Detailed Schedules
Priority and loading
Detailed scheduling assigns orders to resources in the short term. Priority rules: FCFS; SPT (minimizes average flow time and WIP but ignores due dates); EDD (minimizes maximum lateness); critical ratio = time remaining ÷ work remaining, where below 1 means behind. Finite loading respects capacity; infinite loading ignores it and leaves resolution to the planner. Forward scheduling starts at release and projects completion; backward scheduling starts at the due date and sets the latest release.
Constraints and lean scheduling
Theory of constraints: identify the bottleneck, exploit it, subordinate everything else, elevate it, repeat. Drum-buffer-rope schedules the bottleneck and protects it with buffers. Lean scheduling uses takt time (available time ÷ customer demand), level or mixed-model sequencing, pull kanban, and setup reduction (SMED) to shrink queues.
Traps
- Optimizing a non-bottleneck and creating excess WIP - local efficiency is not throughput.
- Assuming SPT meets due dates; it optimizes flow, not lateness.
- Treating 100% utilization at the bottleneck as a problem - it is the plan.
- Sizing kanban on average demand instead of demand during replenishment lead time.
- Confusing standard time with actual time when the process is still on a learning curve.
Worked example
Critical ratio: due in 12 days, 8 days of work remaining → CR = 12 ÷ 8 = 1.5, comfortably ahead. An order with CR = 0.8 needs expediting. Kanban sizing: demand during lead time 400, 10% safety, container 50 → (400 × 1.10) ÷ 50 = 8.8, so 9 kanbans. Rule of thumb: always round kanban counts up.