Keep arrivals and dock schedules in step.
DockTrellis is dock-appointment scheduling software for warehouses and distribution centers. It combines your appointment records with arrival updates from drivers and dispatchers, flags the conflicts a late truck will cause, and lays out schedule adjustments for a manager to compare and approve.
One late truck can unravel a whole dock schedule.
Most warehouses book dock appointments in advance, then manage the day by phone, radio and spreadsheet. When a truck runs late, someone has to work out what it collides with, which doors are free, and who to tell — usually while the yard is already filling up.
How it is handled today
- 01
Appointments live in one place, arrival news in another
The booked schedule sits in a calendar, spreadsheet or portal. ETA changes arrive as calls, texts and emails to whoever happens to pick up.
- 02
Conflicts are found by eye
A coordinator scans the board to see which later appointments a delayed truck will collide with — often only after the truck is at the gate.
- 03
Unloading time is a guess
Slots are booked at a fixed length regardless of load. A mixed-SKU inbound with 26 pallets and a 10-pallet top-up get the same hour.
- 04
Reshuffles happen without a record
Doors get swapped on the fly. Nobody can easily see later why a slot moved, who approved it, or whether it worked.
What a single delay sets off
- 07:52Dispatcher calls: the 08:30 inbound is stuck in traffic, new ETA about 09:40.
- 09:40Truck arrives into the 09:30 slot already booked on the same door.
- 09:45Next carrier waits in the yard. Detention clock starts. Gate queue grows.
- 10:30Unloading crew is split across doors; labor plans for the afternoon slip.
- LaterOutbound loads waiting on that inventory leave late. Nobody has a clean record of what happened.
The information needed to avoid most of this — the booked schedule, the updated ETA, rough unload times and which doors are free — usually exists. It just isn't in one place, early enough, for someone to make a decision. That is the gap DockTrellis is built for.
From an arrival update to an approved schedule change.
Below is an illustrative planning view for a fictional distribution center. A dispatcher reports a delayed truck, DockTrellis checks it against the dock schedule, finds two available docks and proposes adjustments. The manager compares options and decides. Try the options.
- Booked
- 08:30 · D3
- Reported ETA
- 09:40
- Variance
- +70 min
- Booked slot
- 60 min
Dispatcher reports APT-1182 will arrive at 09:40, 70 minutes late.
Rules check overlaps on D3; unload duration estimated from the load.
One conflict found; D5 and D2 are free; three adjustments ranked.
Manager approves one option. The change and the reason are logged.
Six steps, with a person making the final call.
DockTrellis does the checking and comparison work. Managers keep control of the schedule.
Import appointments
Upload a CSV export of the day's or week's dock bookings, or add them in the calendar.
Collect updates
Drivers and dispatchers send arrival updates through a simple link; coordinators can log calls manually.
Estimate delays & durations
Each appointment gets a delay-risk indicator and an expected unloading time, with the basis shown.
Identify conflicts
Overlaps, double-bookings and slots too short for the load are flagged on the dock board.
Recommend adjustments
Options to move doors or times are ranked and shown side by side, with their knock-on effects.
Manager approves
Nothing changes until a manager accepts. Every decision is recorded with who, when and why.
Built around the dock board, not a dashboard of charts.
Appointment calendar and dock assignments
See every booked appointment by door and by hour, with carrier, load type and status on one board.
Why it matters: coordinators stop reconciling a spreadsheet, a whiteboard and a phone log to know what is happening at each door.
CSV appointment imports
Bring in bookings from an existing yard system, carrier portal or spreadsheet. Columns are mapped once and validated on every import.
Why it matters: a team can start without an integration project or replacing the system they already book with.
Driver and dispatcher arrival updates
A no-login link lets drivers or dispatchers report a new ETA. Coordinators can also log updates taken by phone.
Why it matters: delays reach the schedule as structured data, not as a message one person heard.
Delay-risk indicators
Each upcoming appointment shows whether it is on time, at risk or late, based on reported ETAs and how far the slot has already slipped.
Why it matters: managers see which trucks need attention before they reach the gate.
Predicted unloading durations
Estimate how long a load will actually occupy a door from pallet count, load type and door — shown as a range, with its basis.
Why it matters: a 60-minute slot booked for a 90-minute unload is a conflict waiting to happen.
Schedule-conflict detection
Automatically flags overlapping appointments, double-booked doors and slots too short for the expected unload.
Why it matters: conflicts are visible as soon as an update arrives, not when two trucks meet at one door.
Suggested changes for manager approval
Proposes door and time changes, ranks them by knock-on impact, and applies one only when a manager approves it.
Why it matters: decisions are faster, but stay with the people accountable for the dock.
For the people who own the dock schedule.
DockTrellis is focused on inbound and outbound dock operations at single and multi-door sites.
Keep the yard and doors moving
- Their problem
- Late trucks and early arrivals pile up in the yard while doors sit idle or double-booked.
- How DockTrellis helps
- A live dock board with delay risk and free-door options, so trailers can be directed to an available door quickly.
Replan without starting over
- Their problem
- They rebuild the schedule by hand each time a carrier calls, and labor plans drift with it.
- How DockTrellis helps
- Conflicts are found automatically and adjustments come pre-compared, with expected unload times to plan crews against.
Know what the site can absorb
- Their problem
- They learn about dock bottlenecks late and have little visibility into how a delay was handled.
- How DockTrellis helps
- One place to send arrival updates and see approved changes, with a record of each decision.
Everyday dock situations DockTrellis is designed for.
A truck reports it will miss its slot
The update is logged, the conflict with the next booking is flagged, and free doors that fit the expected unload are suggested for approval.
A morning of back-to-back inbound loads
Imported appointments are checked against estimated unload times so that slots too short for the load are spotted before the shift starts.
An early arrival with a free door
Coordinators can see whether pulling a truck forward would clear the yard without colliding with later bookings.
Reviewing why the schedule changed
Every approved or rejected suggestion is recorded with the update that triggered it, so teams can review what happened after the day ends.
Rules first. Forecasting when the data earns it.
Most sites starting out have limited clean appointment history. DockTrellis begins with transparent, rule-based scheduling and only introduces predictive models once they demonstrably beat simple baselines on a site's own data.
Rule-based scheduling
- Conflict detection from booked times, reported ETAs and door assignments.
- Unload duration estimated from configurable rules, e.g. minutes per pallet by load type.
- Delay risk from the gap between reported ETA and slot start.
- Suggestions ranked by simple, explainable criteria: fewest affected appointments, buffer to the next booking.
Validated forecasting
- Models for unload duration and arrival delay trained on a site's historical appointments.
- Each model is compared against the rule-based baseline before it is switched on.
- If a model does not clearly improve accuracy, the baseline stays in use.
Every output is labeled by how it was produced
Deterministic
Overlaps, double-bookings and time arithmetic. Given the same inputs, always the same answer.
Estimated or model-assisted
Delay risk and unload durations. Shown as ranges with their basis, never presented as certain.
Human decision
Schedule changes. Only applied after a manager accepts them, and logged with who decided.
Designed around AWS.
DockTrellis is being built on AWS to store appointment data, process arrival updates as they come in, run scheduled planning jobs and, later, train and run forecasting models on historical appointments. The architecture is sized for early pilot workloads and is meant to grow with the number of sites, appointments and updates it handles.
| Service | Why DockTrellis needs it |
|---|---|
| Amazon S3 | Keeps the original CSV files and builds up the historical appointment dataset that forecasting is later trained on. |
| Amazon DynamoDB | Low-latency reads and writes for the live schedule: appointments, door status and every suggestion and decision. |
| API Gateway + Lambda | Arrival updates are bursty and event-driven; serverless functions process each update and re-check conflicts without idle servers. |
| Amazon EventBridge | Runs planning jobs on a schedule, such as checking the next shift's appointments for conflicts before it starts. |
| Amazon SageMaker AI | Training and batch prediction for delay and unload-duration models — introduced only when a site has enough history and models beat the rule-based baseline. |
| Amazon Cognito | Authenticates warehouse users and separates access by workspace. |
| Amazon SES | Sends the email notifications a team chooses to configure. |
| Amazon CloudWatch | Monitors API errors, function failures and scheduled-job health. |
Why the workload grows with each site.
Infrastructure needs rise in line with how many warehouses use DockTrellis and how busy their docks are — not all at once.
More sites and doors
Each warehouse adds its own appointments, door configuration and users, increasing stored records and API traffic.
More arrival updates
Every ETA change triggers a conflict re-check. Busy sites with many carriers generate frequent, bursty processing.
Growing history
Every completed appointment is added to the historical dataset, increasing storage and the size of training data.
Scheduled planning & forecasting
Pre-shift planning runs and batch predictions scale with the number of appointments being planned.
Schedule data handled carefully, decisions kept with people.
Authentication
Warehouse users sign in through Amazon Cognito. Driver update links are scoped to a single appointment and expire.
Access controls
Role-based permissions separate who can view schedules, log updates and approve changes.
Data isolation
Each customer's appointments and decisions are kept in a separate workspace and are not shared across customers.
Encryption
Data is designed to be encrypted in transit (TLS) and at rest using AWS-managed encryption for S3 and DynamoDB.
Audit trail
Every update, suggestion, approval and rejection is recorded with source, user and time.
Human review
DockTrellis never moves an appointment on its own. Estimates are labeled, and approvals require a person.
On estimates and certifications: delay-risk and unload-duration figures are estimates and can be wrong, particularly for sites with little history. DockTrellis does not currently hold security certifications such as SOC 2 or ISO 27001.
Subscriptions by warehouse and appointment volume.
DockTrellis is early-stage and there is no self-service signup yet. We are inviting a small number of warehouse teams to pilot the first release, and pricing is agreed with each pilot site.
Work with us on the first release
For a warehouse or distribution center that books dock appointments and deals with late arrivals regularly.
- Set up from your existing appointment CSV exports
- Arrival-update links for your drivers and dispatchers
- Conflict detection and suggested changes for approval
- Direct line to the founders for feedback and configuration
How subscriptions will be priced
After the pilot, plans are planned to be based on:
Published pricing will follow once the pilot phase is complete.
Build plan
Core scheduling
- CSV appointment imports
- Calendar and dock assignments
- Manual arrival updates
- Rule-based conflict detection
Suggestions & approvals
- Delay-risk indicators
- Rule-based unload estimates
- Ranked adjustments for manager approval
- Email notifications
Validated forecasting
- Delay and unload-duration models, tested against baselines
- Planned: GPS and transport-system integrations
Building practical scheduling software for the loading dock.
DockTrellis is an early-stage software company focused on one operational problem: keeping dock schedules workable when trucks don't arrive on time.
Late arrivals cost warehouses in detention, yard congestion, idle labor and late outbound loads. The information to manage them is usually spread across schedules, calls and messages. We are building a tool that pulls that together, does the checking work, and leaves the decision with the people running the dock.
Over time, we aim to make DockTrellis the planning layer between appointment booking and the dock floor — adding validated forecasting and integrations only where they make schedules more reliable.
- Building
- AI-assisted dock and appointment scheduling
- Focus
- Logistics, warehousing, wholesale distribution
- Users
- Yard managers, warehouse coordinators, transport planners
- Stage
- Early-stage; first release in development; recruiting pilot sites
- Model
- Subscriptions by warehouse and appointment volume
- Contact
- hello@docktrellis.com
Team
Stephen Ogunleye
Founder
Joel Hamani
Co-founder
Tell us how your docks run today.
If late arrivals regularly disrupt your dock schedule, we'd like to hear about it. Share a few details and we'll get back to you to talk through your site and whether a pilot makes sense.
- Email: hello@docktrellis.com
- Best fit: sites that book dock appointments and can export them to CSV.