IT systems · Practical guide

Validating Körber PAS-X: linking the correct recipe version to permitted execution

An approved recipe alone does not authorise a production step in Körber PAS-X MES. Equipment status, timing and unresolved events may also affect execution. Trace those conditions through to the electronic batch record. A timing example shows why recipe changes and expiring equipment approvals need to be tested together.

Table of contents

At a glance

What does validation of Körber PAS-X MES cover?

Demonstrate that the configured production workflow applies the correct instructions, equipment conditions and quality decisions. Include recipe versions, permitted transitions, data capture, exceptions, permissions and traceable batch records. Test changes between instruction sets and time-dependent equipment states. Supplier functions and test documents can contribute evidence where they cover your process, installed version and actual integration.

  • The MBR instruction, the actually executed procedure, and the electronic batch record are distinct audit objects.
  • A recipe-free release and an acceptable equipment status must be concurrent at the specific point in time.
  • An exception-based review requires substantiated coverage; an empty overview is not a standalone completeness justification.
  • For ongoing and new batches, prior to a recipe change, it must be established which version is currently valid.

1. Define the boundaries between PAS-X MES, EBR, Equipment Management, and Neo

Körber presents PAS-X MES as a solution for networked production environments. The public presentation highlights features such as Electronic Batch Recording (EBR) and Equipment Management. Various operating models are described within the product portfolio; however, this does not imply that every variant or module is offered in every model. The specific offering and the documented system status are decisive.

PAS-X Neo was announced by Körber in May 2026 as an additional offering for smaller and mid-sized enterprises and sites. It is not a simple renaming of PAS-X MES. Prior to selection, clarify the actual delivery status, features offered, and scope of validation evidence with the manufacturer. Features of the announced product line do not constitute proof of compatibility or suitability for your existing MES installation.

The professional reference framework is GMP manufacturing of human medicinal products. Annex 11, Chapter 4 and ICH Q9(R1) provide the applicable bases. The public PAS-X website pages do not specify a concrete numerical MES version. Therefore, obtain release, modules, and project-specific documentation prior to deriving detailed functional statements or test instructions.

Sources: [1] [2] [3] [4] [5] [6] [7]

2. Specification, execution, and documentation to be described separately

A Master Batch Record (MBR) defines the intended manufacturing process. The actual batch execution implements this specification using specific materials, users, times, and equipment. Electronic documentation records the actual activities that took place. Therefore, an approved formulation design is still not sufficient evidence of the subsequent batch execution and its documentation.

The following internal work aid segregates these levels. Supplement your process with the leading systems, responsible roles, and locations. If information arises outside of PAS-X, such as a laboratory decision, its significance must be known upon entry. Technical success of the handover alone does not address this technical question.

The traceability chain for a configured PAS-X configuration
Item under testProcess questionSuitable evidenceTypical open boundary
MBR and recipe instructions Which approved batch applies to this batch? Identifiable specification, release, and defined transition Behaviour of batches already in progress
Equipment conditions Is the specific device suitable for the intended step? Status, relevant deadlines and authorized transitions Outdated display or expiry during use
Execution Were steps, entries, and deviations appropriately handled? Observed workflow with user, time, and acceptance criteria Interruption, resumption and an unauthorised path
EBR and Review Can the actual execution be reconstructed? Recording, events, decisions, and associated history Anomalies that were not captured or are hidden
System handover Does the target system correctly interpret status and data? Assignment of fields, professional comparison, and error handling Misidentification of closure, inspection, and legal certification

Sources: [3] [5] [6]

3. Test recipe modules and the change to a new version

Körber describes graphical master batch record design with reusable modules and configurable approval workflows in its EBR offering. Identify which modules your recipes share and which instructions depend on them. Changing one parameter can affect several execution variants even under a single change request.

Determine the professionally justified transition point prior to testing. From what point may a new batch commence use of the updated status? What is the status of a batch that has already been initiated? Can a change affect only future activities, or is an assessment of ongoing processes also required? The answer depends on the specific change; a technical version lock is not yet a business decision.

Keeping two recipe versions as documents does not establish which instructions ran. Select actual executions and show the version in effect and the change history retained. Justify which recipes and variants you check. Repeating every test or referring to a shared standard module without assessing it will not answer this question.

Sources: [3] [6] [5] [7]

4. Understand equipment status as a time-dependent condition

The PAS-X Equipment Management manufacturer description specifies state models for cleaning, sterilization, and maintenance, as well as timers, counters, and barcode verification. The actual states used by your installation and their association with operational execution must be verified on a project-specific basis. A visible status color does not fully explain the logic underlying the equipment's usage decisions.

Therefore, for each relevant condition, specify the trigger, the validity period, and the consequence. Does a cleaned device remain valid until a specific deadline, until the next use, or additionally under specific storage conditions? Must a planned phase be fully completed within the allowable time period? Review these technical specifications wherever an operator initiates a step late.

Consider the origin of the time and the status. A barcode can identify the correct device, even when its displayed status is no longer current. Therefore, a targeted test case is the transition from permitted to not permitted, including delayed display or interrupted connectivity, provided such behavior aligns with your architecture. The expected response must be defined prior to the test.

Sources: [3] [5] [6]

Further reading:

5. Worked example of a recipe change and a cleaning time limit

Example scenario, no executed PAS-X test: The mixing device M01 is recorded as cleaned in the test dataset at 08:00. The pre-approved example rule permits usage for 120 minutes, i.e., until 10:00. A phase must be completed prior to the expiry of this period. This time limit is not a general GMP requirement and is not derived from the PAS-X standard.

V2 requires a 20-minute phase. V3 requires 30 minutes and applies to newly started batches from 09:40 Uhr in the example. The change decision in the example permits a batch A started at 09:30 Uhr to complete under V2; no retroactive quality issue is present in the scenario. The specific implementation of this binding would need to be assessed.

For Batch B, a planned start at 09:45 Uhr under V3 results in a planned end time of 10:15 Uhr. Although M01 is still within the cleaning validity period at 09:45 Uhr, this execution does not meet the example requirement. A subsequent cleaning at 10:20 Uhr extends the deadline to 12:20 Uhr. A new start at 10:25 Uhr results in an end time of 10:55 Uhr and falls entirely before the extended deadline.

Expected results for recipe version, time calculation and permitted path
Test procedureSpecification and Time AccountingEquipment conditionExpected behaviourRequired documentation
Batch A starts at 09:30 V2: 09:30 + 20 min = 09:50 M01 valid until 10:00 Continue from Example Ruling under V2 Active Status, Start, End, and Reference for Changes
V3 will be effective at 09:40. New batches receive 30 min A is still running; its end remains 09:50 A not unnoticed transition to V3 V2 Implementation and separate, traceable activation of V3 functionality
Batch B would like to start at 09:45 V3: 09:45 + 30 min = 10:15 Deadline expires at 10:00 AM. Do not allow start after example rule Rejected path, rationale and open handling
Retry at 10:00 V3 remains valid Cleaning period expired No initiation through simple repetition Device status, time, and observed reaction
Retain 10:20; Start 10:25 V3: End 10:55 New deadline: 10:20 + 120 min = 12:20 Allow based on confirmed new status Cleaning validation, system status change and new configuration
Later review of batch A Historically, V2 with 20 min was considered The subsequent device status is not the one that existed at the time. The original workflow remains clearly understandable. Historical instruction and previous status reference

Sources: [5] [6] [7]

6. Review interruptions, exceptions, and decision-making authority

The example matrix deliberately includes a rejected start. Define how the team handles it. Who identifies the cause? Who may correct it, and what conditions permit another attempt? A general administrator role does not explain who may make that decision.

In addition, assess a disruption following the start of a phase. Are previously completed activities still documented? Does a duration need to start anew, continue, or be professionally re-evaluated? These are requirements derived from your manufacturing process. Define them prior to a test that merely confirms a screen can reopen.

An intended exception function, together with its prerequisites and consequences, shall be evaluated. A justification input field does not validate a technically unjustified manufacturing process. A clear distinction must be maintained between technical operability, responsible process decision-making, and additional quality procedures. Whether an incident requires further investigation or corrective actions shall be assessed based on its specific content and the applicable procedures.

Sources: [5] [6] [7]

7. Align EBR and Review by Exception to ensure traceable coverage

An exception overview can support the review process. However, its reliability depends on which events are detected, transferred, displayed, and addressed. Specify for the intended review what anomalies must be visible and what additional records the reviewer needs. An empty list does not prove its completeness.

As a standalone validation scenario, the failed execution of Batch B presents a suitable case. Generate the pre-defined error state in the appropriate test environment and follow it through to the review decision. Compare the technical trace with the documented information. Additionally, assess whether filters, permissions, or a subsequently altered status have rendered the original event incomprehensible.

A batch record should link the instructions used, execution and relevant decisions. For each signature, establish the content confirmed. Technical completion and EBR approval do not automatically constitute certification of a medicinal product batch by the Qualified Person. The same system may record separate decisions with different meanings and responsibilities.

Sources: [5] [6] [8]

Further reading:

8. Verify ERP and automation transfers at the execution level

Körber describes interfaces to the ERP and to automation. For your review, you require the specific functions of these interfaces: Does a production order with quantity and unit arrive? Is a machine status transferred? Does a system provide measurement data or confirmation of a completed step? Assign each data point a technical recipient and an authorized usage purpose.

A useful integration validation follows an order from initiation to final feedback. Systematically vary a critical value or state affecting the workflow and independently compare source and target systems. Assess error handling, retry mechanisms, and sequence where the actual data transfer mechanism introduces these risks. The term "interface" does not imply a fundamentally asynchronous architecture nor a specific messaging logic.

For a system failure, it must be demonstrable which actual procedure took place. A retransmitted command must not result in an unintended second execution. Determine the allowable retry based on the process and verify the required documentation. A data transfer test without consideration of the physical or functional impact leaves this question entirely unresolved.

Sources: [3] [5] [7]

9. Consider changes, restoration, and historical batch evaluations together

A version change can affect recipe design, state model, interface or representation of historical processes. Use appropriate manufacturer documentation and your configuration overview to determine impact. In a common recipe module, the affected uses belong in the evaluation. However, this does not automatically repeat all recipes and tests.

Plan a historical comparison for a relevant change: Can Batch A still be reconstructed using V2 and its original conditions, despite V3 now being active and M01 having been cleaned multiple times? A current equipment status alone is insufficient. The version-specific documentation and supporting evidence must demonstrate the continuity of the historical record and establish the relationship between the systems and processes.

Restoration and long-term retention address distinct considerations. For a restart, a consistent operational state including open actions is required. For historical decisions, readable and understandable records together with necessary context are needed. Define scope and retention period appropriately aligned with the data and applicable requirements. A universal export format or a blanket deadline would be overly simplistic.

Sources: [5] [6] [7]

10. Prepare a realistic assessment of resource commitment and local involvement for your site

The effort often arises from variations and decisions: shared recipe building blocks, multiple equipment models, shift changes, manual exceptions, and integration into existing workflows. A small site can be more complex than its user base suggests. Clarify early which processes should be initially covered and which should remain in a controlled, separate procedure.

Production provides valid instructions and typical exceptions. Technology outlines equipment conditions and physical workflows. QA evaluates quality decisions and the scope of evidence. IT describes operations, time availability, access controls, and integrations. The supplier validates the specific mechanism offered. Identify a decision-maker for conflicting process requests to prevent unclear rules from becoming apparent only during test execution.

  • Offer, MES release, modules, and actual operating mode are clearly described.
  • Specifications, recipe components, and variants are documented with relevant technical responsible parties.
  • Rules for new and ongoing batches during a recipe change are decisive.
  • Equipment conditions have clear triggers, deadlines, and consequences for execution.
  • Normal paths, boundary times, interruptions, and exceptions have verifiable acceptance criteria.
  • Review coverage, signature significance, and legal certification decision are defined.
  • Interfaces, retry mechanisms, and historical records are verified with appropriate documentation.

Sources: [1] [5] [7]

Project experience and collaboration

Test how recipe changes affect batch execution

With your manufacturing owners, we examine a PAS-X recipe step as it is actually executed. Which recipe version applies, what decisions does the operator make and what information remains in the batch record? The answers guide testing for a specific change. For a new manufacturing sequence, we first agree the conditions for each step with the process team and IT, including relevant status changes and exceptions.

Frequently asked questions about Körber PAS-X

What should we clarify prior to selecting PAS-X Neo?

Körber announced the first version for November 2026, with an announcement date set for May 2026. An announcement date does not confirm the available delivery scope for your project. Prior to selection, verify the actual delivery status, features, operating agreement, and available evidence for the specific offering.

Sources: [2]

Must a MBR with individually programmed code be validated?

Critical factors are the configured functionality and its impact, not merely the method of development. A graphically composed workflow can influence critical decisions. Capture specifications, employed building blocks, and variants; then define appropriate validation evidence and testing strategies. A one-size-fits-all GAMP category for each MBR cannot be derived from this approach. Similarly, a user interface without programming does not exempt it from validation assessment.

Sources: [3] [5] [7]

What should happen to batches in progress when the recipe changes?

In change control, decide which recipe version applies to batches already in progress and whether those batches need action. A text change for future work differs from discovery of a quality-critical error. Identify affected batches, the permitted transition and the decision owner. Test whether execution follows that decision. The continued processing of batch A is an assumption of this example.

Sources: [6] [5] [7]

Is it sufficient to verify the equipment status at startup only?

An initial validation is sufficient only if it adequately addresses the defined intended use. If a time limit restricts the entire intended scope of use, the execution must align accordingly. Additionally, clarify how state changes during a process are handled. An initial validation may be part of the concept, but it is not automatically sufficient for every process. Evaluate time-bound milestones and the actual impact of an active condition.

Sources: [3] [5] [6]

Is a successful barcode scan evidence of appropriate equipment?

The scan can support object identification. Additionally, the decision requires the correct and sufficiently up-to-date state of the identified object, as well as the appropriate process rule. Therefore, verify not only readable barcodes but also incorrect equipment, an unauthorized state, and the relevant time boundary. Your configuration must demonstrate how these conditions are logically linked.

Sources: [3] [5]

Can QA, during a review by exception, solely evaluate the displayed exceptions?

Such a scope requires a justified and demonstrated coverage of the specific process. Demonstrate that relevant events are reliably captured and that the review decision receives the necessary information. Additionally, identify necessary checks outside the standard oversight. An empty display alone or the manufacturer designation "Review by Exception" does not constitute evidence that all relevant facts have been fully considered.

Sources: [5] [6] [7]

Is an EBR approval equivalent to QP certification?

Not automatically. EBR approval may indicate review of manufacturing records. Certification of the finished medicinal product batch by the Qualified Person has its own legal scope. If both decisions are recorded in one system, define what each confirms, who may make it and how it is recorded.

Sources: [8] [5]

How can the simulation of a process be used as evidence?

First, evaluate which aspects of the actual implementation are represented by the test environment. A simulation may provide appropriate logical evidence, but it may not reflect real device interlocks or on-the-job user conditions. Assign the evidence to the covered requirements and specifically supplement missing proof where necessary. The tool's name does not determine its transferability to your intended operation.

Sources: [5] [7]

What information must a small and medium-sized enterprise provide for a reliable cost estimation?

Outline typical recipe workflows, including the number of distinct functional variants, common components, equipment models, and interfaces. Supplement existing evidence, open process questions, and the proposed operating model. This enables a more accurate assessment than user count alone. Complex, inconsistent state rules or unclear exception decisions can be particularly resource-intensive; such areas should be explicitly planned as clarification tasks.

What should be specifically verified during the upgrade for old batches?

Select historical activities that allow assessment of relevant changes. For example, review the original recipe protocol, modified parameters, equipment assignments, and associated decisions. Not every change requires the same extent of historical testing. Crucially, existing documentation and supplementary evaluations must demonstrably cover the impact of the changes, and historical records must remain clearly accessible and understandable for future review.

Sources: [5] [6] [7]

Sources and review date

The official manufacturer and regulatory sources were made available on 6 October 2026. For PAS-X MES, no numerical release is documented from the publicly accessible product pages. PAS-X Neo was announced on 18 May 2026 for release in November 2026. Tables, deadlines, and batch information are standalone audit aids and do not constitute documented product testing.

  1. Körber: Life Sciences MES PAS-X Suite
  2. Körber: With PAS-X Neo, Körber introduces an additional MES solution
  3. Körber: Implement paperless manufacturing – PAS-X EBR and Equipment Management
  4. European Commission: EudraLex Volume 4
  5. EU GMP Annex 11: Computerised Systems
  6. EU GMP Chapter 4: Documentation
  7. EMA: ICH Q9(R1) Quality Risk Management, Correction 2
  8. EU GMP Annex 16: Certification by a Qualified Person and Batch Release

Sources checked: .

Daniel Herrmann, Daniel Herrmann Consulting

Daniel Herrmann Consulting

Check the recipe, equipment and batch record together

Daniel Herrmann Consulting supports you in aligning technical implementation guidelines and existing PAS-X evidence, and in systematically reviewing open transition points.

  • Computer system validation →

    Bring your GxP systems into validated operation, from laboratory software, MES and eQMS to ERP. We define the scope, plan risk-based tests using GAMP 5 and hand the evidence to QA for release.

Send a project enquiry Customer testimonials →