Public Demo
You are welcome to explore this solution. Prior to official go-live, you may use it freely — however, your data may be lost in future updates. No guarantees or liability are given. Use at your own discretion.
Intermediate

Auto-schedule a backlog of requests

Run the solver over a real backlog, read the preview like a diagnosis, handle the rejects, and commit with confidence.

Published

Outcome

A batch of unscheduled requests placed via the auto-schedule preview, every reject understood and either fixed or consciously deferred — and a repeatable routine for doing this weekly.

Before you begin

  • A backlog: several unscheduled requests with real requirements.
  • Scheduling settings configured for the site (working hours at minimum) — the solver can only plan inside configured availability (Scheduling and availability).
  • The Editor or Admin role. Background: Auto-scheduling.

Step 1 — Groom the backlog first

Five minutes here saves a noisy preview:

  1. On Utilization, filter the Requests panel to Unscheduled Only.
  2. Scan for requests with impossible constraints (a Latest End in the past) or requirements you know nothing satisfies — fix or defer them now.

The solver honors every constraint you recorded; stale constraints become rejects.

Step 2 — Run and read the preview

Press Auto-schedule unscheduled requests. In the Auto-schedule preview:

  1. Note the summary — Solver (OR-Tools CP-SAT, or Greedy as fallback), Scheduled vs Unscheduled counts.
  2. Scan Assignments for placements that are valid but unwise — the solver optimizes the model, and the model doesn’t know a customer visit makes Hall 2 a bad idea on Thursday.
  3. Read Unscheduled reason by reason. This is the valuable part:
Reason What it tells you
No compatible space The requirement set has no satisfier — fix capabilities or relax the requirement
Insufficient capacity Suitable resources are full — widen the window or add capacity
Blocked by existing assignments Existing schedule leaves no room — consider re-planning the blockers too
Invalid duration Constraints are tighter than the duration — the request contradicts itself
Solver limit reached Budget ran out — re-run, or place the rest by hand

Step 3 — Decide: apply, or fix and re-run

  • Preview looks right → Apply commits every assignment at once.
  • Model problems surfaced → Cancel, fix the data (capabilities, constraints, availability), and re-run. Previews are free; bad commits are not.

If the preview goes stale because someone else edited the schedule meanwhile, Orkyo refuses to apply it — close and re-run.

Step 4 — Sweep the remainder

After applying, whatever stayed unscheduled is now a short, understood list. Place the must-haves by hand (Schedule to… or drag), defer the rest explicitly. Check Insights → Conflicts afterwards — a clean run should add none.

Making it a routine

Weekly works well: groom → run → read rejects → apply → sweep. Watch the reject reasons over time — they are a running critique of your model, and a shrinking reject list means the model is converging on reality.