
http://www.sar-dog.org.nz/pdf/SAR-DOG_PoD.pdf
But there are far too many variables in most search areas to apply his formulas, so it is harder to come up with [and the search team that is out there determines what they think their POD is, not the incident command] for a search dog than a simple grid search. So I guess it is more like your best guess given your experience in similar situations.

http://www.sarbc.org/probabil.html

The incident commanders field handbook has charts for calculating cumulative POD from multiple searches of same area. The combination of these tools allows you to assign resources with the highest probability of sucess. There is software that also does this.

[ladylaw203]

Forgive me but I have always gotten a kick out of that. It really sounds good,but at the end of the day, trust me, the minute anyone relys on that,the person will not be found. I have seen bodies do the darndest things and victims do not read books:-)

Also, FYI. With some exceptions, a body in the water will be within 100yds of where they went down. Seen this for 30yrs. I had a team call me last week on a search. Authorities were insisting because of current,yadda yadda that the body had done this or that. I told them back up,start your search where witnesses saw him go down. Guess what? There he was. We have a term in police work KISS Keep It Simple Stupid. Works Also take witness acounts with a grain of salt. Try their info for a while and if it doesn't pan out,chunk it. Folks are upset and their memories are not that great under pressure

[Catu]

Thanks!! this is exactly what I was looking for

To be honst... I don't iagine myself or anyone on my team doing formulas in the middle of a seach. I want it more like a didactic way to tech the people: Those are the factors that influenciate the search work and this is the importance they have.

Originally Posted By: ladylaw203Forgive me but I have always gotten a kick out of that. It really sounds good,but at the end of the day, trust me, the minute anyone relys on that,the person will not be found. I have seen bodies do the darndest things and victims do not read books:-)

I remember someone told me once "There is a moment you have to stop trying to be logical, and trying to be on the victims shoes, shut up your mind and let the dog do the job"

[ladylaw203]

Quote: I remember someone told me once "There is a moment you have to stop trying to be logical, and trying to be on the victims shoes, shut up your mind and let the dog do the job"

I will remember that one I was finding dope and HR long before we used dogs a lot so I use everything I have ever seen on a search . I always tell my narc dog handlers who keep insisting on their dog going over and over some area if they know where the dope is why do you have the dog?

[jocoyn]

The team north of us DOES use these tools but they don't rely on the "standard" book info to estimate where the highest probability areas are to search - they rely on experience from over 80 searches a year in their area for almost 30 years.

The people using these formulas are in IC and are the folks making team assignments. They put highly experienced searches in the IC to do the search management [some of these folks were doing this 30 years ago and are getting a bit arthritic to be huffing up and down the mountains]

The numbers do help when you have someone insisting to search and re-search and are that has been searched repeatedly or when a decision is made to stand down the search. They do help with pinpointing areas that have not had adequate coverage.

Yes victims move that always changes things. The dog can do its job but the dog has to be worked in an area where they can locate the victim.

A big issue I have seen with poorly managed searches is the lack of containment as the search area grows each hour the victim[if live] is out there. If we could get all those untrained community volunteers to do containment instead of throwing them in the woods as some departments do, there may be more successes.

[jocoyn]

It also helps with assigning which resources will search which areas. If there is a lower POA, you are going to look at your available resources and put you less experienced ones there and put the more experienced teams in the areas with highest POA.

[ladylaw203]

Well, all that sounds great, BUT. One should search everything equally and forget all the fancy stuff. Guess how many times I have been on a search and been told,oh gee, that guy would NOT have made it over there,no way. Guess where the body was found. Again, I have searched for too many crooks and too many bodies over 30 years to ever use that as a yardstick. Again, folks who are missing,suicidal etc do not follow what we consider a logical path
One way that I do try to clear an area is use a helo with FLIR. You can clear a lot of ground with that (no forrest canopy) and use the ground pounders and dogs in areas where FLIR is not useful