Origin and Cause
CAUSE AND ORIGIN
Don't ask FOCUS to find the Cause and Origin of a fire or explosion. It can't be done! (almost never)
The 2021 edition of NFPA 921 states in Chapter 18 "The origin of a fire is one of the most important hypotheses that an investigator develops and tests during the investigation. Generally, if the origin cannot be determined, the cause cannot be determined, and generally, if the correct origin is not identified, the subsequent cause determination will also be incorrect."
ORIGIN AND CAUSE!
The origin of any incident; fire, explosion, building collapse, even a water leak, must be found before you can even think about what the cause of the incident was.
The area of fire origin is the room, or portion thereof, where the fire started. We can further narrow the area of origin down to the point of origin. That is, the specific point at which the fuel, the oxidizing agent, and the heat source all came together and started the fire.
The analysis of fire damage patterns will lead us to the point of origin. Fuel loads, ventilation, and the direction of the fire's progression can combine to confuse an investigator and his/her determination of the area and point of origin. At FOCUS we understand all the factors that play out in determining the point of origin and will use those factors to find the true point of origin.
Improper analysis of these patterns can result in an incorrect finding of the point, and in some cases even the area or room, of origin. FOCUS investigators have spent time fighting fires and understand how confusing the patterns may be at times. FOCUS investigators have constructed numerous training fires to study the fire's progression and the patterns left by the fire. We know how to get to the point of origin. We have also conducted experiments to determine if an ignition source was a competent enough ignition source to have started the fire.
NFPA 921 directs fire investigators to use the scientific method when conducting fire investigations. There are seven steps to the scientific method as described in NFPA921. Step ONE, recognize the need and step TWO, define the problem are pretty straightforward; We had a fire and we need to find out what caused it. Another integral part of this phase is the recognition of who may be an interested party in the investigation, especially from any civil aspect.
Step THREE is the collection of the data associated with the fire. We collect as much relevant data as possible. This includes data we gather from our own observations, our own experiments that we have conducted, interviews of witnesses and other interested parties, and even weather are just a few of the necessary points of data we need to collect. If an investigator enters the investigation with the idea of what happened (bias) he/she may not collect all the data necessary and only collect the data that supports the bias.
Step FOUR is the analysis of all the data collected. Is it valid data? The analysis of the data should be based on the investigator's knowledge, training, experience, and expertise. This step of analysis is where individual investigators need to remember to stay in their lane of expertise. For example; I'm not an electrical engineer, I know a great deal about electrical, but I'm not an electrical engineer. I may be able to decide if an electrical motor failed but I lack the expertise to decide why that motor failed; I'll leave that up to the electrical engineers.
Step FIVE. Developing hypotheses. Notice that it doesn't say the singular, hypothesis. The investigator needs to develop as many hypotheses as possible. If I can't develop any hypothesis then I haven't collected enough data and need to go back to step THREE.
In step SIX we will test each one of the hypothesis developed in step FIVE. In this testing step we will compare each one of the hypothesis to what we know as a body of scientific information (far too great to get into in this short explanation!) and we may rule out some of the hypotheses quickly, others will need to be more thoroughly tested. (The testing can be the really fun part or the very tedious part.) If, after testing there are no hypotheses left then the investigator needs to go back to step THREE. If, after all the testing, there are two or more hypotheses remaining then the fire cause is undetermined per NFPA 921 Section 188.8.131.52. Only when the investigator has one (and only one) hypothesis remaining can the investigation be moved to step SEVEN.
Step SEVEN is the selection of the final hypothesis. NFPA 921 Section 4.3.7 states "The investigator should document the facts that support the final hypothesis to the exclusion of all other reasonable hypotheses." This seventh step is totally dependent on the correct application of steps ONE through SIX. If the investigator has missed anything or not completed anything in the other steps then the result in this step seven is worthless.
Coulda, Woulda, Shoulda & DID
There are a great number of investigators who will find multiple possible sources of ignition and contend that any of them "could" have started the fire. This provides for a lot of speculation.
The scientific method tells us it is proper to develop multiple hypotheses during the typical investigation and the investigator must test each one. (we won't get into the test methods here) If a hypothesis can't be ruled out then it has to be counted as possible. When two or more hypotheses remain then the fire cause is undetermined per NFPA 184.108.40.206.
There is a big difference between "could" the thingamajig cause the fire and "did" the thingamajig cause the fire. FOCUS will make every effort to reduce all of the "could"s down to one single "did". Using science we will eliminate "could" and, again with science (part of our name), leave only one "did". Will there be times we can't find that one single "did"? YES! We will also let you know that we may never find that one "did" and at that point the best option may be to not spend more time and money running down a dead-end street.
Over a beer I'll tell you about the counsel that built his case around "could".