How to Spec a Door Schedule for a 24/7 Facility With Three Shifts
Most access control schedules are written for facilities that close. Doors unlock at 7:00 AM, lock at 6:00 PM, and everything outside that window is card-only. A 24/7 facility with three shifts breaks every one of those assumptions, and I keep seeing the same result: the integrator copies a template schedule from a nine-to-five office job, the facility runs on it for two weeks, and then the 11:00 PM shift change turns into a nightly pile-up of forty people badging through a single door that should have been unlocked for the transition window.
A door schedule for a three-shift operation is not a bigger version of an office schedule. It is a different design problem, and it starts with data you have to go collect, not assume.
Step 1: Map Real Shift Boundaries
The HR shift schedule and the actual movement of people are two different documents. On paper, first shift is 7:00 AM to 3:00 PM. In reality, people start arriving at 6:20 AM, the outgoing shift doesn't clear the building until 3:25 PM, and there is a 20 to 40 minute overlap where the building population nearly doubles.
Before you write a single schedule interval, pull two weeks of badge transaction history if the site has any existing system, or stand at the main entrance during two shift changes with a counter. You are looking for three numbers per boundary: when the first early arrival badges in, when the peak flow occurs, and when the last straggler from the outgoing shift leaves. Your unlock windows get built from those observed numbers plus margin — typically 30 minutes before the earliest arrival and 30 minutes after the last departure — not from the HR document.
Three shifts means three boundaries per day, and the overnight boundary is the one that gets botched. A schedule that runs 23:00 to 07:00 crosses midnight, and not every panel handles a midnight-spanning interval the same way. On Kantech EntraPass, for example, you build the overnight window as two intervals inside one schedule — 23:00 to 23:59 and 00:00 to 07:00 — rather than trusting a single wrapped interval. Whatever platform you are on, test the midnight crossing explicitly before cutover. I have watched a distribution center's dock doors relock at exactly 00:00 for three nights because nobody verified the wrap behavior.
Step 2: Holiday and Exception Handling
A 24/7 facility usually does not close on holidays — it runs a reduced crew. That is worse than closing, because the standard holiday feature on most panels substitutes a "holiday schedule" for the normal one, and the default holiday schedule is often "locked all day." Apply that blindly and the skeleton crew on Thanksgiving is locked out of the break room, the tool crib, and sometimes the exit-adjacent doors they need for material flow.
Spec the holiday behavior door by door. Perimeter doors typically go card-only for the full 24 hours on holidays. Interior circulation doors should follow a reduced-crew schedule, not a locked schedule. And critical infrastructure doors — electrical rooms, IT closets, sprinkler risers — should not have a schedule at all; they are card-only every hour of every day, so holidays change nothing.
Also decide who maintains the holiday calendar and write it into the handover documentation. Panels ship with the current year's holidays at best. If nobody owns the update, the system runs correct schedules for 12 months and then silently degrades in January.
Step 3: Group vs Door-Level Schedules
The temptation on a 60-door facility is to write one schedule per door. That produces 60 schedules that drift apart over years of one-off edits, and nobody can answer "what should the machine shop door be doing at 2:00 AM" without reading raw configuration.
Instead, define schedule classes and assign doors to them. In practice a three-shift industrial facility needs surprisingly few:
| Schedule class | Behavior | Typical doors |
|---|---|---|
| 24/7 card-only | Never unlocked; valid credential required always | Perimeter, IT, electrical, records |
| Shift-transition unlock | Unlocked 30 min around each shift boundary, card-only otherwise | Main employee entrance, turnstile bypass gate |
| Production-hours unlock | Unlocked continuously while lines run, card-only during shutdown windows | Interior circulation, break rooms |
| Business-hours unlock | Classic 8-to-5 unlock for the office wing | Lobby, front office suite |
| Escort/receiving | Unlocked only during scheduled receiving windows | Dock man-doors, vendor entrances |
Five classes cover most of the building. When operations changes a shift boundary, you edit one schedule and every door in the class follows. Door-level exceptions still happen, but they should be exceptions you can count on one hand, each with a comment field explaining why.
Step 4: Reader Mode Per Schedule
A schedule does more than lock and unlock. On most enterprise panels the schedule can also switch reader mode: card-only, card-plus-PIN, PIN-only, or unlocked. Use that. The pattern that has held up best for me on three-shift sites is card-only during staffed hours and card-plus-PIN on high-security doors during the overnight window, when there are fewer witnesses and the facility population is at its minimum. A stolen card is far more useful to someone at 3:00 AM than at 3:00 PM.
Be honest about the throughput cost: card-plus-PIN roughly doubles transaction time at the reader, from about 1.5 seconds to 3 or 4 seconds per person. That is irrelevant on an electrical room door that sees six events a night. It is a disaster on a main entrance during shift change, which is exactly why the mode switch is scheduled rather than permanent.
Worked Example: A Hospital Loading Dock
A concrete case from a regional hospital I worked on: the loading dock serves food service deliveries from 4:30 AM, general receiving from 7:00 AM to 3:00 PM, and linen pickup at 9:00 PM. The dock area has a roll-up (not on access control), a dock man-door, a corridor door into the service hallway, and a pharmacy receiving cage inside the dock.
The original spec had the man-door on a single 4:30 AM to 9:30 PM unlock. That meant the dock was effectively public for 17 hours a day, and the pharmacy cage — card-only, correctly — was the only barrier between a propped roll-up and a controlled-substance storage area.
The rebuilt schedule set: man-door unlocked 4:15–5:30 AM (food service window), card-only 5:30–7:00, unlocked 7:00 AM–3:15 PM (receiving), card-only until 8:45 PM, unlocked 8:45–9:45 PM (linen), card-only overnight. The corridor door into the hospital proper went card-only 24/7 — dock traffic badges into the building, no exceptions. The pharmacy cage went card-plus-PIN 24/7 with a two-person rule enforced in software. Total schedules created: three, all reused elsewhere in the facility. The security director can now answer "who could reach the pharmacy cage at 2:00 AM" in one sentence, which is the real test of a schedule design.
Schedule Complexity Diagnostic
Before you commit the design, run each door through this quick tree. First: does the door protect life-safety or high-value assets? If yes, it gets no unlock schedule — card-only or card-plus-PIN around the clock, and you stop there. Second: does the door see more than roughly 30 people in any 15-minute window? If yes, it needs an unlock window covering that surge, because a reader-per-person bottleneck at that volume backs up 10+ minutes and people start holding the door — which trains the workforce to defeat your system. Third: does anyone other than badged employees use the door — vendors, drivers, contractors? If yes, its unlock windows must align to the visitor process, not the shift schedule. Any door that falls through all three questions defaults to the production-hours class.
If you finish the exercise with more than about eight distinct schedules on a single-building site, the design is probably over-fit. Merge classes and handle the outliers with operator overrides instead.
Step 5: Override and Threat Response
Every schedule needs a documented override path, because the day the facility needs to deviate is the day nobody remembers how. Three overrides matter on a 24/7 site. Lockdown: a single operator action or input trigger that drops every unlocked door to card-only (or locked) regardless of schedule — and be explicit about whether card access still works during lockdown, because "locked" and "card-only" are very different postures during an incident. Extend: a supervisor-level action to hold a door unlocked past its window for a late delivery, ideally with an automatic expiry of 1 to 4 hours so the extension cannot become permanent by neglect. Weather/emergency unlock: for facilities in tornado or hurricane regions, a path to release specific interior shelter doors fast.
On Kantech hardware this maps cleanly to operator-level manual operations and input-triggered events in EntraPass; other platforms have equivalents. The platform matters less than the paperwork: the override procedure, with screenshots, printed and stored where the overnight supervisor can find it. The overnight shift will not call the integrator at 2:00 AM — they will prop the door.
One boundary the schedule must never touch: egress. Schedules govern entry; leaving the building stays free at all hours through the door hardware itself, regardless of what the access system thinks the time is. This sounds obvious until a well-meaning schedule change puts a delayed-egress or interlock behavior on a path the overnight skeleton crew uses to reach the parking lot, and the first fire drill on third shift discovers it. Every schedule revision on a 24/7 site should be walked against the egress plan for the lowest-population shift, because that is when a blocked path has no witnesses and no second exit habit. Where local code review is required for hardware like delayed egress, that review binds the schedule design too — a device that is compliant on the day-shift traffic pattern can be a violation on the overnight one.
Step 6: Audit and Documentation
The schedule you commission is not the schedule that exists in three years unless somebody audits it. Deliver a one-page schedule matrix — every door, its class, its reader modes by time band — as part of closeout, and put a recurring 6-month review on the maintenance contract. The review is simple: export the current schedule configuration, diff it against the documented matrix, and investigate every delta. One-off changes made during a crisis have a way of never getting reverted; the audit is how a 30-minute extension from last November stops being a permanent hole.
Also audit the transaction log against the schedule. Doors that show forced-open or held-open events clustered at a specific time of day are telling you the schedule fights the actual workflow. Fix the schedule; the workforce has already voted.
Deployment takeaway: Build three-shift door schedules from observed badge-flow data, not the HR calendar: pull two weeks of transactions, set unlock windows 30 minutes either side of real arrival and departure peaks, and verify midnight-spanning intervals on the actual panel before cutover. Collapse 60 doors into five or so schedule classes, schedule reader-mode escalation (card-plus-PIN overnight on high-security doors), give critical-asset doors no unlock schedule at all, and put a printed override procedure in the overnight supervisor's hands. Then diff the live configuration against the documented matrix every six months — the schedule that drifts silently is the one that gets exploited.
Where This Fits in a Deployment Program
Schedule design sits between hardware selection and commissioning, and it deserves its own line item in the project plan rather than an afternoon at the end. The panel platform you choose constrains what the schedules can express — interval counts, holiday sets, reader-mode switching — so review those capabilities while you are still selecting controllers from the Access Control catalog. If the site is standardizing on Kantech access control, the class-based approach above maps directly onto EntraPass schedules and door groups, and the same catalog covers the controllers and readers to build it out — browse all Kantech products to match controller capacity to your door count. If you are working through a three-shift schedule design and want a second set of eyes on the door matrix or the hardware list, send over the project details and door schedule draft — happy to help you spec it before the first controller ships.