Project

Supplier Change Qualification and Production Readiness Project

Industrial engineering project for qualifying a supplier change with requirement traceability, first-build evidence, capability screening, incoming inspection, ramp capacity, risk controls, and release gates.

This project prepares a supplier change qualification and production readiness package. The deliverable is the evidence file an engineering, quality, manufacturing, supply chain, and operations team would use to decide whether a new or changed supplier can be released into production.

A supplier change is not only a purchasing action. It can change material condition, process route, inspection method, packaging, lot traceability, dimensional capability, defect escape risk, logistics behavior, operator workload, and production recovery time. The project therefore treats the supplier as part of the engineering system, not as an external price quote.

The central project question is:

Can the proposed supplier change be released without creating unacceptable quality, delivery, safety, traceability, or production-readiness risk?

The answer must be evidence-based. A clean purchase order, a good price, or one acceptable sample is not enough.

Project Objective

Prepare a supplier-change qualification package for a critical machined bracket used in an electromechanical assembly. The final deliverable must include:

  1. change scope and affected requirements;
  2. urgency and stockout exposure;
  3. first-build and dimensional evidence;
  4. capability screen for critical characteristics;
  5. incoming inspection and containment rule;
  6. ramp-capacity check;
  7. supplier risk controls and traceability;
  8. readiness score with hard gates;
  9. final release, conditional release, or hold decision.

The project is written for industrial and management engineering students and early-career engineers. The numbers are simplified, but the logic matches real production change control: release is credible only when requirement, product, process, logistics, quality, and operations evidence point to the same decision.

Baseline Scenario

A factory builds motorized actuator modules. The current supplier of the aluminum support bracket has a capacity constraint after a tooling failure. A qualified alternate supplier is proposed. The alternate supplier uses the same alloy and drawing revision, but a different machining fixture, different anodizing subcontractor, different packaging, and a different logistics route.

The bracket is not safety-critical by itself, but loss of dimensional control can cause bearing misalignment, high motor current, assembly rework, noise, and early warranty returns.

ItemValue
weekly customer demand2400\ \text{modules/week}
current supplier available output1400\ \text{brackets/week}
finished bracket inventory4500\ \text{brackets}
alternate supplier pilot lot125\ \text{brackets}
target qualification lead time3\ \text{weeks}
production release requirementC_{pk}\geq1.33 on critical dimensions
incoming containment for first three lotstightened inspection with c=0 for critical defects
maximum allowed packaging damage0.2\% during transport trial

The proposed supplier change affects four engineering domains:

  1. Product evidence: drawing revision, material grade, critical dimensions, surface finish, coating thickness, cleanliness, and functional fit.
  2. Process evidence: machining fixture, tool control, anodizing subcontractor, inspection method, operator training, and reaction plan.
  3. Supply-chain evidence: capacity, lot size, packaging, route time, receiving inspection, labeling, and escalation.
  4. Operations evidence: assembly fit, line-side presentation, traceability, nonconformance containment, and ramp stability.

Deliverable Structure

The supplier-change file should contain:

SectionEvidence required
change requestreason for change, affected part number, drawing revision, supplier process route
requirement mapcritical characteristics, test method, acceptance limit, owner, record location
risk assessmentfailure modes, severity, occurrence, detection, containment and prevention actions
first-build evidencepilot build data, material certificates, dimensional results, coating and functional fit
capability screensample size, mean, standard deviation, C_p, C_{pk}, stability notes
logistics readinesspackaging trial, labeling, lot traceability, transport route, receiving workflow
ramp planrate trial, capacity margin, inventory bridge, inspection escalation rule
release decisionopen gates, accepted risks, hold conditions, follow-up monitoring

The package should be short enough to review, but complete enough that a later quality escape, line stop, or customer issue can be traced back to the release assumptions.

Step 1: Define the Change Boundary

The first engineering task is to state what is actually changing. “New supplier” is too broad. A useful boundary lists changed and unchanged variables.

ElementStatusEngineering meaning
part numberunchangedexisting drawing remains the requirement source
drawing revisionunchangedno design release expected
alloy specificationunchangedmaterial certificate must still be verified
machining fixturechangeddimensional capability must be re-established
anodizing subcontractorchangedcoating thickness, color, corrosion and masking evidence required
packagingchangeddamage and label-traceability risk must be tested
logistics routechangedlead time and handling exposure must be validated
assembly stationunchangedfit trial can use the existing station and work instruction

Commentary: a supplier change often looks small until the process route is decomposed. The supplier name is not the controlled variable. The controlled variables are material, geometry, process, measurement, handling, records, and reaction rules.

Step 2: Screen Stockout Exposure

The current supplier cannot meet demand. Weekly demand is:

D=2400\ \text{brackets/week}

Current supplier output is:

Q_c=1400\ \text{brackets/week}

The weekly supply gap is:

G=D-Q_c

Substitute:

G=2400-1400=1000\ \text{brackets/week}

Inventory is:

I_0=4500\ \text{brackets}

Time until stockout is:

\displaystyle t_s=\frac{I_0}{G}

Substitute:

\displaystyle t_s=\frac{4500}{1000}=4.5\ \text{weeks}

The alternate supplier qualification target is 3\ \text{weeks}, so the nominal inventory bridge is:

M_t=4.5-3.0=1.5\ \text{weeks}

Engineering Comment

The supplier change is urgent, but not uncontrolled. A 1.5\ \text{week} bridge gives time for evidence review if qualification stays on schedule. If capability correction, packaging redesign, or audit delay consumes that margin, the release decision becomes a production-continuity risk.

The correct response is not to skip qualification. It is to run qualification work in parallel, define hard gates early, and protect operations with inventory and inspection controls.

Step 3: Map Critical Characteristics

The bracket has many dimensions, but not every dimension should drive the same release burden. The project classifies characteristics by consequence.

CharacteristicRequirementRisk if uncontrolledRelease evidence
bearing bore center distance80.00\pm0.20\ \text{mm}misalignment, motor current, noiseC_{pk}, fixture check, assembly fit
mounting flatness\leq0.08\ \text{mm}rocking, bolt preload variationsurface plate or CMM evidence
coating thickness12 to 20\ \mu\text{m}corrosion, assembly interferencecoating record, thickness sample
thread qualitygo/no-go gage passassembly cross-thread, line stopthread gage and torque trial
packaging orientationno metal-to-metal contactdents, coating damagetransport trial and receiving check
lot traceabilityheat, machining batch, coating batchcontainment failurelabel and record audit

Commentary: this table prevents a common mistake: treating every drawing note as equally important while missing the few characteristics that control field performance, assembly flow, or containment.

Step 4: Capability Screen for a Critical Dimension

The critical bore center distance has:

LSL=79.80\ \text{mm}
USL=80.20\ \text{mm}

The pilot lot gives:

\bar{x}_1=80.04\ \text{mm}
s_1=0.045\ \text{mm}

Potential capability is:

\displaystyle C_p=\frac{USL-LSL}{6s}

Substitute:

\displaystyle C_p=\frac{80.20-79.80}{6(0.045)}=\frac{0.40}{0.270}=1.48

Upper-side capability is:

\displaystyle C_{pu}=\frac{USL-\bar{x}}{3s}
\displaystyle C_{pu}=\frac{80.20-80.04}{3(0.045)}=\frac{0.16}{0.135}=1.19

Lower-side capability is:

\displaystyle C_{pl}=\frac{\bar{x}-LSL}{3s}
\displaystyle C_{pl}=\frac{80.04-79.80}{0.135}=\frac{0.24}{0.135}=1.78

Therefore:

C_{pk}=\min(1.19,1.78)=1.19

Engineering Comment

The variation is not the only issue. C_p=1.48 suggests the spread could be acceptable, but the process is shifted toward the upper limit, so C_{pk}=1.19 fails the 1.33 release target.

This is exactly why a supplier qualification should not stop at “all parts passed.” The pilot parts may be conforming, but the process is not centered enough for production release.

Step 5: Correct and Recheck Capability

The supplier adjusts the machining fixture offset and repeats a validation run. The revised data are:

\bar{x}_2=80.01\ \text{mm}
s_2=0.040\ \text{mm}

Recalculate:

\displaystyle C_p=\frac{0.40}{6(0.040)}=\frac{0.40}{0.240}=1.67

Upper-side capability:

\displaystyle C_{pu}=\frac{80.20-80.01}{3(0.040)}=\frac{0.19}{0.120}=1.58

Lower-side capability:

\displaystyle C_{pl}=\frac{80.01-79.80}{0.120}=\frac{0.21}{0.120}=1.75

Therefore:

C_{pk}=1.58

Engineering Comment

The corrected process passes the simplified capability target. The release file should still note limitations:

  • the sample must represent the intended machine, fixture, tooling, operator method, material, and inspection system;
  • capability assumes the process remains stable;
  • the first production lots should remain under tightened incoming control;
  • the reaction plan must define what happens if the mean drifts back toward the upper limit.

Capability is release evidence, not a permanent guarantee.

Step 6: Incoming Inspection Decision Rule

For the first three lots, the team proposes sampling:

n=80,\quad c=0

where n is the inspected sample size and c=0 means the lot is rejected or held if any critical defect is found.

If the true critical defect probability is p, the probability of accepting the lot is:

P_a=(1-p)^n

For a poor incoming lot with:

p=0.01

the acceptance probability is:

P_a=(1-0.01)^{80}=0.99^{80}=0.448

For a much better lot with:

p=0.001

the acceptance probability is:

P_a=0.999^{80}=0.923

Engineering Comment

The c=0 plan is a useful containment screen, but it is not proof that every accepted lot is safe. At a true 1\% defect rate, the plan still accepts about 44.8\% of lots because the sample may not include a defective unit.

For critical characteristics, sampling must be combined with process evidence: capability data, fixture control, material traceability, supplier reaction plans, first-article review, and early production monitoring.

Step 7: Ramp Capacity Check

The alternate supplier must support:

D_w=2400\ \text{brackets/week}

The supplier proposes one machining cell with:

ParameterValue
working days5\ \text{days/week}
net time per day420\ \text{min/day}
nominal cycle time42\ \text{s/bracket}
expected OEE during ramp0.78

Available weekly time is:

T_w=5(420)(60)=126000\ \text{s/week}

Nominal weekly capacity is:

\displaystyle Q_{nom}=\frac{T_w}{42}=\frac{126000}{42}=3000\ \text{brackets/week}

Effective capacity during ramp is:

Q_{eff}=Q_{nom}(OEE)
Q_{eff}=3000(0.78)=2340\ \text{brackets/week}

Capacity shortfall:

\Delta Q=2400-2340=60\ \text{brackets/week}

The supplier proposes 2\ \text{h/week} of planned overtime during the first month:

T_{ot}=2(3600)=7200\ \text{s/week}

Additional effective output:

\displaystyle Q_{ot}=\frac{7200}{42}(0.78)=134\ \text{brackets/week}

Adjusted capacity:

Q_{adj}=2340+134=2474\ \text{brackets/week}

Capacity margin:

M_Q=2474-2400=74\ \text{brackets/week}

Engineering Comment

The supplier can meet the ramp only with a small overtime margin. That is acceptable for a temporary bridge, but weak for long-term production. The release decision should include a follow-up action: either improve OEE, qualify a second fixture, reduce changeover loss, or protect demand with a controlled inventory buffer.

Capacity evidence should be based on good units, not machine run time alone.

Step 8: Risk Priority Review

A simplified FMEA table identifies major supplier-change failure modes.

Failure modeEffectInitial SInitial OInitial DInitial RPN
bore center driftassembly rework, motor current rise844128
wrong alloy or heat lotstrength or corrosion risk92590
coating thickness out of rangeinterference or corrosion645120
packaging damagecosmetic reject, handling delay45480
lot traceability breakcontainment failure836144

The highest initial RPN is lot traceability break:

RPN=8(3)(6)=144

The proposed controls are:

  • barcode label linking receiving lot, machining batch, coating batch, and material heat;
  • blocked receiving status until label scan and certificate match;
  • first three lots reviewed by quality engineering before release;
  • supplier shipment checklist with photo record of packaging and labels;
  • reaction rule for missing records: hold lot, do not consume.

With controls, score traceability as:

S=8,\quad O=2,\quad D=2

Then:

RPN_{controlled}=8(2)(2)=32

Engineering Comment

Severity usually does not change because the consequence of a traceability failure remains serious. The controls reduce occurrence and improve detection. The project should not claim that risk is gone. It should show that the release decision is now supported by a controlled detection and containment path.

Step 9: Readiness Score and Hard Gates

A readiness score helps organize the review, but hard gates control release.

Readiness itemWeightScore before closure
drawing and revision confirmed55
material certificate and heat traceability54
first-build dimensional report55
critical-dimension capability53
coating thickness and masking evidence43
incoming inspection plan44
packaging transport trial42
supplier ramp-capacity evidence43
label and lot-traceability audit44
assembly fit trial44

Total possible score:

R_{max}=5+5+5+5+4+4+4+4+4+4=44

Actual score:

R=5+4+5+3+3+4+2+3+4+4=37

Readiness index:

\displaystyle RI=\frac{37}{44}=0.841

Engineering Comment

An 84.1\% readiness score is not enough by itself. Two items are hard gates:

  1. critical-dimension capability must pass after centering correction;
  2. packaging trial must demonstrate damage below the allowed limit.

If those gates are open, the supplier should not receive unconditional production release even if the average score looks acceptable.

After the offset correction, capability evidence improves from 3 to 5. After packaging separators are added and the transport trial passes, packaging improves from 2 to 4. The revised readiness score is:

R_{closed}=37+2+2=41
\displaystyle RI_{closed}=\frac{41}{44}=0.932

The supplier is now above a practical 90\% readiness threshold and the hard gates are closed.

Step 10: Release Decision

The recommended decision is:

Conditional release for the first three production lots, followed by normal release only if capability, incoming inspection, traceability, packaging, and assembly feedback remain stable.

Conditions:

  1. use corrected fixture offset from the second validation run;
  2. hold first three lots under n=80, c=0 incoming inspection for critical characteristics;
  3. verify every incoming lot has material heat, machining batch, coating batch, and shipment label traceability;
  4. maintain packaging separators and photo evidence until damage data are stable;
  5. review assembly fit, motor current, rework, and bracket defect records daily during ramp;
  6. require quality engineering signoff before moving from conditional to normal release.

The project does not recommend emergency bypass. The stockout margin is narrow, but not narrow enough to justify releasing an uncentered process or unproven packaging.

Final Deliverable

The final supplier-change qualification file should include:

  • controlled part number, revision, supplier process route, and change reason;
  • matrix of critical characteristics, acceptance criteria, evidence source, and owner;
  • first-build report with material, dimensional, coating, thread, packaging, and assembly evidence;
  • capability calculations for critical dimensions before and after correction;
  • incoming inspection plan with explicit accept, reject, hold, and escalation rules;
  • ramp-capacity calculation using effective good-unit capacity;
  • supplier risk table with controls and residual RPN;
  • readiness score, hard gates, release decision, and follow-up monitoring plan;
  • record-retention path for lots produced under conditional release.

Common Mistakes

  • Treating supplier approval as a commercial task instead of an engineering release.
  • Accepting one conforming sample without checking process centering, variation, traceability, and ramp capacity.
  • Calculating C_p but ignoring C_{pk} and process offset.
  • Using incoming inspection as the only control for a critical characteristic.
  • Forgetting that packaging and labels are part of product quality when damage or traceability loss can stop production.
  • Releasing a supplier because the readiness score is high while a hard gate remains open.
  • Failing to define who can unblock a held lot and what evidence they must review.
  • Removing containment before the first production lots show stable capability and assembly feedback.

Validation Checklist

Before final release, confirm:

  1. drawing revision, specification and supplier process route match the approved change;
  2. measurement method is adequate for the tolerance being judged;
  3. first-build samples represent intended production conditions;
  4. critical dimensions meet capability targets with stable centering;
  5. material, coating, packaging and traceability records are complete;
  6. incoming inspection plan has a clear decision rule and owner;
  7. ramp capacity is based on good units after downtime and yield loss;
  8. assembly trial uses production operators, tools and work instructions;
  9. nonconformance containment is defined before the first shipment;
  10. release conditions, follow-up checks and escalation paths are written.

Transferable Lesson

Supplier qualification is a systems problem. The supplier may make a good part, but production readiness also depends on stable process evidence, credible measurement, lot traceability, incoming controls, packaging, logistics, capacity margin, assembly feedback, and clear reaction rules.

The engineering discipline is to connect those elements into one release decision. A supplier change is ready when the evidence shows that the product can be built, inspected, shipped, received, assembled, traced, contained, and improved under normal production conditions.

REF

See also