Fog, friction and firefighting
Too much information, not enough information and bad information all contribute to uncertainty in the incident commander’s decision-making process
If you thought this article was about nozzles and hoses, you’ll be disappointed. Many articles and books cover those subjects adeptly. Instead, I’m here to discuss uncertainty (fog) and all the little things that make the simple task difficult (friction) at fires and other emergencies.
Two hundred years ago, a Prussian general named Karl von Clausewitz wrote the seminal book on the philosophy of war. Regarding the fog of war, he noted that too much information, not enough information, and bad information all contribute to uncertainty in the commander’s decision-making process. (More on too much information later.) On the other hand, friction is all about Murphy’s Law—if anything can go wrong, it will. Murphy is Edward A. Murphy, an engineer working on a U.S. Air Force project who once quipped, “If there’s more than one way to do a job, and one of those ways will result in disaster, then somebody will do it that way.”
Fog and Friction as Applied to the Fire Service
Take a moment to think about the number of things that can go wrong at an emergency incident. Next, think about which ones will force you to change your strategy and tactics on the fly. Here are some examples:
- An interlock fails and the apparatus operator can’t lift the aerial out of the bed; meanwhile people are hanging from the fourth-floor window.
- A hose line bursts at a critical moment.
- The next-due company gets into a crash or breaks down.
- There’s a misunderstood or ignored radio transmission.
Any one of these things can turn a simple room-and-contents fire into a nightmare. Therefore, how you, the incident commander, overcome the failure of little things is paramount to a successful conclusion.
Let us move on to the fog of firefighting. Of course, anyone who has been an incident commander always wishes they had more data, more information. But to what end? How much of that data and information can we turn into knowledge applicable to the incident at hand?
From Clausewitz to today, commanders struggle with both too much and too little information, as well as inadequate or incorrect information—all of which creates uncertainty for the incident commander. Bad or incorrect information is, arguably, a leading challenge to making sound command decisions, whether the incident commander is dealing with too much information or too little.
The Observe, Orient, Decide, Act (OODA) Loop
The OODA Loop is a mechanism for interpreting data and turning that data into actionable information to be used in the analysis of and the selected strategy for the incident. Colonel John Boyd (USAF) developed the OODA Loop to simplify a four-step cycle for a one-on-one dogfight in jet aircraft. The original OODA Loop assumes the characteristics of the jet, weapons, and training for each pilot are the same. Based on this assumption, Boyd concluded that to win, a pilot had to be quicker in deciding what action to take than the opponent. Thus, for our purposes, the OODA loop is a method of decision-making.
The elements of the OODA Loop are Observe, Orient, Decide, and Act. OODA elements are present in your 360-degree size-up. Simplified, the 360 looks something like this:
- OBSERVE: You, the assigned member, walk around the scene and note the factors influencing your overall strategy.
- ORIENT: You take the data gathered from the walk around, along with data received from dispatch or bystanders or the data terminal in your vehicle and combine that information with your knowledge of firefighting, including building construction, fire dynamics, and tactics.
- DECIDE: All these elements together inform your preferred strategy.
- ACT: You put your incident action plan into action.
For our purposes, using the OODA Loop is not a one-and-done process. Instead, operating at any incident requires the incident commander to continuously observe, orient, and decide whether the current strategy is working, requires modification, or needs to be completely changed. A common example is the decision to go from offensive to defensive operations. The incident commander observes the current strategy is not working and the fire is growing beyond the ability of the resources to contain and extinguish it, so they select a different strategy.
Boyd, Clausewitz and the Internet of Things (IoT)
Right about now, you may be thinking, how did he get from the 19th century to the 21st century? Boyd and Clausewitz were thinkers ahead of their time. You may have heard about big data, smart cities/devices, and artificial intelligence (AI). For our purposes, the internet of things (IoT) is the combination of the three. Loosely defined, big data is large amounts and varied information resources that use AI to provide knowledge and aid in decision making. AI is a form of computer science that uses machines to perform tasks that typically require human intelligence, usually through pattern recognition and problem-solving.
The IoT connects computing devices embedded in everyday devices through the internet, allowing them to send and receive data. AI is used to turn this data into knowledge. These devices may include everything from your smartwatch and automobile to building fire safety systems to fire apparatus. In the foreseeable future, firefighters and dispatchers will be able to receive real-time data from specific sprinkler zones, smoke detectors, and other building systems en route to the emergency.
In the meantime, several cities are using artificial intelligence to scour all types of data for patterns in data sets too large for a human to analyze. Atlanta, New Orleans, New York, and Pittsburgh have successfully created systems that use historical fire data, building and code enforcement data, and demographic data to identify and prioritize buildings for inspections. Much of this data can also be transmitted to the responding apparatus.
The Department of Homeland Security (DHS) and NASA’s Jet Propulsion Laboratory are field-testing location-tracking technology. The POINTER system uses AI to identify the location of firefighters in three dimensions. The system provides real-time information to a base station laptop at the command post. DHS expects this technology to be commercially available sometime in 2022. The drawback to the system is that it can cover only three floors of a building.
Let’s go back and visit with Clausewitz and Boyd and our bright and shiny information technology. This article is not a criticism of technology. Rather, it is a reminder to every incident commander that Murphy inhabits all that we do. Remember that too much information is not necessarily a good thing. Technology will provide us with enormous amounts of information that we must convert to knowledge to inform our decisions and actions. AI may help us process it but consider if there is a fault in the AI training data set, it will negatively affect the processing of live data it is using to assist you.
According to the OODA Loop, data and information run through the orientation (analysis) phase to make decisions. Dr. Roderick Wallace posits that it is at this point where AI may become overwhelmed with data. If you consider the building has a vote as well as contributes to Clausewitz’s friction, the randomness of failure combinations can lead to an AI system collapsing. Moreover, friction in two-sided conflicts has been around for millennia. John Boyd said, “[Combat] is dialectic in nature generating both disorder and order that emerges as a changing and expanding universe of mental concepts matched to a changing and expanding universe of observed reality.” Replace “combat” with “firefighting,” and you have a good description of what the incident commander faces.
The Fog Remains
Incident commanders need to comprehend and cope with firefighting space. To do this, we develop mental patterns of meaning. Then, we use these mental patterns to identify situational changes and respond to the changes. On the other hand, artificial intelligence uses algorithms to identify and respond to situational changes. To illustrate this point, consider a tiger. Show a child one or two pictures of a tiger, and the child will be able to pick out the tiger posed in any position on any background.
On the other hand, for AI algorithms to recognize the same tiger reliably, AI must have images of the tiger in every pose on every possible background. Now take this concept and apply it to the elements and dynamics of the fireground. The combinations are endless as the number of variables increases with the complexity of the incident.
I have little doubt that big data, AI and IoT will be helpful to the fire service. They will help us identify and manage high-risk properties and other community risk-reduction priorities. Internet-connected sensors will provide live building data and accuracy in tracking firefighters operating in an emergency. However, history shows that friction in dynamic systems has been around since the Battle of Thermopylae. Every incident commander will have to deal with fog and friction no matter how much technology surrounds the command post.
- Angerman W. (2004) Coming Full Circle with Boyd’s OODA Loop Ideas: An Analysis of Innovation Diffusion and Evolution. Washington, DC: Defense Technical Information Center. Accessed 11/1/21 at https://apps.dtic.mil/sti/citations/ADA425228.
- Carr C, ed. (2000) Karl von Clausewitz: On War. The Book of War. New York: Random House Modern Library: 257–973.
- Dayton Innovation Legacy. Edward A. Murphy, Jr.—The Engineer Who Accidentally Coined Murphy’s Law. Accessed 11/1/21 at http://www.daytoninnovationlegacy.org/murphy.html.
- Jay J. (2017) How we predicted building fires in Baton Rouge, LA — working version. Urban Analytics Blog. Cambridge, MA. Accessed 11/1/21 at https://scholar.harvard.edu/jonjay/blog/how-we-predicted-building-fires-baton-rouge-la-working-version.
- Madaio M. (2018) Fire Risk Analysis: Predicting Fire Risk to Prioritize Commercial Property Fire Inspections. Metro21: Smart Cities Institute. Pittsburgh: Carnegie Mellon University. Accessed 11/1/21 at https://www.cmu.edu/metro21/projects/fire-risk-analysis.html.
- Newcombe T. (2016) 4 Reasons Data Analytics Often Fail. Governing. Accessed 11/1/21 at https://www.governing.com/archive/gov-data-analytics-failure.html.
- Wallace R. (2018) Carl von Clausewitz, the Fog of War, and the AI Revolution: The Real World Is Not a Game of Go. Cham, Switzerland: Springer International Publishing, AG.