Production Scheduling and Sequencing: How APS Re-Sequences When Demand Changes
Production scheduling decides when each job runs and on which resource. Production sequencing decides the order of jobs on each resource. Sequencing is the part of scheduling that changes most when demand moves, because a new order, a cancelled order, or a date change all alter which job should run next.
When demand keeps changing, manufacturers keep the schedule stable where it has to be, and re-sequence the rest on a regular cycle against real capacity. That approach is called dynamic scheduling.
How should manufacturers sequence production when demand keeps changing?
Manufacturers should sequence production when demand keeps changing by freezing the near-term schedule, re-sequencing the flexible horizon on a rolling cycle, and choosing each sequence against capacity, changeovers, and due dates together.
- Set time fences. Freeze the next few shifts or days, so work already released runs as planned. Allow limited changes in the next band, and full re-sequencing beyond it.
- Level what you can. Where demand is steady enough in total, spread the mix of products evenly across the period (heijunka), so a demand swing on one product stays contained within the pattern.
- Re-sequence on a rolling cycle. Run a new sequence on a fixed rhythm, such as each shift or each day, and whenever a major event occurs, such as a large order, a breakdown, or a material delay.
- Sequence around constraints. Respect setups that depend on the previous job, shared tools and operators, and material arrival, alongside due dates.
- Change as little as the new demand requires. Keep jobs in place where they still work, and move only what the change forces.
- Measure stability as well as delivery. Track how much of each new schedule differs from the last one, alongside on-time delivery, so the plant can see whether re-sequencing helps or creates churn.
The balance is the hard part. Re-sequence too rarely, and the plant builds products ahead of real need. Re-sequence every hour, and the shop floor loses trust in the schedule.
What is the difference between production scheduling and sequencing?
The difference between production scheduling and sequencing is scope: scheduling sets the start and finish time of each job on each resource, and sequencing sets the order of those jobs.
Sequencing sits inside scheduling. A plant can have the right jobs scheduled for Tuesday and still lose hours to changeovers because the order is wrong. In plants with long or sequence-dependent setups, the sequence often decides how much the bottleneck produces in a shift.
Which sequencing rules do manufacturers use?
Manufacturers use priority rules to decide which job runs next, and each rule favors a different goal, such as due dates, throughput, or fewer changeovers.
| Rule | How it orders jobs | What it favors |
| First in, first out (FIFO) | In the order jobs arrive | Simplicity and predictable flow |
| Earliest due date (EDD) | Earliest due date first | Meeting due dates on a single resource |
| Shortest processing time (SPT) | Shortest job first | Lower average waiting time and WIP |
| Longest processing time (LPT) | Longest jobs first when assigning jobs across eligible parallel machines | Reducing overall completion time (makespan) in parallel-machine scheduling |
| Least slack | Smallest gap between time remaining and work remaining first | Jobs at greatest risk of running late |
| Critical ratio | Lowest ratio of time remaining to work remaining first; a ratio under 1 means the job is behind | Recovering late work as conditions change |
| Changeover grouping | Similar jobs run together, ordered to shorten each setup | Fewer and shorter changeovers on the bottleneck |
Least slack and critical ratio are dynamic rules: their values change as time passes and work completes, so the order updates itself. In practice, plants combine several rules, because the job that best meets one goal often works against another. Advanced planning and scheduling (APS) software can evaluate competing objectives together, balancing due dates, throughput, capacity, and changeovers according to the constraints and goals configured for the plant.
What is heijunka, and when does it help?
Heijunka is level scheduling. The Lean Enterprise Institute defines it as “leveling the type and quantity of production over a fixed period of time.”
Instead of building all of one product, then all of the next, a level schedule mixes products in a repeating pattern sized to average demand. A small finished-goods buffer can absorb day-to-day demand swings, helping the plant and its suppliers work to a steadier production pattern.
Heijunka works best when total demand is stable and changeovers are short enough to run a frequent mix. Where demand swings sharply or changeovers are long, plants combine a level pattern for steady products with dynamic re-sequencing for the rest.
How do time fences keep a schedule stable when demand changes?
Time fences keep a schedule stable by dividing the horizon into zones with different rules for change: a frozen zone that runs as planned, a middle zone that allows limited changes, and an open zone that can be fully re-sequenced.
The spans below are illustrative. Each plant sets its time fences according to its material, tooling, setup, and release lead times.
| Zone | Example span | What can change |
| Frozen | The next few shifts or days | Only urgent, approved changes |
| Limited change | The following days to weeks | Sequence and quantities within agreed limits |
| Open | Beyond that | Full re-sequencing against new demand |
As a rule of thumb, the frozen zone covers at least the time needed to stage materials and tooling for the work already released.
What is dynamic scheduling?
Dynamic scheduling is scheduling that updates as conditions change, using current data on orders, capacity, and shop floor status, in place of a fixed plan built once a week.
Research published by the US National Institute of Standards and Technology (NIST) describes dynamic decision-making tools, such as scheduling, as a way for manufacturers to respond to disruption by drawing on the data their facilities already produce. The same paper names integrating and curating that data as a challenge. In practice, that means dynamic scheduling depends on the scheduler seeing current status from the shop floor.
How does Eyelit APS re-sequence when demand changes?
Eyelit APS re-sequences production against current demand while limiting disruption to the schedule already in place.
Eyelit APS replans with memory: when demand or supply changes, it minimizes disruption to the existing schedule, with the degree of allowed change configurable at any level or attribute.
Eyelit APS uses attribute-based changeover matrices with additive or maximum setup policies and per-resource setup times.
For plants with several linked zones, Eyelit APS performs multi-zone production sequencing — for example Body, Paint, and Trim — optimized holistically in a single run, and models re-entrant flows, merges, splits, buffers, and differing line speeds across zones. Eyelit APS generates near-optimal weekly sequences at the scale of 15,000 vehicles.
For suppliers delivering in sequence, Eyelit APS supports reverse-order (‘bat order’) build constraints for LIFO in-sequence delivery by automotive suppliers. Our guide to just-in-sequence for automotive suppliers covers OEM sequencing in detail.
An automotive OEM increased paint batch sizes from 3 to 9 while respecting Trim spacing constraints. Our guide to APS capabilities for automotive OEMs covers this result and the 15,000-vehicle sequencing scale.
For selection criteria, see our guide on how to evaluate APS software, or compare vendors in our comparison of the 10 best APS software platforms for manufacturers.
