ME 134 2SAY2324 L23 Variation and Quality.pdf

VARIATION AND QUALITY  3 INTRODUCTION Variation and quality are essential part of manufacturing. Quality: Conformity to requirements or specifications. The ability of a product or service to consistently meet customer needs....

VARIATION AND QUALITY  3 INTRODUCTION Variation and quality are essential part of manufacturing. Quality: Conformity to requirements or specifications. The ability of a product or service to consistently meet customer needs. Variation: refers to the change in outcome of a process Tolerance: allowable or permissible limit of variation of a process.  4 MEASURE OF QUALITY LEGO LEGO GILETTE RAZORS RAZORS Special Bricks Classic Bricks  5 CAR BODY BUILD VARIATION: PRODUCT LAUNCH 6 standard deviations from the mean: 3.4 defects per million!  6 CAR BODY ASSEMBLY HIERARCHY  7 WHAT DO WE NEED TO KNOW? Customers wants. Their perceived “good quality” should be related to the specifications of the product. Quantify the variation statistically. Factors that cause variations in processes and how to minimize the variations as needed. Monitor the variation and maintain process control.  8 VARIATION AND QUALITY Normal distribution Error stackup and simple fits Lognormal distribution Process sensitivity Principle of measurement  NORMAL DISTRIBUTION NORMAL DISTRIBUTION Helps us quantify the outcome of a process. Helps in understanding the measurement of variation from process. MEASURED VARIATION: HEX NUTS Nuts are not identical even if they should have the same nominal sizing. Pocess variation which is a natural outcomeof a processs. Measurement procesdure might have some variations too.  12 NORMAL DISTRIBUTION Bell curve Probability distribution of the occurrence of a data. Naturally occurring phenomena tends to approximate the normal distribution. In manufacturing, normal distribution is a means of quality control.  13 NORMAL DISTRIBUTION  14 NORMAL PROBABILITY DENSITY FUNCTIONS http://work.thaslwanter.at/Stats/html/statsDistributions.html  15 CUMULATIVE DISTRIBUTION FUNCTION (CDF) 16 Presentation title 17  18 EXAMPLE Car tires have a lifetime that can be modeled using a normal distribution with a mean of 80,000 km and a standard deviation of 4,000 km. What fraction of tires can be expected to wear out within ±4,000 km of the average? What fraction of tires will wear out between 70,000 km and 90,000 km?  19 ANSWER TO QUESTION 1  20 ANSWER TO QUESTION 2  21 STACKUP AND SIMPLE FITS MEASURED VARIATION: HEX NUTS  23 MODELING ‘STACKUP’  24 ASSEMBLY EXAMPLES  26 LOGNORMAL DISTRIBUTION  LOGNORMAL DISTRIBUTION Logarithm of x is distributed normally Image: By Krishnavedala|Wikimedia Commons  28 LOGNORMAL DISTRIBUTION: METAL POWDER FOR 3D PRINTING Ti6Al4V Specification: 15-45 um  29 PROCESS SENSITIVITY HOW HEX NUTS ARE MADE FACTORS THAT AFFECT PROCESS VARIATIONS Process: inherent capability; change of settings Material: raw material variation, defects Equipment: tool wear, equipment maintenance or calibration Operator: procedure, distraction, fatigue Environment: temperature, humidity, vibration Measurement: capability of the tool (resolution), need for calibration 33  34 MACHINE, TOOL AND WORKPIECE  35 INJECTION MOLDING PROCESS WINDOW  37 Systematic (“special cause”) variation: influences of process parameters or external disturbances that can be isolated and possibly predicted or removed. Random (“common cause”) variation: caused by uncontrollable factors that result in a steady but random distribution of output around the average of the data. In other words, this is the ‘noise’ of the system.  38 GENERAL MODEL OF PROCESS VARIATION  40 EXAMPLE SENSITIVITIES Injection Molding Relationship between molecular weight of polymer (determines viscosity) and accuracy (final part dimension compared to mold) Relationship between injection pressure and accuracy  41 EXAMPLE SENSITIVITIES Machining Relationship between depth of cut and surface roughness (= spatial frequency of tool marks) Relationship between tool life (sharpness) and accuracy (= workpiece deformation via higher force and temperature rise) PRINCIPLES OF MEASUREMENT  44 Accuracy = “the ability to tell the truth” Difference between the measured and true value Repeatability = “the ability to tell the same story many times” Difference between consecutive measurements intended to be identical Resolution = “the ability to tell the difference” Minimum increment that can be measured