Can People Get Out in Time?
Using modeling to move beyond intuition and strengthen evacuation readiness
We’re using technology to answer the questions that matter most. When you can measure evacuation time, you stop hoping your plan will work and start seeing whether it will. - Leo Zlimen | CEO of Ladris
Every evacuation centers on one critical question: can people get out in time?
Whenever an incident forces people to leave their homes, neighborhoods, or cities, it sets off a race between two clocks: the time it takes people to reach safety, and the time the incident allows for them to get there.
The problem is that we rarely know how fast the first clock actually runs—or how much time evacuating a community truly requires.
As evacuation risk shifts from a rare disruption to a recurring stressor in many communities, this uncertainty creates a growing blind spot for the professionals responsible for preparing for disasters and managing complex response operations.
Yet recent advancements in artificial intelligence and modeling technology are beginning to change that. Today, public safety leaders, planners, and operators can move beyond intuition and experience alone and start putting measurable bounds around evacuation time. This shift supports more disciplined planning, clearer communication, and more informed decisions under pressure.
To explore what this shift looks like in practice, I spoke with Leo Zlimen, co-founder and CEO of Ladris, about how communities are using evacuation modeling to better understand risk before they are forced to confront it in real time.
The Blind Spot | We Don’t Know “the Number.”
One of the most important—and often overlooked—features of an evacuation is that it is not a standalone operation. Evacuation is a protective action taken in parallel with another incident: a wildfire advancing toward a community, a hazardous materials release spreading downwind, or a hurricane approaching the coast.
That distinction matters because evacuation is always constrained by an external incident with its own dynamics.
The time available clock is set by the incident itself. How long until a wildfire reaches the edge of a neighborhood? How long before a toxic plume moves over a city? How many hours remain before hurricane-force winds make travel unsafe?
These timelines are shaped by factors largely outside of our control—weather, topography, ignition point, population density, and the effectiveness of mitigation or suppression efforts. Small changes in any of these elements can have outsized effects: a slight shift in wind direction, a slight change in wind speeds, or an ignition point moving a few hundred yards can dramatically alter how much time is available.
As a result, the time-available clock is most useful once an incident is underway, when those variables collapse into the conditions that exist in that moment. Trying to establish a single, definitive baseline benchmark for time available in advance quickly turns into an infinite series of “what ifs,” offering limited value as a foundation for preparedness.
The more useful number is the time required. Knowing how long it actually takes to evacuate a neighborhood, corridor, or community provides a concrete way to quantify risk.
As Leo explained to me, “Knowing that number changes everything. It’s how you start to understand how bad ‘bad’ actually is.”
If evacuating a neighborhood takes three hours and the fire is projected to arrive in five, that is a fundamentally different problem than needing three hours when the fire may arrive in two. Neither scenario is comfortable, but the distinction directly shapes response decisions. Who needs to evacuate first? How should alerts be phased? Where should limited traffic control resources be prioritized? Are suppression efforts focused on stopping the fire’s advance, or on protecting evacuation routes to preserve time for people to get out?
While the time required to evacuate a community is not entirely controllable, it is highly influenceable. It reflects readiness: the clarity and operational usefulness of evacuation plans, the degree to which agencies and partners have trained and practiced together, and the availability of resources to implement those plans quickly and effectively.
Once this number is known, it establishes a baseline for performance. It moves communities beyond narrative confidence (“we’ve done this before”), institutional memory (“this worked last time”), professional intuition (“my gut says we have a few hours”), or—too often—no estimate at all. It replaces assumption with understanding, and guesswork with something leaders can actually work with.
How Do You Measure the Time Required?
For decades, evacuation planning has lived with a fundamental limitation: the first time strategies truly meet physics is often during a real-world incident. While some communities conduct large-scale evacuation drills, those rare and resource-intensive exercises usually provide only one chance to test assumptions. The true operational consequences of evacuation decisions—where congestion forms, how long movement takes, and who gets delayed—are still most often discovered when conditions are already changing and the margin for adjustment is limited.
Advances in AI are beginning to close that gap as they allow planners and operators to move beyond describing intent by testing assumptions and seeing what actually happens when a plan is implemented.
At its core, measuring evacuation time begins with simulation.
A simulation is not a prediction, and it’s not a promise. It is simply a structured way to take a set of assumptions—what you believe might happen and how you intend to respond—and observe the outcomes those assumptions produce. Instead of asking, “Will this work?”, simulation asks a more useful question: “If we do this, what happens as a result?”
As Leo demonstrated, in a platform like Ladris, agencies can adjust a wide range of inputs that meaningfully affect evacuation time. Traffic control strategies and the presence or absence of control points. Time of day and real-world traffic conditions. New housing developments that add vehicles without adding road capacity. Construction zones, lane closures, and chokepoints. Differences in how evacuation orders are issued, including phased or staggered alerts. Even human behavior: how long it takes people to decide to leave, how quickly they get into their cars, and how unevenly those decisions occur across a community.
The goal is not to eliminate uncertainty, but to represent it honestly—and still see how the system behaves.
As Leo explained to me, “Out of a simulation, you receive both a visualization and quantification of the outcome, for the assumptions the user puts in.”
That combination matters. Visualization makes consequences visible. Quantification makes them comparable.
Once this base case is established—this is how long evacuation takes under these assumptions—the work becomes iterative. Agencies can test changes one at a time and observe their effect. What happens if traffic control points are repositioned? If evacuation orders are issued earlier or in phases? If response resources are surged sooner? Each adjustment can be evaluated against the same baseline to see whether it meaningfully reduces evacuation time or simply shifts the problem elsewhere.
Leo likens this process to how risk is managed in other domains.
“This is the same approach used in every other field of risk management,” he told me. “You start with a base case, test changes, and see whether the outcome improves. Not doing that for operations that directly impact people’s lives doesn’t make much sense.”
This approach also changes how decisions are made during planning. In my experience, emergency operations planning discussions often gravitate toward the loudest voice or the most confident person in the room. Quieter perspectives—especially those that can’t be easily proven or are outranked by others—tend to fall away. Simulation shifts this dynamic as ideas no longer need to win on confidence alone. Proposed changes can be run, compared, refined, and run again.
That is the real shift underway. For the first time, evacuation planning can move beyond narrative confidence and professional instinct into a disciplined, testable process, and allow communities to explore consequences in a controlled environment.
Conclusion | A More Grounded Conversation
Public expectations around evacuation and disaster response are evolving. As evacuations become a more familiar feature of life in many communities, there is a growing desire for clarity. People want to understand what is possible, what is planned, and what tradeoffs are being made on their behalf.
Much of the tension that emerges during evacuations stems from a simple mismatch: assumptions about what can be done, expectations about what will be done, and the realities of time, resources, infrastructure, and human behavior. When those are misaligned, even well-intentioned decisions can appear confusing or inadequate after the fact.
Tools that help quantify evacuation time offer a way to close that gap. By establishing a shared, commonly understood picture of how evacuation actually unfolds, they support more consistent decision-making across agencies and more transparent communication with the public. Instead of relying on intuition alone, leaders can ground conversations in evidence and explain both capabilities and constraints with greater confidence.
This does not remove uncertainty or eliminate hard choices. But it does elevate the practice of evacuation planning. Knowing the time required allows communities to approach preparedness as a disciplined, testable capability that can be improved over time and discussed honestly before it is tested in real conditions.
That shift, from assumption to shared understanding, is one of the most meaningful advances now underway in evacuation readiness and disaster preparedness.
Enjoyed This Article? Pass It On.
If this article sparked ideas, share it with your network, a colleague, or on social media. Sharing is how we expand the community of professionals committed to staying left of bang.