Principles of Egress

The heart of fire safety is a single, stubborn idea: from any point a person can occupy in a building, there must be a continuous, protected path that leads all the way to a place of safety - and it must stay usable for long enough, in the dark and the smoke, for everyone to walk it. Everything else in fire design exists to make that path work or to buy it time. This is the means of escape (or means of egress), and it is the single most important thing an architect designs for life safety.

'Continuous' is the first discipline. The route cannot have a gap - a locked door, a dead-end, a step no one can see, a stair that stops at a floor and makes you cross an open office to find the next flight. You trace it as a line, unbroken, from the furthest desk to the open air. 'Protected' is the second: as you move along the path, it should become more safe, never less. You leave the room where a fire might start, pass into a corridor that resists smoke, then into a stair enclosure that fire and smoke are kept out of, and finally out of the building. The protection ratchets up; you are always moving toward greater safety, never stepping from a protected stair back into a smoke-logged space.

'A place of safety' is the destination, and it has a precise meaning. Ultimately it is the open air, well clear of the building, where people can disperse - the final place of safety. But a route often passes through relative safety first: a protected stair where fire and smoke cannot follow, a refuge, or in some buildings an adjacent fire compartment. Reaching relative safety stops the race against the fire's clock even before you are outside.

Hold the whole route in your head as you design, not just the exit door. The worst egress failures are not missing exits; they are good exits reached by a path that breaks, narrows, or fills with smoke halfway along. Frame the figures - how far, how protected, how wide - as typical guidance to verify against NBC 2016 Part 4 and the AHJ; the principle of the unbroken, ratcheting-up path is universal.

It helps to name the three parts of the route, because codes and fire engineers talk in these terms and each part has a different job. The exit access is the first leg - the travel within the occupied space, from wherever you are to the door of a protected exit: across the room, along the corridor, to the stair door. This is the least protected leg and usually the most dangerous, so it is kept short and as direct as the plan allows. The exit itself is the protected portion - most often a fire-resisting stair enclosure, or on the ground floor a protected passage or simply a final exit door. Once you are in the exit, you should be in relative safety, sealed from the fire. The exit discharge is the last leg, from the foot of the stair to the final place of safety in the open air - and it is easy to forget, yet a stair that discharges into a locked yard or back through the burning lobby has failed at the finish line.

Two properties govern every part: it must be available long enough and wide enough. Available means usable for the time people need - a function of how well smoke is kept out (detection buys early warning; protection and venting keep the route tenable). Wide enough means the route and its exits can pass the number of people relying on them fast enough, without a crush - exit capacity, which you size to the occupant load (Lesson 3.2 and Module 2).

A useful mental test on any plan is to stand, in your mind, at the furthest occupied point and walk out: count the legs, notice where protection begins, check the route never doubles back toward the fire or narrows to a pinch-point, and confirm it ends somewhere genuinely safe and open. If you cannot trace that line cleanly, neither can a frightened person in smoke.

The exact definitions, permitted lengths and the point at which 'protection' must legally begin differ by occupancy, height and jurisdiction - verify them in the current NBC Part 4 and with the AHJ. What does not change is the shape of the idea: access, then a protected exit, then a safe discharge.

A single escape route is a single point of failure, and fire is exactly the kind of hazard that blocks routes. So a foundational principle of egress is alternative escape: wherever people might be, they should generally have a choice of at least two independent ways out, leading in different directions, so that one fire cannot cut off every route at once. If smoke fills the corridor to your left, you turn right; if the stair at one end is involved, you reach the one at the other.

The word that matters is independent. Two exits that both funnel through the same lobby, or two stairs reached only by one corridor, are not true alternatives - a fire at that shared pinch-point takes both. Real alternative escape means the routes separate early and stay apart, ideally remote from one another, so no single event compromises both. On a floor plate this usually drives stairs toward opposite ends or corners; in a large room it drives exits to separate walls.

The dangerous opposite is the dead-end: a pocket of the plan from which you can travel in only one direction before reaching a point of choice. A dead-end corridor, an inner room whose only way out is through an outer room, a seating area with exits on one side only - in each, a fire in the wrong place leaves no alternative and people are trapped. Dead-ends cannot always be eliminated, so codes limit how deep they may be; treat every dead-end you draw as a liability to justify, not a detail to ignore.

There are controlled exceptions. Small, low-occupancy, low-risk spaces are sometimes permitted a single exit, and some strategies rely on protected routes or phased evacuation rather than simple alternatives. But these are exceptions granted by the code and the AHJ, not defaults you assume - the exact dead-end limits, travel distances and single-exit allowances are binding figures to verify in NBC 2016 Part 4 for your occupancy and height, and on any complex building to confirm with a fire engineer. The instinct to build in is simple and never wrong to start from: give people a choice of direction.

Pull the principles together into a set of golden rules you can carry to any plan. First, keep the route short and direct - every extra metre and every turn is time lost and a chance to get lost in smoke. Second, give a choice of direction - alternative, independent escape, with dead-ends kept shallow. Third, protect the route and keep it usable - the further along, the safer; stairs and key corridors sealed from smoke, and the route tenable for the time people need. Fourth, keep it unobstructed and openable - no storage in corridors, no furniture narrowing an exit, and above all never a locked exit: the recurring killer in real fires is a final door chained or bolted against the people trying to leave. Fifth, make it obvious - clear, lit exit signage and emergency lighting so the way out is findable when the normal lights fail and smoke obscures everything (Lesson 3.3). Sixth, size it to the people - enough exits, wide enough, for the occupant load (Lesson 3.2). Seventh, land it somewhere genuinely safe - a real place of safety in the open, not a trap.

Behind the rules sits a human truth from Module 2: in an emergency people tend to leave the way they came in, by the familiar route, and they delay before they move. So the best-protected fire exit is useless if nobody knows it is there or it feels like a back-of-house door. Good egress design works with behaviour - routes that are intuitive, well-signed and part of daily circulation where possible.

These rules are the architect's to own from the first sketch, because they are spatial decisions that fix the plan. The numbers that make them precise - how short, how wide, how protected, how deep a dead-end - are code and fire-engineering matters that you verify against NBC 2016 Part 4 and the AHJ, and escalate to a fire engineer on anything large, tall or unusual. But the rules themselves are the instinct; learn to see any plan through them.