Planning Time Fence: Meaning, Length and Exceptions

Home » Resources » Planning Time Fence: Meaning, Length and Exceptions

Estimated reading time: 9 minutes

A planning time fence is a time boundary that changes how a manufacturing plan is calculated or revised. In many MPS and MRP systems, it protects near-term supply proposals from automatic changes, leaving the planner to decide which exceptions are worth making.

That protection matters before a machine starts. Material may already be picked, fixtures prepared and the next job agreed with the operator. A new planning run can produce a different answer without accounting for all the preparation it would undo.

The aim is not to stop production from responding. It is to stop routine recalculation from casually overturning commitments. SkyPlanner APS supports the scheduling side of that decision: protect upcoming jobs, let Arcturus recalculate schedulable work, and see the effect on later operations in the Gantt instead of rebuilding the queues by hand.

What does a planning time fence actually control?

The fence marks a change in planning policy, not the end of the plan. You still need to see demand, supply and capacity beyond it.

In an MRP firming context, firming means protecting a supply proposal from automatic alteration by subsequent planning runs. It does not mean the order is already manufactured, or that enough capacity exists to complete it. How existing proposals and new requirements are handled depends on the system and its firming settings.

The name is not completely uniform across ERP systems. For example, JD Edwards uses its planning fence to change the demand-netting rule, with a separate freeze fence controlling order creation and replanning messages. Oracle Fusion can also derive a natural fence from firm supplies. Check what your system’s fence governs before treating it as a production freeze.

For daily decisions, keep these controls distinct:

Control	Question it answers
Business change policy	Which changes to near-term product quantities and dates require an explicit decision?
MRP firming	Which supply proposals should subsequent planning runs leave unchanged?
Detailed scheduling lock	Which scheduled operations should the scheduler leave in place?

A firmed product quantity and date do not, by themselves, secure a machine slot. Conversely, locking a machining operation does not approve a change to the MPS quantity. Connect the two decisions rather than assuming one control does both.

Give near-term work different rules from later work

A useful production policy often separates three levels of flexibility. These are working agreements, not three settings that every ERP implements identically.

• Frozen work: routine changes are not allowed. A named planner or production manager decides exceptions because preparation or execution is already committed.
• Controlled work: changes are possible after checking materials, capacity and other delivery commitments.
• Flexible work: the plan can adapt more freely as demand becomes clearer, within the available resources and business priorities.

Decide what each agreement protects. Does a change to quantity require review? What about moving the completion date, changing a resource or inserting another batch? Who can accept the consequences?

Without those answers, a fence becomes a date that everyone interprets differently. With them, purchasing and production can prepare against a plan they understand, while sales knows when an earlier delivery request needs an explicit tradeoff.

How long should the planning time fence be?

Base the length on how quickly you can change the work responsibly, not on a universal number of days.

Start with the relevant material and production lead times. A batch may need bought-in components, outside processing and several internal operations before it becomes available. The protection decision must account for commitments made along that route, not just the duration of the final operation.

Then identify the point where changing the plan starts to waste real work: a supplier commitment, a prepared kit, a reserved fixture, a completed setup or an agreed shift sequence. Ask how much notice the affected people need to change direction without unnecessary cost.

A fence that is too short lets the plan move after those preparations have begun. A fence that is too long preserves avoidable delays and leaves less freedom to accommodate a valuable new order. Different products or supply routes may justify different planning policies; check how your ERP supports that distinction.

Also distinguish the business planning window from the shop-floor protection window. You might control changes to a product commitment several working days ahead, while protecting the detailed sequence over the next few hours. They serve different purposes and need not have the same length.

Review the policy when exceptions keep recurring. If prepared work is repeatedly displaced, protection may start too late. If jobs are protected but regularly cannot run, investigate material readiness, capacity and the realism of the original plan before extending the fence.

A prepared batch versus an earlier delivery request

Imagine a precision-components manufacturer preparing a batch of valve housings. The material is picked and the fixture is ready for an upcoming machining run. A second customer asks whether its batch can be brought forward.

The MPS/MRP policy requires review of near-term quantity and completion-date changes. On the floor, the immediate question is narrower: should the prepared housing batch lose its place?

First check whether the second batch can fit without disturbing that preparation. Its machining slot is only part of the answer: it still needs deburring and inspection before dispatch. Starting it earlier is useful only if the remaining route supports the requested delivery.

If bringing it forward displaces the housing batch, compare the benefit with the consequences. Will its material need to be returned? Will the fixture preparation be wasted? Which customer’s completion moves, and by how much? A changed priority should buy something worth that disruption.

This is where a visible, connected schedule is more useful than a firmed date alone. In SkyPlanner, you can keep protected operations in place while Arcturus recalculates the schedulable work under your priorities. The Gantt shows the resulting sequence across workstations, so you can judge the later completion dates before accepting the tradeoff.

Protect upcoming work in SkyPlanner without losing the wider plan

SkyPlanner’s Running Time Lock protects upcoming jobs within a configured number of hours from rescheduling. The window rolls continuously from the present moment. It gives the floor a stable near-term sequence while later schedulable work can still be recalculated.

On the Gantt, pink jobs identify the protected work and orange lines mark the end of the time lock on each workstation. You can see what the protection covers alongside the work that follows it. That saves repeated queue preparation and makes it easier to explain why a new priority has not simply displaced the next job.

For the valve-housing example, a rolling lock can protect the imminent prepared work. It is a detailed scheduling control alongside the business planning policy, not an MRP firming rule or an MPS change-approval workflow.

Two other distinctions matter when applying that protection:

• A workstation lock protects the resource assignment. Locking a process step to its current workstation keeps it on that resource during rescheduling; do not confuse that with fixing every timing decision.
• Started work is already a commitment. A job that has started, including one that has started and then been paused, remains in progress and locked in its Gantt position. Arcturus does not treat Pause as permission to move it. Waiting, not-yet-started jobs remain schedulable unless another rule protects them.

You decide the priorities and the protected decisions; Arcturus handles the scheduling calculation around those constraints. The gain is less manual rearrangement without losing sight of what the remaining work will require.

SkyPlanner integrates with any ERP or MES through its open REST API. Keep the item-level planning policy in your existing process and connect the production orders to a detailed schedule that shows how those commitments will be carried out.

What stable work queues changed for Mamec

Mamec, a Finnish manufacturer of industrial mixers and high-tech contract products, had thousands of interdependent work orders. Its earlier planning relied on paper work orders and delivery-date prioritisation on the floor, without a reliable overall view of load.

With SkyPlanner integrated with Oscar ERP, AI-powered finite scheduling provides the groundwork and work queues locked a week at a time give machinists a clearer sequence. Knowing the next job lets them prepare upcoming work during long runs. Management can also see the production load rather than rely on subjective estimates.

Mamec reports a clearly lighter planning workload and improved delivery reliability. Its weekly locking practice is not a recommended fence length for every factory. It shows the practical value of stability: people can prepare useful work while the organisation retains the bigger capacity picture.

When should you make an exception?

Make an exception when preserving the current plan is no longer feasible, or when the benefit of a change justifies its full cost. A material shortage, quality hold or lost resource may make the protected sequence impossible. An important earlier delivery request may justify changing it, but urgency alone does not make capacity available.

Before changing protected work, establish three things:

1. What has changed? Confirm the demand, material, quality or resource situation with the people responsible.

2. What would the change displace? Check prepared and in-progress work, the remaining route and other promised completions, not just the first open slot.

3. Who accepts the new commitment? Agree the decision, communicate the revised sequence and update the quantities or dates used by sales, purchasing and production.

Do not remove all protection merely to get a different scheduling result. Resolve the affected commitment, apply the necessary rule or priority change, and recalculate the schedulable work. Started or paused jobs need their own explicit production decision; disabling a rolling lock does not turn them into untouched waiting work.

Track why exceptions occur and what they undo. The useful outcome is fewer avoidable changes and clearer decisions when a change genuinely matters, not a plan that stays unchanged while the factory falls behind it.

To see that balance in your production, book a SkyPlanner demo. Bring a typical prepared batch, the window you want to protect and a recent request that challenged it. You can examine how protected work and later load stay visible in one schedule. Or try SkyPlanner with demo data to explore the scheduling controls first.