Top Banner
POLAROID A Case on Process Control Presented by: Anish Kirolikar Jaspreet Kaur Nrupen Kamtekar Prakhar Nigam Uttam Kumar Das
36

TQM Polaroid

Nov 28, 2015

Download

Documents

uttam.nift

sjjfljsjllalk
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript

Slide 1

PolaroidA Case on Process ControlPresented by:Anish KirolikarJaspreet KaurNrupen KamtekarPrakhar NigamUttam Kumar DasIntroductionPolaroid - manufactures cameras, photographic equipment and accessories

Two main divisions The Technical and Industrial DivisionThe Consumer Photograph Division

The company produced two main types of films:The peel apart filmThe integral film (R2 facility)Factory Layout1st floorSpring ProductionStamped out of sheet metal in a ten-staged metal pressing operation.Pod ProductionDesigned such that reagent would be released from the front seal of the pod.Flows between negative and positive to develop image.Operates when film frame are ejected from the camera.Packaging

2nd and 3rd floorCartridge Production and AssemblyQuality and Process Control at R2All films that were produced were checked by the Quality Control departmentSampled 15 finished cartridges out of every lot of 5000.These finished cartridges each contained 10 frames

If defects were found,Lot was put on hold for further testingEither reworked or rejected

Subsequent lots were subjected to more intense and rigorous testing Also increased the sample size on these lotsQuality and Process Control at R2The QC department ended up rejecting just over 1% (50 cartridges per 5,000) in 1984

Used to check 32 random cartridges for each lot at each stage of the process

Took specific measurements of different characteristics

Measurement combined with the operators knowledge were used to determine if a lot should be rejected or not

Problems with existing quality controlTesting of the cartridges was very destructive and expensiveResulted in sample crap - $1.28 mnRejected lots additional $2 mn

Handling of the cartridges increased the chances of developing defectsPassed repackaged handled again increased defects

Use of a perfect testing camera did not match customers cameras

Sampling did nothing to improve qualityLead to high costsIf the control sampling was cut in half, the outgoing defects would be 0.03% of production.

Project GreenlightFocused on improving the quality control processes while reducing the number of samples

3 parts : Adoption of statistical process control principles

Production operators would be given the process control tools that the process engineering technicians had been usingexpected to make disposition decisions themselves

Quality control auditors concentrate on training operatorsoperationalizing specifications on new products

Project GreenlightOperators responsible for taking measurements and recording them

If the machine was operating outside of the control limits, the machine was automatically shut down

Maintenance would be called and the machines had to be cleaned, re-calibrated and restarted

Operators were no longer allowed to tweak the machines.

Results Project GreenlightMixed results Was supposed to cut down the defects and testing losses. The reported defective rate from the operators and auditors was reduced from 1% to 0.05% The defective rate from the central process auditors had shot up from 1% to 10%

Lack of trust between the auditors and the operators

The nature of defects also changedThe variability in the kinds of defects detected increased

Integral Film Production+ Film FrameNegativePositive+ Thick long strips were placed between the two+ Craft Material at the top+ Lamination+ Top Cover Sheet+ Pod+ Spring+ BatteryPlastic BoxPlastic Box Subassembly+ End CapsCartridgeProcess Quality & Capability levelAt 1% defect level10000 DPMO.process would be under 4.1 sigma level.

ZL = (-LSL)/sigma = 2.33

Process Capability Cpk = Min(ZL,ZU)/3

Cpk = 2.33/3 = 0.7766

Going beyond 1%Customer specification limit would remain same, thus to increase Cpk process variation must be reduced as much as possible.This means making process more robust and decreasing defects.Try to achieve 6sigma level3.4 DPMO

Control Chart for Pod WeightSample1234562.7922.812.7772.7992.8032.7882.7742.7832.7992.822.8122.8072.7972.792.7852.7952.8662.8262.8192.7872.8092.8622.8232.8162.7542.7932.822.8462.8232.8072.7842.7812.7332.8012.8232.8442.8442.7992.7812.8022.822.8132.8062.7862.8362.8152.8362.8082.8432.7662.7952.7782.8352.7832.8162.792.8232.8022.782.804Range0.090.0440.1030.0840.0860.025 CL2.805SD 0.025UCL 2.879LCL2.730ChartAnalysisAcc to chart,Performance variability is within the control limitHence process is in control

Reducing quality control expenseWastage of checked samplesReworking costs of tested samplesIncrease in defects due to reworking process

No reliance on operators individual judgementStandardized procedures and settings to be followed

Control Chart for Finger HeightSample1234562.0212.1582.0491.9592.1071.8751.8362.2562.0992.2692.1932.1932.0042.1661.9552.1251.9882.0092.1772.1712.0682.1431.9792.2782.1672.0322.0321.9552.0182.0072.0162.1082.1052.0371.9571.8811.9392.3022.0192.1542.1041.832.1792.1891.972.0672.0881.9031.9622.1281.9762.2282.0361.9492.261.991.8632.1832.021.889Range0.4240.3120.150.3140.2360.448CL2.060SD0.118UCL 2.415LCL1.705ChartFinger Height Control Chart Shift AMost of the data in shift A exceeds UCLFinger Height Control Chart Shift BMost of the data in shift B is within the control rangeFinger Height Control Chart Shift CMost of the data in shift C is below the LCLTypes of defectsAcc to auditor, excess reagent defectPrior to Greenlight 10%Post implementation of Greenlight 22%Leads to increased rejection rates of non-defective product as wellCustomer is unaware of the existence of such a defectHigher wastage and losses

DefectsType of defectOperator DefectsAuditor DefectsCumulative Defects% of Cumu. DefectExcess reagent4111521%Frame feed failure291115%Dirt from assembly0568%Insufficient reagent1457%Damaged spring3357%Double feed2357%Malformed box1357%Misalignment on assembly1357%Positive Sheet defect2346%Excess flash on box2246%Marginal lamination1246%Negative Sheet defect3234%Grand Total225072100%Analysis of DefectsInferencesImproper proportion of reagent is one of the frequently occurring defects

The top 5 defects constitute around 60% defect

Almost 60% defects is produced by the operator himself

FMEA: Risk Priority NumberShingos Transformational Process

ComparisonOperator can change the speed temperature, and pressure of the machineMaintenance has unique set of solution with each machineOperators produce consistent quality by tweaking based on experienceOperators have to shut down the machineMaintenance has to determine standardized best practice procedure for all the machineProcesses have been standardized OLD ProcessGreenlight ProcessChange ManagementChange managementapproach to transitioningindividuals,teams, andorganizationsto a desired future state with the help of new process/changesPolaroid Operator based statistical process control (SPC)Cut CostsSuperior product quality

ImplicationsHard to change the concept of work mapped in the head of the workersOperator thinks that shutting down the machine is more expensiveMaintenance thinks their job is depersonalizedQC personnel do not trust the operatorsEngineering technicians feels that they are better than the operatorsIncrease in defect rate from central process auditors (1% to 10%)

Developed by Edward Deming and derived from PDCA cycleQuality circles were first established inJapanin 1962 by Kaoru Ishikawa

31Quality CirclesQuality circle - PolaroidOLearyMurrayBud Rolfs

RecommendationsOperator responsibility:Need to treat downstream process as a customerShould aim at Doing it Right the first timeShould shut machine as soon as process is out-of-controlManagement responsibility:Convince operators about potential impact of their process on overall qualityIntroduce SPC training modules for OperatorsInculcate a culture of Quality in the Organization at all levels

RecommendationsProcess Control measures:Apply SPC at the injection moulding machines to control defectsAutomation of data collection methodsProcess Documentation & Product Quality Rating should be introducedSchedule maintenance periodically Use checklist Change Management:Motivate operators towards achieving Polaroids Quality objectiveLink incentives to operators salaries to ensure adherence to revised Quality norms

Thank You!!!Who did whatJaspreet Intro+ Control ChartsAnish Change management & quality circleNrupen Process Quality and capability, Assembly chartPrakhar Fema, Transformational process, RecommendationsUttam Fema, Deftects analysis, Putting together presentation, Recommendations