The Approval Queue Nobody Puts on the Capacity Plan
Updated: Aug 26
Approval Is Work
In a regulated environment, the operating model must be very clear: AI assists, humans decide. Systems of record remain authoritative.
That observation from a recent ISPE discussion of pharmaceutical batch disposition captures a constraint many operations plans still miss. A facility can manufacture a batch on schedule, complete analytical testing, and have finished goods physically available. But it does not have product available to supply until the required evidence is complete, exceptions are resolved, and the quality unit makes the final disposition decision.
The batch may be complete. The release system is not.
Most capacity plans model equipment hours, labor availability, raw material lead times, and fill finish slots. Far fewer model the actual capacity of the people and workflows that review batch records, assess deviations, approve change impacts, reconcile data, and make release decisions.
That omission creates a hidden queue.
When the queue grows, manufacturing appears to be performing while the business is still waiting for product.
The Problem
Life sciences operations teams rarely choose to under resource QA review or validation work. The issue is that the workload is dispersed across the operating model.
A new product campaign increases batch documentation. A process change adds assessments and approvals. A late deviation creates investigation work. A new digital system adds validation evidence. A site transfer introduces new quality agreements, master data checks, and comparability questions.
Each item can look manageable in isolation. Together, they create a release bottleneck that is visible only after inventory begins waiting.
The usual response is to ask quality teams to work harder, approve overtime, or create a temporary release task force. Those actions may clear an immediate backlog, but they do not change the mechanism that created it.
The real problem is not that quality review exists. It must exist. The problem is that operations treat it as an after the fact administrative step rather than as a capacity constrained production process with defined inputs, flow, service levels, and escalation rules.
Root Cause One: Capacity Planning Stops at Manufacturing Completion
Most site plans define a batch as complete when processing, packaging, or testing ends. From a supply perspective, that definition is incomplete.
A released batch requires more than physical output. It requires a complete, reviewable, traceable set of evidence. Under FDA CGMP requirements, production and control records, including relevant audit trails, must be reviewed and approved by the quality unit before batch release.
That means release capacity is real capacity.
If a site plans for twenty batches but its quality review system can reliably disposition only fifteen under the expected level of exceptions, the site has planned for fifteen saleable batches, not twenty. The remaining five are inventory in limbo.
This gap becomes especially costly during launches, demand surges, remediation programs, and new site ramp ups. Manufacturing leaders see a full schedule. Finance sees inventory. Customers see a shortage. QA sees an overloaded review queue.
All three perspectives can be true at the same time.
Core Insight: A batch is not capacity realized when it leaves the line. It is capacity realized when it is released for supply.
Root Cause Two: Exceptions Arrive Too Late and Too Expensively
In a paper based or fragmented workflow, an exception often becomes visible at the end of the batch. By then, the operator may be on another shift, the equipment context has changed, supporting records are spread across systems, and the original decision has to be reconstructed.
That is not simply a documentation issue. It is a flow issue.
Late discovery turns a small discrepancy into a cross functional investigation. Manufacturing has to explain execution. Quality has to assess impact. Engineering may need to confirm equipment behavior. QC may need to revisit laboratory data. Supply chain may have to revise commitments.
The work multiplies because the issue was not captured, classified, and resolved while the relevant people and evidence were available.
Review by exception can help, but the principle is broader than a specific technology. The objective is to identify meaningful exceptions early, route them to the right owner, and preserve the evidence needed for a timely decision.
When information is fragmented, each release becomes a scavenger hunt.
Core Insight: The cost of an exception rises sharply once the batch is over and its context has disappeared.
Root Cause Three: Every Batch Enters the Same Queue
Many quality systems process clean batches and difficult batches through the same review path. A routine batch with complete records waits behind a batch with an unresolved deviation. A high priority market shipment receives no different treatment than stock for a low urgency replenishment run. An approver begins work based on the order documents arrived rather than the risk and supply consequences of delay.
That is a queue design problem.
Not every batch should receive the same level of attention, but every batch should receive the attention its risk requires. The answer is not to automate final release decisions or weaken oversight. The accountable QA reviewer or Qualified Person must remain responsible for final disposition.
The answer is to make the work visible and prioritize it intelligently.
A practical release queue distinguishes between clean records and true exceptions, highlights incomplete evidence, identifies batches tied to supply risk, and shows which decision or input is blocking release. It gives reviewers a defensible order of operations instead of a growing undifferentiated backlog.
Core Insight: Quality rigor does not require treating every record as equally uncertain.
The Real Cost
The clearest cost is time, but the business impact compounds beyond time.
Industry sources report that paper based batch release can take from 10 to 40 days, with a cited average of 48 hours of review effort for a single batch report. These figures vary substantially by product, process, and site, so they are not a universal benchmark. They are a warning about the scale of work that can be hidden in a release process.
Consider a simple operating example. If a site produces twenty batches a month and each batch spends an avoidable two extra days waiting for review or clarification, that is forty batch days of finished goods tied up every month.
The immediate cost may include working capital, storage, temperature controlled space, expediting, and missed shipment windows. The larger cost is loss of operating control. Leaders cannot distinguish a genuine quality risk from an avoidable approval delay. Teams start escalating everything. Priorities become political. Quality becomes positioned as the obstacle even when the root cause is a poorly designed upstream process.
That dynamic is dangerous. It encourages pressure on the release decision rather than improvement of the system that feeds it.
The Fix
Sarga II would approach this as an end to end release flow diagnostic, not a software selection exercise.
First, map the path from batch completion to disposition. Identify every handoff, system, required record, review, exception, signature, and decision point. Measure actual queue time separately from active review time.
Second, classify the recurring sources of delay. Are batches waiting for missing execution records, laboratory results, deviation closure, audit trail review, master data reconciliation, second person review, or a specific approver? A queue without delay codes becomes an opinion. A queue with delay codes becomes an improvement agenda.
Third, establish risk based release triage. Define which conditions require immediate escalation, which clean batches can follow a streamlined review path, and which supply critical batches require coordinated review without compromising quality standards.
Fourth, move exception detection closer to execution. The best time to resolve a missing signature, unexpected parameter, or documentation inconsistency is while the operator, equipment state, and process context are still accessible.
Finally, manage release as an operating system. Give it a daily management cadence, visible work in progress limits, clear ownership, measurable service levels, and escalation rules that protect both patient safety and supply reliability.
Case Pattern
A growing biopharma site expands its production schedule to meet a new demand forecast. Manufacturing is confident because equipment utilization is manageable and staffing plans are in place.
Within weeks, finished goods begin accumulating in quarantine.
The initial assumption is that QA needs more reviewers. The diagnostic shows a more complex picture. Nearly half of delayed batches are waiting for clarifications that could have been resolved during execution. Another group is delayed because laboratory and manufacturing evidence sit in separate systems with no shared release readiness view. A smaller but urgent group is held by deviations that lack a clear decision owner.
Adding reviewers helps temporarily, but it does not remove the avoidable work.
The site redesigns its release flow around early exception capture, defined readiness criteria, daily risk based triage, and clear ownership for unresolved items. QA retains final release authority. The operating system around QA becomes more reliable.
The result is not less compliance. It is less waiting created by preventable ambiguity.
What Good Looks Like
A healthy release system has no mystery inventory.
At any point, operations leaders can see which batches are in review, which are blocked, what evidence is missing, who owns the next action, and what supply exposure exists. Clean batches do not disappear into the same queue as complex investigations. Exceptions are visible near real time, not discovered during final document assembly.
Quality reviewers spend their expertise assessing meaningful risk rather than searching for context across disconnected tools. Manufacturing teams know what review ready means before the batch ends. Supply teams can make commitments based on release reality rather than optimistic assumptions.
Most importantly, the site stops framing quality and speed as opposing goals.
A well designed release system improves both. It protects the rigor of the final decision while removing the avoidable friction around it.
CTA
If finished goods regularly wait longer than the production plan expected, the constraint may not be on the line.
It may be in the approval queue nobody put on the capacity plan.
If this pattern is familiar, we should talk.
https://www.sarga-ii.com
Comments