Actual site
Is this the actual site that will perform the work?
Robotics Factory Assessment Guide
How to verify whether a specific production site can manufacture, calibrate, test and scale a defined robotics product
A factory tour can show that equipment exists.
A capability assessment must determine whether the actual site can repeatedly manufacture, configure, calibrate, test and release the required robotics product under controlled conditions.
This guide explains what evidence should be reviewed before a factory is approved for a defined manufacturing scope.
A factory capability assessment is an evidence-based evaluation of whether a specific production site can perform the processes required to manufacture a defined product or component.
It examines the factory’s engineering support, equipment, tooling, operators, process controls, measurement systems, calibration, testing, quality records, traceability, maintenance and usable capacity.
The result should approve, conditionally approve or reject a defined scope. It should not approve a factory in the abstract.
Factory capability cannot be established from a certificate, equipment list or factory tour alone. It requires product-specific evidence that the actual production site can control the required processes, people, equipment, measurement systems, calibration, testing, traceability and usable capacity. See the parent robotics manufacturing lifecycle for how factory assessment fits production readiness.
| Activity | Main question | Typical scope |
|---|---|---|
| Supplier qualification | Should the organization be approved? | Identity, ownership, capability, risk, quality and commercial suitability |
| Factory capability assessment | Can this site perform the required manufacturing scope? | Processes, equipment, people, measurement, testing and capacity |
| Factory audit | What evidence does the site provide against defined criteria? | Structured on-site or remote evidence gathering |
| Process audit | Is a defined manufacturing process adequately controlled? | One process or connected process sequence |
| Product audit | Does output meet the product requirements? | Product, sample, batch or unit |
| Product inspection | Does the inspected product conform at this point in time? | Specified units or lot |
| Social-compliance audit | Does the facility meet defined labor or ethical criteria? | Employment and workplace practices |
| Environmental audit | Does the facility meet defined environmental criteria? | Environmental system and compliance |
A factory audit is a method. Factory capability is the decision that the collected evidence must support.
provides current guidance on audit principles, audit-program management, audit conduct and auditor competence. It is guidance for auditing management systems, not a product-specific factory-capability certification.
Organization-level approval belongs to qualifying the supplier organization. Control plans, inspection and corrective action depth belong to building a robotics production quality system.
The assessment should answer ten questions before a manufacturing scope is approved.
Is this the actual site that will perform the work?
Does the factory understand the product requirements?
Are all required processes internal, controlled or transparently outsourced?
Does the equipment have the necessary range, accuracy and condition?
Are tooling, fixtures and work instructions suitable for repeatable production?
Are the people performing critical operations demonstrably competent?
Can the measurement systems reliably evaluate the required characteristics?
Can the factory configure, calibrate and test the product correctly?
Can the site maintain traceability and control changes?
Can it achieve the required output without losing process control?
A factory may possess suitable equipment without having demonstrated that its complete manufacturing system can repeatedly meet the relevant robot specification.
A factory is not simply capable or incapable.
Capability exists in relation to a defined product, process, tolerance, material, production volume, test requirement and risk level.
A site may be capable of assembling a service robot but not manufacturing precision robot joints. It may be capable of machining housings but not validating gearbox geometry or robot accuracy.
For every step, record the process owner, inputs, outputs, equipment, tooling, operator qualification, parameters, inspection, records, failure reaction, subcontractor dependency and capacity.
Compare the documented process flow with what actually occurs on the factory floor. Capability gaps are often hidden in informal rework, shared equipment, manual adjustments or outsourced steps that do not appear in the original process map.
| Evidence level | Example | Relative strength |
|---|---|---|
| Claim | “We can hold this tolerance.” | Weak |
| Document | Procedure or equipment specification | Limited without implementation evidence |
| Observation | Process seen on the factory floor | Stronger but time-limited |
| Record | Historical measurement, maintenance or production record | Strong |
| Demonstration | Controlled trial using representative requirements | Strongest for unproven capability |
Capability decisions should not rely on a supplier statement when stronger evidence can reasonably be obtained.
Evaluate whether the site can interpret drawings and specifications; perform design-for-manufacturing and process-feasibility reviews; translate critical characteristics into process controls; develop process flows; select suitable equipment; design tooling and fixtures; allocate tolerances; diagnose manufacturing failures; develop inspection methods; support software provisioning; develop calibration methods and production tests; implement engineering changes; support product variants; and manage product and process documentation.
For design-transfer and industrialization context, see moving a robot prototype into production.
For each critical machine, record the process, equipment category, manufacturer, model, serial number, age, working range, rated accuracy, demonstrated performance, control system, maintenance status, calibration status, utilization, backup resource and qualified operators.
Assess tooling design ownership, manufacture, revision, identification, storage, preventive maintenance, expected life, actual wear, spare tooling, repair capability, calibration where applicable, transfer rights and capacity.
A fixture that works only with experienced manual adjustment may support a prototype build but may not provide a controlled production process.
Review manufacturing engineers, process engineers, quality engineers, software or firmware engineers, calibration technicians, test engineers, maintenance technicians, inspectors, production supervisors, critical-process operators and assembly operators.
For each critical process, verify input and output requirements, process sequence, equipment, tooling, materials, operator qualification, controlled parameters, environmental conditions, measurement method, inspection frequency, acceptance limits, records, nonconformance reaction, change control, maintenance and capacity.
A process audit should connect documented requirements to actual factory practice and objective production evidence.
VDA 6.3 is one recognized process-audit framework, particularly in automotive supply chains. It may be referenced as an example of a risk-oriented process-audit approach, but it should not be presented as mandatory for all robotics factories.
Process capability evaluates how the output of a stable manufacturing process compares with defined specification limits.
It is not established from machine specifications alone.
The data should represent the actual process, material, equipment, operators and conditions relevant to the manufacturing scope.
Process capability should not be calculated first and justified afterward. Measurement validity and process stability must be evaluated before the capability index is treated as decision evidence.
NIST describes process capability as comparing the output of an in-control process with specification limits through indices such as Cp and Cpk.
Cp compares the specification width with estimated process variation. Cpk also considers whether the process mean is centred between the specification limits.
Cp = (USL − LSL) ÷ 6σ
Cpk = min[(USL − μ) ÷ 3σ , (μ − LSL) ÷ 3σ]
Where USL is the upper specification limit, LSL is the lower specification limit, μ is the process mean and σ is the estimated process standard deviation.
NIST notes that capability indices depend on assumptions about process stability, data quantity and distribution. This guide does not apply a universal Cpk threshold without context.
Measurement system analysis evaluates whether the complete measurement process can generate sufficiently reliable data for the intended decision.
The measurement system includes the instrument, fixture, method, software, operator, reference, environment and part interaction.
AIAG describes measurement system analysis as a set of methods used to assess measurement-system quality so that manufacturing decisions are based on suitable data.
A calibrated instrument does not automatically prove that the complete measurement system is suitable for the application.
Gage repeatability and reproducibility is a measurement-system study used to estimate how much observed variation comes from the measuring equipment and from differences between appraisers or measurement conditions.
It is one part of measurement system analysis, not the complete MSA framework.
Do not treat gage R&R as a standalone decision topic. The purpose of this section is to determine whether factory measurement evidence can be trusted within the broader capability assessment.
Review the measurement-equipment register, calibration intervals and status, reference standards, traceability chain, out-of-calibration response, environmental control, internal and external laboratory competence, measurement uncertainty where relevant, software version and record retention.
specifies requirements for a measurement-management system intended to provide confidence in the validity and reliability of measurement results. addresses competence, impartiality and consistent operation of testing and calibration laboratories.
For this guide, treat a process as special or critical where failure has a significant product or safety consequence; output cannot be fully verified later; verification would be destructive; process variation is difficult to detect at final inspection; or performance depends strongly on controlled parameters or operator technique.
AIAG publishes special-process assessment frameworks for several automotive processes. These can be used as reference structures where relevant, but they are not automatically mandatory for robotics manufacturing.
No.
ISO 9001 certification can provide evidence that a defined quality management system has been assessed.
It does not by itself prove that a specific factory has the engineering, equipment, process capability, measurement systems, calibration, testing or capacity required for a particular robotics product.
remains the current published requirements standard in July 2026, with its replacement expected in September 2026. The scope, site and certified entity should be checked rather than relying on the certificate logo alone.
Detailed inspection, nonconformance and corrective-action frameworks are covered in robotics quality systems.
Review approved suppliers and manufacturers, purchase specifications, incoming verification, material certificates, lot traceability, storage conditions, shelf life, counterfeit prevention, supplier changes, nonconforming incoming material and sub-tier process approval.
Organization-level sub-tier supplier qualification should follow robotics supplier qualification. Component-level sourcing context is covered under robot components.
Factory capability depends on the sustained condition of production, measurement and test equipment.
A machine that achieved the required result once may not continue to do so without controlled maintenance, tooling replacement, calibration and failure-response processes.
Manufacturing traceability connects the finished robot or component to the materials, components, process conditions, software, calibration, tests and rework history used to create it.
Select one finished unit. Can the factory trace backward to critical components, process records, calibration, software version, test results and rework?
Then select one critical supplier lot. Can the factory trace forward to every affected production unit?
Production capacity is the output a factory can achieve under defined operating conditions.
Usable capacity must consider yield, downtime, maintenance, changeover, testing, calibration, rework, material availability and the constrained operation—not only nominal machine speed.
Usable output ≈ Available production time ÷ Effective constrained-cycle time
The model must include assumptions and should not be presented as a production guarantee.
Final assembly is not automatically the capacity constraint. A short, specialized calibration or test operation may limit the output of the complete factory.
Production-line readiness is the demonstrated availability and control of the resources required to execute the released manufacturing process.
It is more specific than general factory capability and should be assessed for the relevant product, line, volume and launch stage.
| Concept | Question |
|---|---|
| Product readiness | Is the design ready? |
| Supplier readiness | Is the organization approved? |
| Factory capability | Can the site perform the required scope? |
| Line readiness | Are the required production resources prepared now? |
| Production readiness | Can the complete system proceed to the next production stage? |
See prototype to production for DFM, pilot builds and ramp sequencing.
Possible calibration scopes include joint-zero calibration, encoder-offset calibration, robot-geometry calibration, tool-frame calibration, sensor calibration, force-torque sensor zeroing, camera-to-robot calibration, mobile-base calibration, payload configuration and battery or energy calibration.
Calibration fixture exists ≠ calibration process is controlled. A calibration result produced ≠ the result is sufficiently accurate and repeatable. One calibrated sample ≠ production calibration capability.
Where industrial robot performance claims are involved, provides performance criteria and related test methods for manipulating industrial robots. Safety scope may also involve and .
Review test requirements, coverage, sequence, limits, reference equipment, fixtures, software, software revision control, hardware compatibility, data retention, false-pass and false-fail risk, failure reaction, rework and retest.
End-of-line testing verifies the individual production unit. It does not replace validation of the design, manufacturing process or test system itself. Sampling plans such as may support lot inspection decisions but do not substitute for process capability.
| Subsystem | Factory capabilities to examine |
|---|---|
| Mechanical structure | Machining, casting, welding, dimensional inspection, surface treatment |
| Robot joints | Bearing fits, reducer alignment, motor integration, encoder alignment, lubrication, calibration |
| Motors | Winding, magnet installation, balancing, electrical testing, thermal testing |
| Reducers | Gear manufacture, heat treatment, precision metrology, backlash testing, endurance |
| Controllers | PCB assembly, programming, electrical test, firmware control, traceability |
| Sensors | Assembly, alignment, calibration, environmental testing, signal verification |
| Cables and harnesses | Crimping, shielding, routing, continuity, flex-life controls |
| Batteries and power systems | Cell traceability, joining process, BMS programming, safety testing |
| Machine vision | Camera mounting, optics handling, lighting, intrinsic and hand-eye calibration |
| Final robot assembly | Mechanical integration, software provisioning, configuration, calibration, functional test |
| Mobile robot platform | Drive alignment, navigation sensors, battery system, localization calibration |
| End-effector system | Interface geometry, pneumatic or electrical test, load and functional verification |
Each finding should record the assessment criterion, observed condition, objective evidence, requirement or expected control, capability impact, risk, required action, owner, due date, verification method and qualification consequence.
Adequate relevant evidence was observed.
No current nonconformity, but future risk or weakness exists.
Limited weakness that does not currently invalidate the complete scope.
Systemic or significant weakness affecting capability confidence.
Evidence indicates that the site cannot safely or reliably perform the scope.
These categories are the Yana reporting model for factory-capability decisions. They are not presented as universal audit-standard definitions.
Relevant capability has been demonstrated for the defined scope.
Capability is acceptable subject to defined controls, limitations or corrective actions.
The site may become capable, but substantial evidence or process development is still required.
The site may perform controlled prototype or pilot activity but is not approved for unrestricted serial production.
Critical capability is missing or the evidence is insufficient for the defined risk.
Do not convert the complete assessment into one artificial percentage unless the scoring model, evidence weighting, disqualifying conditions and decision rules are publicly defined.
| Failure | Likely consequence |
|---|---|
| Audit begins without product requirements | Assessment becomes generic |
| Supplier chooses the demonstration product | Evidence may not represent the intended scope |
| Equipment list accepted without verification | Capability remains unproven |
| One conforming sample treated as process evidence | Variation remains unknown |
| ISO certificate accepted as product capability | Site-specific gaps remain hidden |
| Process flow does not include subcontractors | Responsibility and traceability gaps |
| Machine specification treated as process capability | Material, tooling and operator effects are ignored |
| Capability calculated using weak measurement data | Statistical conclusions are unreliable |
| Gage R&R treated as complete MSA | Bias, stability and other issues may remain |
| Operators are interviewed but not observed | Actual practice is not verified |
| Rework area is excluded from the audit | Process instability is hidden |
| Final yield reported without rework | Process performance is overstated |
| Quoted capacity is accepted without bottleneck analysis | Production commitments fail |
| Shared equipment is treated as dedicated capacity | Availability is overstated |
| Calibration method is undocumented | Unit-to-unit variation |
| Test software is not revision-controlled | Release results may be inconsistent |
| Findings are not linked to qualification | Audit becomes administrative |
China has broad manufacturing capability across machining, castings, gears, motors, encoders, electronics, batteries, cables, sensors, robot joints and final robot assembly.
However, the commercial supplier name may not identify the factory that owns or performs each critical process.
A capability assessment should therefore establish the operational structure below the level of the sales entity.
The objective is not merely to confirm that a factory exists.
The objective is to determine which entity, site, equipment, processes and technical knowledge will be responsible for manufacturing the product and whether that complete system has demonstrated the required capability.
This page does not publish rankings of Chinese factories.
It is an evidence-based evaluation of whether a specific production site can perform the processes required to manufacture a defined product or component. See the definition section.
A factory audit is an evidence-gathering activity. A capability assessment uses that evidence to decide whether the site can perform a defined manufacturing scope. See assessment vs audit.
Supplier qualification evaluates the organization as a whole. Factory assessment evaluates the processes, resources and evidence at a specific production site. See supplier qualification.
No. A factory audit examines systems and processes. Product inspection examines specified units or a production lot at a particular point in time. See the comparison table.
No. It provides evidence relating to a defined quality management system. Product-specific engineering, equipment, processes, measurement, calibration, testing and capacity require separate evaluation. See ISO 9001 and QMS.
Manufacturing capability is the demonstrated ability of a defined production system to perform the required processes and meet the relevant product requirements. See what the assessment must prove.
Capability concerns whether the process can meet requirements. Capacity concerns whether it can do so at the required output and timing. See production capacity.
Process capability compares the variation of a stable process with defined specification limits. See process capability.
Cp compares process variation with specification width. Cpk also reflects whether the process mean is centred within the specification limits. See Cp and Cpk.
No universal value should be applied without considering the characteristic, risk, data assumptions, customer requirements and applicable industry rules. See Cp and Cpk.
Measurement system analysis evaluates whether the complete measurement process can produce sufficiently reliable data for its intended use. See MSA.
Gage repeatability and reproducibility is an MSA study used to estimate variation associated with the measuring equipment and appraisers or measurement conditions. See gage R&R.
Not necessarily. Calibration addresses the relationship to a reference. Suitability also depends on the method, fixture, operator, environment, resolution and intended decision. See calibration and metrology.
It is a process requiring additional control because failures may have serious consequences or may not be fully detectable through later inspection. See special and critical processes.
Review real cycle times, uptime, yield, rework, testing, calibration, maintenance, staffing, material supply and the constrained operation rather than relying only on nominal machine speed. See capacity and bottlenecks.
Each robot should be linked where required to its product revision, critical components, supplier lots, hardware, software, calibration, tests, rework and release records. See manufacturing traceability.
There is no universal validity period. Reassessment may be required after site, ownership, process, equipment, product, volume or critical-supplier changes, or after significant quality failures. See capability decision.
Confirm the legal and production entities, actual site, equipment ownership, internal and outsourced processes, engineering responsibility, calibration, test systems, traceability and realistic production capacity. See assessing factories in China and the supplier qualification service.
If you have defined the product or process scope and need structured factory capability assessment, process evidence review, measurement-system evaluation or China manufacturing-site verification, Yana can help build an evidence-led assessment package.