Articles
12 minutes
Copy Link
How to Route Production Scheduling Exceptions Without Waiting for the Plant Manager
TL;DR
Resolve routine production scheduling exceptions within predefined plant constraints. Route changes that exceed those limits for human approval.
Let an approved rule handle a reversible change within its limits. Require planner sign-off when a change affects capacity or delivery commitments, and management sign-off for larger tradeoffs. Stop changes that could violate safety or quality requirements.
Require current evidence before any decision. Escalate when evidence goes stale, and record who approved each consequential change.
The exception matrix below covers machine downtime, late materials, labor gaps, rush orders, and quality holds. A worked example follows the matrix from signal to sign-off.
Why schedule exceptions stall on one person's desk
Production scheduling exceptions reach the plant manager when a plant has rules for building a schedule but no clear authority for changing it. The ERP records order priorities, and the MES reports a machine stoppage. Neither record tells the planner whether moving the next job requires approval. Without a defined owner and deadline, the planner sends the decision upward, even if the change stays within approved capacity and delivery limits.
A breakdown or late component can disrupt a shift’s job assignments. The planner needs current evidence of the disruption and its effect on other orders. Without an agreed evidence requirement, the request waits for someone with broader authority.
The guides to ERP scheduling and AI decision layers and ERP integration cover system responsibilities and data handoffs. Each plant must also decide who can approve a proposed change, what evidence they need, and when the request escalates. Explicit decision rights let planners resolve bounded changes while sending consequential tradeoffs to an authorized signer.
Who can approve a production scheduling exception?
Route each production scheduling exception by its effect on approved capacity, cost, delivery, safety, and quality limits. The plant should set thresholds, name an owner and backup for each approval level, and specify when a proposal must stop rather than escalate. The time limits below are plant examples, not industry standards.
A dispatcher or scheduling rule can resequence work without case by case approval when the change stays inside previously approved capacity, material, labor, and delivery constraints. For example, a rule could move an order to another qualified machine if the operator is available, no other committed order slips, and no extra cost is incurred. The rule must record the triggering signal and the constraints it checked. If evidence becomes stale or a constraint fails, the rule must stop and route the proposal to a person.
A production planner signs feasible changes that affect another order, consume a protected capacity buffer, or incur overtime within the planner’s approved limit. If a change moves a customer commitment, the planner also needs the authority expressly assigned for that commitment. The planner checks the affected orders and current ERP and MES records before approving. A plant might require a response within 15 minutes and then route an unanswered request to the designated backup planner. No schedule change should proceed merely because the approval window expires.
The plant manager or another named manager signs changes that exceed the planner’s cost or delivery authority. A safety concern, an active quality hold, or missing evidence requires an immediate stop to the affected work, not a management approval to bypass the constraint. Quality owns release of held material. If an approved change loses its required material or capacity before execution, the planner withdraws it. The planner restores the prior schedule only if it remains feasible and approved. Otherwise, the planner submits a new proposal. A person signs every consequential decision.
The exception and approval matrix
The matrix below gives each exception an owner, an evidence requirement, and an escalation trigger. The limits are plant-configurable examples, not industry standards. In this example, a verified change may proceed without approval only when it stays within the current shift, preserves committed dates, and uses approved resources. A planner must sign changes outside those limits. Escalate an unanswered planner request after 15 minutes, but keep the affected order on its last approved schedule until someone decides. A person with the relevant authority must sign any change to a customer commitment, any cost above a planner’s approved limit, and any quality hold release. The matrix identifies the responsible role for each case.
Exception | Required evidence | Decision authority | ERP and MES updates | Escalation or stop trigger |
|---|---|---|---|---|
Machine downtime | MES fault time, maintenance estimate, alternate machine qualification and status, and order due dates | Use an approved resequencing rule within its limits. Planner signs an alternate route outside them. | Record the fault in MES. After authorization, revise MES dispatch and applicable ERP schedule fields. | Route downtime beyond the plant’s threshold to the planner. Escalate missing repair evidence to the backup; stop an unqualified route. |
Late materials | ERP receipt status, supplier ETA and timestamp, available stock, and approved substitute list | Use an approved resequencing rule within its limits. Planner signs a permitted substitute or internal date change. The authorized commitment owner signs a customer date change. | Record the material status in ERP. After authorization, revise applicable ERP dates and MES dispatch. | Route an unconfirmed ETA to the planner; escalate a proposed commitment change to its authorized owner. |
Labor gap | Attendance record, current qualifications, minimum staffing rule, and affected orders | Use an approved reassignment rule within its limits. Planner signs cross-line moves within assigned authority. | After authorization, update MES assignments and any affected ERP schedule fields. | Route an unanswered approval to the backup after the plant’s limit. Stop the affected operation if staffing or qualifications fall below requirements. |
Rush order | Requested due date, finite capacity check, displaced orders, and cost estimate | Planner signs within approved capacity and cost limits if commitments hold. The named manager signs a proposed commitment change or cost above the planner’s limit. | After authorization, enter priority and dates in ERP, then update MES dispatch. | Route an unanswered planner request to the backup after the plant’s limit; send commitment or cost exceptions to the named manager. |
Quality hold | Hold record, affected lots, inspection results, and disposition | The designated quality owner signs release. Planner may schedule unaffected work within approved limits. | Block affected lots in MES and ERP. Record the authorized release in both systems before restoring dispatch. | Stop affected work immediately. Route missing or disputed disposition evidence through the plant’s quality process. |
If new evidence invalidates an approved change, the planner should pause affected dispatch and check whether the prior approved sequence is still feasible. If it is, the planner can restore it and record the rollback. Otherwise, the planner must seek approval for a new sequence. A quality owner must separately authorize any change to a held lot’s status.
How does a machine downtime exception reach approval?
An illustrative plant rule routes downtime beyond 30 minutes to the planner, rather than directly to the plant manager. Suppose the MES reports that Press 4 stopped at 10:05 and a maintenance work order estimates a 40 minute repair. The planner has 15 minutes to respond before the request goes to a backup planner. These are plant configured examples, not industry thresholds.
The planner needs evidence before signing. The exception record brings together the MES stop event, the maintenance estimate, the affected orders, and current capacity on the proposed alternate press. It also checks material availability, operator qualifications, and due dates. If the alternate press can run the next order without breaching those constraints, the planner checks any displaced orders, signs the revised sequence within the planner’s authority, and records the evidence and decision. The approved change then updates MES dispatch instructions and the relevant ERP schedule fields through the plant’s authorized interfaces.
An unknown repair estimate or an alternate press without an approved setup would block that route and trigger escalation under the plant’s rule. If a quality hold arrives after approval but before the alternate order starts, dispatch stops. The planner withdraws the revision and checks the latest schedule state before issuing another. If work has already started, the planner records a new revision rather than treating rollback as though production could be undone.
How should a plant test and limit scheduling autonomy?
Run each exception rule in shadow mode before it changes a production schedule. The rule should assemble ERP and MES evidence, propose a route, and record its reasoning while a planner makes the actual decision. Compare each proposal with the planner’s decision and the shop floor outcome. Include cases where the rule should have stopped, such as a quality hold or conflicting machine status. Before granting authority, a plant might require 30 reviewed exceptions with no missed stop condition, including cases with stale or conflicting evidence. The sample size is a plant configured example, not an industry standard.
Grant autonomy only for the exception type and limits the plant has tested. For example, a rule could resequence jobs on one line when materials and qualified labor are confirmed, delivery commitments remain intact, and the change stays within a time window set by the plant. A planner still signs changes outside those limits. Management signs decisions with larger cost or customer exposure, while the designated quality owner retains authority to release a hold.
Record every proposal and decision so you can review whether the rule stayed within its authority. The log should capture the source evidence and its timestamp, the constraint version, the proposed change, the approval or rejection, the person who signed when required, and the resulting ERP and MES updates. Limit who can edit rules or expand their authority, and retain the previous rule version for review.
Withdraw automatic authority when source data passes the plant’s freshness limit, ERP and MES signals conflict, a constraint changes without review, or a proposed change breaches an approved limit. Route new exceptions to a planner until someone verifies the evidence and retests the rule. If dispatch has not begun, restore the prior schedule version. If work has begun, the planner must assess the impact and approve a corrective schedule rather than assume a software rollback can undo production.
How does Humble support scheduling approvals alongside ERP and MES?
Humble provides a decision intelligence layer above your existing ERP and MES. The ERP retains orders and planned dates, and the MES retains execution status. Humble can bring scheduling evidence and plant constraints together to propose a route and retain a decision record. Confirm which evidence, approval, and ERP or MES writeback steps your deployment supports before granting the rule authority. The ERP integration guide covers the data handoff in more detail.
You can describe plant constraints in natural language, such as which alternate machines are permitted and which changes require planner approval. When constraints change, Humble’s scheduling logic can revise a proposed response for review. The decision record should identify the evidence, applicable constraint, approval route, and person who signed a consequential change. If machine status is stale or a quality hold lacks release approval, plant rules should stop or escalate the proposal.
Humble does not replace a standalone advanced planning and scheduling system. Your plant sets the approval limits, and authorized people decide when changes enter ERP or MES. For background on the scheduling decisions that precede approval, see the production scheduling use cases and the guide to production scheduling challenges.
Check whether Humble fits your scheduling approval process
If your plant already runs ERP and MES, check whether Humble fits the way you handle schedule exceptions. The Humble fit test offers a starting point for assessing whether to test this approval model.
Discuss a scheduling exception pilot with Humble
If one type of schedule exception keeps reaching the plant manager, you can discuss a shadow mode pilot with Humble focused on that approval path. Bring the current approval path and a few recent cases. You can then test who should decide, what evidence they need, and when the exception must escalate before changing live scheduling authority.
Stay Ahead on Scheduling Governance
Get practical guidance on setting approval limits and reviewing production scheduling exception rules in the Humble newsletter.
FAQ: production scheduling exception routing
Who can release a quality hold?
Your plant’s designated quality authority should sign the release after checking the required inspection and disposition records. A planner can propose a revised sequence, but production must not resume on the held material until quality records the release.
How should you set escalation time limits for schedule changes?
Set each limit early enough for the backup approver to decide before the next operation needs a schedule, and name that backup in the rule. For example, if a replacement job must start within 30 minutes, route an unanswered planner request to the backup before that deadline. Treat the limit as a plant rule, not an industry standard.
When should you withdraw authority from an automated scheduling rule?
Pause the rule when its evidence becomes stale, ERP and MES signals conflict, or a proposed change breaches an approved constraint. Route the exception to the named planner, and reverse any unstarted schedule change that relied on invalid evidence. Record what triggered the pause before you restore the rule.
Put the approval matrix into operation
Choose one frequent exception type and document its owner, required evidence, approval limit, backup, and stop condition. Test the rule against past and live cases in shadow mode before allowing bounded schedule changes. Keep consequential decisions with the authorized signer and record each decision alongside the resulting ERP and MES updates.