In an automotive project, successfully passing a test does not mean that its result will remain valid indefinitely or for every subsequent version of the product. A component or vehicle may have met the relevant requirements during Design Verification (DV), Production Validation (PV), or a standards-based evaluation, but several months later, the material may change, a new supplier may be introduced, the thickness of a component may be modified, a mounting point may be relocated, or the system software and calibration may be updated. Under these circumstances, an important question arises: Can the previous test results still be used for the modified product, or must some of the tests be repeated?
There is no universal answer such as “always repeat the test” or “minor changes do not require retesting.” The decision should begin with an Impact Analysis. First, the exact nature of the change must be identified. Then, its potential effects on requirements, functions, failure modes, and the conditions under which the previous evidence was generated must be assessed. Only after this evaluation can a decision be made about retaining previous results, performing additional analysis, conducting partial retesting, or repeating a broader set of Verification activities.
This issue becomes more sensitive in regulatory projects. If a product already has Type Approval or another approval, assessing the impact of a change is not merely an internal engineering decision. The applicable regulatory requirements and approval procedures must also be reviewed. Therefore, change management and retest planning must consider both engineering and, where applicable, regulatory aspects.
Why Can a Design Change Affect the Validity of Previous Test Results?
Every test result relates to a specific set of conditions. The specimen had a defined part number, design revision, and configuration; the test was performed according to a particular method; and the result was generated to evaluate a specific requirement. If a characteristic affecting the evaluated behavior changes, it cannot automatically be assumed that the previous evidence remains representative of the modified product.
Suppose a vehicle seat assembly, Revision B, has undergone a structural strength test and met the relevant requirement. After testing, the thickness of a bracket or its connection specifications are modified. The previous result remains valid evidence for Revision B, but whether it can also be used for Revision C depends on the effect of the change on structural behavior and the relevant requirement.
Conversely, a change may have no technical relationship to the requirement that was tested. In such cases, all tests should not automatically be repeated simply because the design revision has changed. An appropriate analysis should demonstrate why the previous results remain applicable.
Therefore, the main question is not “Has the product changed?” but rather “Which characteristics and requirements are affected by this change?”
What Is Impact Analysis?
Impact Analysis is a process for evaluating how a change affects relevant parts of the product and existing evidence. The change must first be clearly defined, after which its relationship with requirements, component interfaces, failure modes, Verification activities, manufacturing processes, and, where relevant, existing approvals must be examined.
For example, if a component's material has changed, it is necessary to determine which material properties are important to product performance and which requirements may be affected. If the software has changed, the potentially affected functions and behaviors must be assessed. If the supplier has changed, it is necessary to establish whether the design definition, material, manufacturing process, or other relevant characteristics have also changed.
The purpose of Impact Analysis is not to identify a mandatory retest for every change. Its purpose is to ensure that decisions about retaining or repeating evidence are based on clear, technically justified, and traceable relationships.
First Step: Define the Change Precisely
Statements such as “we only made a small change” or “the supplier has changed” are not sufficient for making retesting decisions. The change must be defined in enough detail to allow its impact to be evaluated.
For geometric changes, the modified dimensions, thicknesses, radii, locations, or component features should be identified. For material changes, the differences between the previous and new specifications must be clear. For software changes, the modified functions and new configuration should be identifiable. For manufacturing process changes, the affected production, assembly, or control operations must be specified.
The less clearly a change is defined, the weaker its Impact Analysis will be. Engineering Change Requests, Engineering Change Notices (ECNs), or other change-control documents may be used to record this information according to the organization's processes.
Do All Design Changes Have the Same Impact?
No. Design changes can range from minor cosmetic modifications to major changes in product architecture. However, even the physical size of a change is not, by itself, a reliable indicator of its significance. A very small geometric modification in a sensitive area may affect stress concentration or component interfaces, while a larger change in an area unrelated to the requirement may have no effect on the previous result.
Therefore, general rules such as “changes of less than five percent do not require retesting” are not technically defensible without a specific reference. The significance of a change must be evaluated in relation to the relevant function and requirement.
This approach helps avoid two mistakes: unnecessarily repeating all tests after every design revision and using old results without assessment for a product whose relevant characteristics have changed.
When Does a Material Change Require Re-evaluation?
A material change can affect various product characteristics, and the nature of its impact depends on the component and the applicable requirement. For a structural component, mechanical properties may be critical. For another component, thermal, electrical, chemical, or environmental performance may be more important.
Therefore, stating that “the component dimensions have not changed; only the material is different” is not sufficient to decide whether retesting is necessary. It must be determined which material properties affect the product requirements and whether the new material specifications could alter the relevant behavior.
In some cases, engineering analysis or comparison of reliable data may support the decision, while in others, retesting may be necessary. The appropriate method should be selected based on the requirement and available evidence.
What Is the Impact of Changes in Thickness or Geometry?
Geometry can directly affect structural behavior, dimensional fit, component clearances, field of vision, fluid flow, heat transfer, or other characteristics. However, which tests are affected depends on the component's function and the nature of the change.
For example, changing the thickness of a structural bracket may be important for a strength requirement, but it does not necessarily mean that every product test must be repeated. Impact Analysis should identify which load paths and requirements are associated with the modified feature.
Indirect effects must also be considered. A geometric change may alter the position of an adjacent component, the assembly process, or an interface with another subsystem. Therefore, if the change has system-level consequences, the analysis should not be limited to the modified component.
Does Changing the Supplier Automatically Require Retesting?
There is no universal answer for every project. A supplier change alone does not provide sufficient information about its technical impact. It is necessary to determine how closely the new supplier's product conforms to the design definition, material specifications, manufacturing process requirements, and other controlled characteristics, as well as what new sources of variation may be introduced.
The part number and drawing may remain unchanged while the manufacturing process, tooling, raw materials, or process controls change. In such cases, the impact of these differences must be evaluated. In another project, a new supplier may manufacture the product according to the same specifications and defined qualification process, resulting in a different Verification decision.
In addition to engineering requirements, customer-specific requirements or supplier approval procedures may mandate particular activities. Therefore, the decision should not be based solely on the supplier's name changing.
What Is the Difference Between a Manufacturing Process Change and a Design Change?
In a design change, the product definition is modified, for example through changes to geometry, material specifications, or software requirements. In contrast, a Manufacturing Process Change may leave the design definition unchanged while altering how the product is manufactured.
Changes to welding methods, molding parameters, heat treatment, assembly methods, or production tooling are examples of process changes. If the final product characteristics are sensitive to the manufacturing process, such changes may affect performance even when the engineering drawing remains unchanged.
This is particularly important during PV and production because one objective of evaluating production-representative products is to demonstrate that products manufactured using the intended process meet the required performance. Therefore, change management must consider both the product definition and the manufacturing process.
How Are Software or Calibration Changes Evaluated?
In electronic systems and modern vehicles, physical modifications are not the only type of change. Software updates, algorithm modifications, or calibration changes can alter system behavior without any visible change to the vehicle's components.
To decide whether retesting is necessary, it is important to identify which parts of the software have changed, which functions and requirements depend on them, and which interfaces are affected. If the change influences a verified function, evidence from the previous version may no longer be sufficient for the new configuration.
On the other hand, not every software change requires repeating all vehicle tests. Traceability between software requirements, functions, and Verification activities helps ensure that the scope of reassessment remains targeted.
For this reason, recording software versions and calibration information as part of the test specimen configuration is particularly important.
Can Cosmetic Changes Also Require Retesting?
The term “cosmetic change” may suggest that a modification has no engineering impact, but it must first be established that the change is truly cosmetic. Changes to surface geometry, covers, trim, or exterior features may affect requirements such as component clearance, visibility, external projections, or sensor performance, depending on their location.
If the analysis demonstrates that the change is unrelated to the relevant requirements, previous results may remain applicable. However, the decision must be based on the actual impact of the change, not the label assigned to it in the change request.
Therefore, even the term “cosmetic only” does not replace Impact Analysis.
How Does DVP&R Support Retesting Decisions?
One of the main benefits of DVP&R, or Design Verification Plan and Report, is maintaining the relationship between requirements, Verification methods, specimen configurations, and results. When a change occurs, the engineering team can determine which requirements are associated with the modified feature and what evidence has already been generated for those requirements.
Suppose Revision B of a product has 30 Verification results, and a new change affects only three requirements. With appropriate traceability, the team can focus additional evaluations on those areas. Without such a structure, all 30 activities may be repeated unnecessarily, or conversely, none may be properly reviewed.
The article “What Is DVP&R and What Role Does It Play in Automotive Product Development?” explains this relationship and the role of traceability in product development.
How Should the Test Plan Be Updated After a Design Change?
Following Impact Analysis, the Test Plan should be updated according to the new decision. Some previous test results may remain applicable, some activities may require repetition, and new tests may need to be introduced to evaluate the effects of the change.
The new specimen configuration must be identified in the plan. If retesting is performed, the reason and associated requirement should be traceable. If the decision is made not to repeat a test, the technical justification should also be documented in the change management records.
The article “What Is a Test Plan? How to Prepare a Test Plan for an Automotive Project” provides further information about the relationship between requirements, specimens, acceptance criteria, and results.
Must All Tests Always Be Repeated?
No. Full retesting is not a universal response to every change. Impact Analysis may show that only some requirements are affected and that targeted Verification of those areas is sufficient. Conversely, a major change affecting product architecture or several functions may require a broader set of Verification activities.
The term Partial Re-Test refers to repeating only some testing activities, but its scope must be determined through technical analysis. The number of tests should not be reduced without technical justification merely to lower costs.
Similarly, full retesting should not be selected automatically when sufficient evidence shows that large parts of the product have not been affected. Verification activities should be planned in proportion to the risks and impacts of the change.
Can Engineering Analysis Replace Retesting?
In some circumstances, Engineering Analysis can provide appropriate evidence for evaluating the impact of a change. However, there is no universal rule stating that analysis can always or never replace testing. The Verification method must be suitable for the requirement, and regulatory or customer-specific projects may already specify the accepted method.
For example, if a particular change can be evaluated using a validated model and previous data, analysis may form part of the technical justification. However, if the requirement explicitly calls for testing or the impact of the change cannot be assessed with sufficient confidence through analysis, testing may still be necessary.
Therefore, the statement “simulation showed no problems” is not, by itself, sufficient justification for eliminating a test unless the analysis method, model validity, relevant requirement, and applicable acceptance process support that decision.
Is a Pre-Test Result Sufficient to Approve a Change?
Pre-Testing can be highly useful for reducing risk and understanding the behavior of a modified design, but its results do not necessarily replace final Verification evidence. If the specimen, setup, or test conditions differ from those of the main activity, the applicability of the result will also be limited.
For example, after modifying a bracket, the engineering team may perform a Pre-Test on a prototype to investigate its failure mode. A successful result may increase confidence and help improve the design before formal retesting. However, if the original requirement calls for specific Verification, the Pre-Test alone is not sufficient to close that requirement unless the project's process accepts such evidence.
The article “What Is a Pre-Test and How Is It Different from a Formal Test?” discusses the applications and limitations of this activity.
What Is the Impact of Design Changes During DV?
During Design Verification (DV), design changes are a natural part of the development process. If DV testing reveals a design weakness or failure, the design is modified, and the impact of that change on the Verification Plan must then be assessed.
Repeating the test associated with the failed requirement may be necessary, but the change may also affect other requirements. For example, reinforcing a structure may alter the load path, mass, or interface with another component. Therefore, simply repeating the failed test is not always sufficient; the effects of the corrective action on other product characteristics must also be assessed.
This is one reason change control is important during DV. Each design iteration must be traceable so that it is clear which results belong to which revision.
Why Are Design Changes More Sensitive During PV?
During Production Validation (PV), the product is closer to production status, and changes may affect not only the design but also production tooling, manufacturing processes, suppliers, and the Start of Production (SOP) schedule. If PV testing shows that a production-representative product does not meet a requirement, the root cause may lie in the design or manufacturing process.
Changes at this stage require careful evaluation of their impact on production readiness. A geometric modification may affect tooling, a process change may require additional process Validation, and a supplier change may involve its own qualification requirements.
The article “What Are DV and PV in the Automotive Industry? From Design Verification to Production Validation” explains the different objectives of these stages and the importance of production-representative specimens.
What Happens When a Design Changes After Type Approval?
In this situation, a clear distinction must be made between engineering Verification and the regulatory approval process. If a product has Type Approval and its design changes, the continued validity of the existing approval cannot be determined solely from the internal DVP&R.
Within the UN regulatory framework, specific requirements may apply to modifications of the approved product Type and extensions of approval. Depending on the regulation and the nature of the change, it may be necessary to notify the Type Approval Authority, which determines whether the approval can continue without additional testing, whether an Extension of Approval is required, or whether further tests must be performed.
The details vary between regulations, and the applicable regulatory text must be reviewed. Therefore, a general rule such as “minor changes do not need to be reported” cannot be applied to all UN approvals.
The article “What Is the Difference Between Type Approval and Homologation?” explains the roles of the Type Approval Authority, Technical Services, and product configuration in the approval process.
Does Changing a Part Number Invalidate All Previous Results?
Not necessarily. A part number change may result from a significant technical modification or simply from the organization's configuration management structure. Therefore, the part number alone does not provide sufficient information to determine the technical impact.
However, part numbers and design revisions are essential for traceability. If a test result relates to the previous part number, it must be clear why and on what basis that result can also be used for the new part. This justification may involve Impact Analysis, technical comparison, or other evidence accepted by the project's processes.
Therefore, a part number change is a trigger for review, not an automatic requirement for retesting.
What If Only the Drawing Revision Changes?
In this case, the content of the change must also be examined. A new drawing revision may introduce a technical requirement modification or may only correct notes, formatting, or information that has no effect on the product.
For this reason, decisions should not be based solely on the drawing revision number. Engineering change records should identify what has changed compared with the previous version and what the impact is.
If the change is purely administrative, no additional Verification may be required. However, if a relevant characteristic has been modified, the scope of the assessment should be determined accordingly.
How Does a Previous Failure Affect the Retesting Decision?
If a change was introduced to correct a failure, Verification must at least demonstrate that the identified issue has been resolved and the relevant requirement is now satisfied. However, the corrective action may also affect other areas, which must be examined through Impact Analysis.
Furthermore, successfully passing a retest does not, by itself, prove that the Root Cause Analysis was correct. If the original failure was caused by specimen variability, a passing result from another specimen may simply reflect differences between specimens. Therefore, for significant failures, the relationship between the root cause, corrective action, and verification evidence must be clear.
The objective of retesting should not simply be to obtain a Pass result. It should demonstrate the effectiveness of the correction and continued compliance with the relevant requirements.
How Does Change Management Reduce Testing Costs?
Effective change management does not mean performing the fewest possible tests at any cost. Its value lies in making Verification activities more targeted. When requirements and results are traceable, the team can identify which evidence has been affected by the change and focus resources on those areas.
Without this structure, a project may repeat every test as a precaution or, conversely, underestimate the significance of a change and omit necessary verification. Both approaches can result in substantial costs.
DVP&R, Test Plans, configuration management, and change records work together to make retesting decisions traceable and technically defensible.
What Questions Should Be Answered Before Deciding on Retesting?
Before making a decision, the project team should be able to identify exactly what has changed, why it has changed, which functions and requirements depend on it, what evidence exists for the previous revision, and how the change affects the assumptions on which that evidence was based.
The team must then determine whether the relevant requirement or process mandates a particular Verification method, whether engineering analysis can provide sufficient evidence, whether a new specimen is needed, and whether the change affects approvals, supplier qualification, or the manufacturing process.
These questions should preferably be answered before a new specimen is manufactured. If the specimen is built first and Verification activities are planned afterward, its configuration or quantity may not be suitable for the required tests.
For projects where the impact of a change on requirements or the Test Plan remains unclear, TAT Engineering and Consulting Services can support requirements review and test planning. The TAT Table of Available Tests is also available for reviewing relevant tests and standards.
After a Design Change, the Main Question Is Not “Which Test Should We Repeat?”
Retest planning does not begin with selecting a test. It begins with understanding the change and its impact. Every previous result relates to a specific requirement, specimen, and configuration. When the product changes, it is necessary to determine whether the assumptions that supported the previous evidence remain valid.
Sometimes this assessment leads to full retesting; sometimes only certain Verification activities need to be repeated; and in other cases, engineering analysis or existing evidence may be sufficient to retain the validity of previous results. None of these decisions should be based solely on the apparent size of the change, testing costs, or experience from a previous project.
Regulatory projects introduce another consideration: the applicable regulation and the requirements of the Type Approval Authority. A change may be minor from a product development perspective but still require formal review from an approval standpoint.
Therefore, the appropriate engineering question after modifying a product is: “Which requirements and previous evidence are affected by this change, and what Verification activities are necessary to establish sufficient confidence in the new configuration?” When the answer is based on traceability and Impact Analysis, retesting becomes a controlled part of change management rather than a decision based on assumptions or past experience.