Fire department knowledge management is the discipline of making operational knowledge — SOPs, training materials, policies, incident histories — accessible to the people who need it, when they need it, in a form they can use. Most departments have more of this knowledge than they realize; the problem is that it lives in binders, shared drives, and the heads of senior personnel who will eventually retire. FlorianAI approaches fire department knowledge management by making existing documents queryable in plain English — so a lieutenant can get an accurate answer about a hazmat SOP at 0300 without waking up the hazmat coordinator.
The Knowledge Problem in Fire Service
Fire departments generate enormous amounts of operational knowledge over time. Standard operating procedures, emergency response guidelines, pre-fire plans, training records, incident after-action reports, apparatus maintenance histories, mutual aid agreements — the accumulated operational intelligence of a fire department is substantial.
The problem is not that this knowledge does not exist. The problem is that it is almost never organized in a way that makes it easily accessible at the moment it is needed. SOPs live in three-ring binders in the company commander’s office. Pre-fire plans are in the CAD system, if someone remembered to update them after the last site visit. The after-action from the propane tank fire three years ago that established how the department handles similar incidents is in a folder somewhere on the shared drive — if anyone remembers where.
Chief Erron Kinney of Norfolk, Massachusetts Fire Department has spoken directly about this problem: the gap is not in the knowledge itself but in its accessibility. A fire department can have excellent SOPs and still have personnel who, under operational pressure, default to their own experience rather than the documented procedure — simply because finding the procedure in the moment is slower than acting on memory. The goal of knowledge management is to close that gap.
What “Queryable” Means
The central concept in AI-assisted knowledge management is making documents queryable. Instead of searching for a document — opening the shared drive, navigating folder structures, searching for a filename — you ask a question and receive a direct answer drawn from the relevant documents.
“What is our protocol for a natural gas leak in a structure?” Instead of opening the SOPs folder and reading through the utility emergency section, an AI assistant searches the indexed documents and returns the relevant protocol steps, citing the SOP it came from. If the lieutenant needs more detail, they can ask follow-up questions. If they need to review the full document, the AI provides the link.
This is not a novel technology concept — it is the same approach used by enterprise knowledge management systems in corporate environments. What has changed recently is that the capability is now accessible to organizations that do not have enterprise IT departments, and that the language understanding is good enough to handle the domain-specific terminology of fire service operations without extensive configuration.
SOPs Are Only as Useful as Their Accessibility
The standard operating procedure that nobody reads is not protecting anyone. This is a known problem in fire service — departments invest significant effort in writing and maintaining SOPs, and then discover that personnel under operational stress default to personal experience because the SOP is not accessible in real time.
There are structural reasons for this. At 0300 on a working structure fire, no one is consulting a binder. The knowledge that drives the response has to be internalized through training, or it is not operationally available. AI cannot replace training. But it can change the relationship between documented procedures and operational behavior in several important ways.
During training: AI-assisted access to SOPs makes them more usable as training tools. When a firefighter can ask a question about a procedure and get an immediate, sourced answer, the barrier to engaging with documented procedures drops. Curiosity-driven learning — “what exactly is the protocol for X?” — becomes instantly accessible rather than requiring a supervisor lookup or a document search.
During pre-planning and briefings: Company officers preparing for a specific type of incident or a high-risk occupancy can query relevant SOPs, pre-fire plans, and historical incident data in a few minutes rather than assembling documents manually. That preparation improves performance.
After incidents: After-action discussions and training debrefs benefit from accessible knowledge of what the documented procedure was, so the gap between what was done and what the SOP says can be clearly identified and discussed.
The Institutional Memory Problem
Every fire department has people who function as walking institutional memory. The battalion chief who has been with the department for 30 years. The apparatus engineer who knows every quirk of every piece of equipment. The deputy chief who remembers what happened the last time the department tried a particular approach.
This knowledge is invaluable. It is also fragile. When that battalion chief retires, the institutional memory they carry does not transfer automatically to their successor. Some of it lives in documents — incident reports, training records, after-action reports — but much of it was never written down because it never needed to be; the person who held it was always available to answer the question.
AI-assisted knowledge management cannot replace the depth of experience that comes with 30 years in the fire service. But it can do two things that help departments manage the institutional memory problem. First, it can surface the documented knowledge that exists but is hard to find — making the organization less dependent on any individual person knowing where things are. Second, it creates a workflow for capturing knowledge in a form that persists — when an experienced officer provides guidance, that guidance can be documented in a structured way that makes it queryable for future personnel.
What Needs to Be Connected
For AI-assisted knowledge management to work, the relevant documents need to be accessible to the AI system. In practice, this means identifying where the department’s operational knowledge actually lives:
- SOPs and guidelines (often in PDF or Word documents on a shared drive or printed in binders)
- Pre-fire plans (often in CAD or a separate planning system)
- Training materials and records
- Mutual aid agreements and regional protocols
- Apparatus and equipment specifications
- After-action reports and incident histories
Most departments find that the majority of their operational knowledge is in digital form somewhere — the challenge is that it is distributed across multiple systems and storage locations. FlorianAI is designed to connect to these disparate sources and index them for natural language query, so personnel can find what they need regardless of where it is stored.
The Search vs. Ask Distinction
There is a meaningful difference between searching for a document and asking a question. Search returns documents that may contain the answer. Asking a question returns the answer, with a citation to the source document.
For operational use in fire service — where the need is often specific, the time is often constrained, and the environment is often not conducive to reading long documents — the “ask” model is significantly more useful than the “search” model. AI knowledge management tools built on modern language models can make this distinction; older document management systems cannot.
This is why the shift from “a place to store documents” to “a system you can ask questions of” represents a genuine operational improvement, not just a technical upgrade.