1-1._PENGANTAR_PENGENDALIAN_MUTU_PU.ppt

Definition of QualityQuality Control AreasStatistical Process ControlIntroduction to Quality Control

Handbook :Fundamental of quality control and improvementby : Amita Mitra

Evolution of Quality Control Egyptians demonstrated quality through construction of PyramidThe greeks set high standard in arts and craftsStructural design created by Roman for building cities, churches, bridges, roads and so on.In 1800s, the production of good and services was predominantly to a single individual or a small group owned by person or a small group

Evolution of Quality Control1900s to 1920s : - mass production - principle of specialization - process is relatively simple1920 -1940: - Inspection quality control - process is more complicated - Statistical aspects dealing with quality control were being developed1940 1960: - Statistical Quality Control - Standard sampling plan is introduced1960: - TQM era (involvement of several department in quality Control activities1970s: - TQM Organizationwide ( partisipation of every one in the company ) - Fishbone diagram is introduced - Taguchi Method is introduced

Evolution of Quality Control1980s: - The continued growth in computer 1990s: - expanded use of quality measures and increased to customer needs

SIX SIGMA (DMAIC)LEAN-SIGMA

Purposes1.Ability to understand about Quality2. Ability to recognize the quality control areas3. Ability to diffrentiate between the assinable and common causes4. Ability to calculate the control limit5. Ability to calculate the type 1 and type 2 error6. Ability to analyse the control chart

Definition of QualityQualityTranscendentProduct-basedManufacturing-basedUser-basedValue-basedGarvin (1984)Performance, features, reliability, conformance,durability, serviceability, aecthetics, and perceived qualityCrosby (1979)Conformance to requirements or specificationsJuran (1974)Fitness for use

Definition of qualityThe quality of a product or service is the fitness of that product or service for meeting its intended use as required by the customers

TermsQuality Characteristics : One or more elements which define the intended quality level of a product or serviceNonconformity : a quality characteristic that does not meet its stipulated specifications requirement.A nonconforming unit : a unit that has one or more nonconformities such that the unit unable to meet the intended standard and is unable to function as required.

Defect : a quality characteristic that does not meet certain standard.A Standard : a pricise statement that formalize the requirements of the customer .Specification : a set of conditions and requirements of specific and limitted application , that provide a detail description of the procedure, process, material, product, or service for use primarily in procurement and manufacturing.

Quality AspecsQualityQuality of DesignQuality of ConformanceQuality of Performance

Quality of DesignThe product or service must be designed to meet at least minimally the needs of the customerInfluenced by : product, cost, profit policy, demand, availability of parts and materials, and product safety.

Quality of ConformanceThe manufactured product or the service rendered must meet the standards selected in the design phase.

Defect prevention, defect finding, and defect analysis and rectification.

Quality of PerformanceConcerned with the operation of the product when actually put to use or the service when performed and measures the degree to which it satifies the consumer

Quality ControlDefined as a system that is used to maintain a desired level of quality in a product or service

Quality Control AreasOff-line Quality ControlStatistical Process ControlAcceptance Sampling Plan

Choosing the product and Process parametersThis task can be performed through the product and process design.The design experiment and Taguchi method gives the off-line quality control procedure.Off-Line Quality Control

Comparing the process and service output to the standard.On-line SPC Taking the information about the product, process, and service when the system is functioning.SPC uses the control chart for on-line SPC procedure.Statistical Process Control

Related to product or service inspection. Acceptance sampling plan is a planning in determining the number of item in the sample and lot criteria for lot acceptance based on a specific condition, e.g., The producen or customer risk. Acceptance Sampling Plan

A system that ensures that all procedures that have been designed and planned are followedAll those planned or systematic actions necessary to provide confidence that a product or service will satisfy given needsQuality Assurance

Quality CircleAn informal group of people that consists of operators, supervisors, managers, and so on, who get together to improve ways to make the product or deliver the services.

Benefits of Quality ControlThe improvement in the quality of product and servicesThe system is continually evaluated and modified to meet the changing needs of customer A quality Control system improves the productivityReduce cost in the long run The leadtime for producing parts and subassemblies is reducedA quality control system maintains an improvement environment when everyone strives for improved quality and productivity.

Chapter 6Statistical Process ControlUsing Control Chart

Control Chart is a graphical tool for monitoring the activity of an ongoing process Sometimes referred to as Shewhart Control ChartThe value of quality characteristic are plotted along the vertical axis, and horizontal axis represents the samples, or subgroups.

Control ChartQuality CharacteristicsSubgroup01234567151812UCLLCLCL

The chart indicate :When to take the corrective actionsType of remedial action necessaryWhen to leave the process aloneProcess CapabilityPossible means of quality improvement.

Causes Of Variation

The existing variability is not from the process. It is not effecting all the products.Example: The use of the wrong tool, bad quality of raw material, or operator error.If one observation locates outside the control limits or there is a non-random pattern, it is expected that there is an assignable cause process is out of control.Assignable/Special Cause

The existing variability comes from the process it self.It always exists as long as the process does not change and effected all the products. It comes from alot of small causes and can not totally eliminate. Example: variability that come from the raw material, mechanical vibration, work condition fluctuation..Common Cause

Control Limits Calculation

The mean of part thickness of a product is 30 mm with standard deviation equal to 1.5 mm. Make a control chart using the limits of 3s with sample size equal to 5 chosen ramdomly from the process.Example 1

To decide that a process is out of control when it is not.

Probability of type I error is represented by a.Error in Decision Making(1)Type I error

Error in Decision Making(2)a/2a/2CLUCLLCLType 1 error

To decide that a process is in of control when it is not.

Probability of type II error is represented by b.Type II ErrorError in Decision Making(3)

Kesalahan Tipe IIKesalahan Pengambilan Kesimpulan (4)CLUCLLCLABbm

Error in Decision Making

The control chart for nylon strength will be made. A sample with size 5 is taken randomly from a process. The mean strength and standard deviation are estimated by 120 kg and 8 kg, respectively.Example 6-2. p.242

If the control limits uses the 3 sigma from the mean of process, what is the probability of type 1 error ? If the process mean shifts to 125 kg, What is the probability of type 2 error at the first sample plotted since the shifting existed ? What is the probability to detect the shift in the first two samples chosen randomly and independentlyExample 6-2 (continuation)

The OC curve shows the control chart ability to detect the changing of process parameter.

Plotting the probability of type 2 error to the shift of the process parameter value when process incontrol.

Operating Characteristic Curve0.01.00.5140130125145Process ParametersProbability of undetected

The effects of Control Limits to Type I and II ErrorCLUCLLCLLCLUCL

The Effect of Sample Size to The Control LimitsCLUCLLCLLCLUCLbb

To determining the average of the number of Sample needed to detect the out-of-control signal

ARL = 1/PdPd: probability of plotted observation otuside the control limits = alphaAverage Run Length (ARL)

Choosing Sample item: instant-of-time method and period-of-time method.Sample size.Frequency of Sampling.Sampling

If one or more located outside the control limmits out of control.Control Chart Pattern Analysis (1)Quality characteisticsSubgroup01234567151812UCLLCLCL

If two out three consecutive points located outside 2s (warning limit) at the same side of the CL out of control.Control Chart Pattern Analysis(2)Quality characteristicsSubgroup01234567151812UCLLCLCLUWLLWL

If four out five points consecutively located outside 1s at the same side of CL out of control.Control Chart Pattern Analysis(3)Quality characteristicsSubgroup01234567151812UCLLCLCLlimit 1slimit 1s8

If nine or more of consecutive points located at the same side of CL out of control.Control Chart Pattern Analysis(4)Quality characteristicsSubgroup01234567151812UCLLCLCL89

If six or more points consecutively has a tendency to increase or decrease out of control.Control Chart Pattern Analysis(5)Quality characteristicsSubgroup01234567151812UCLLCLCL89

Application of the rules to determine a simultanly out of controlThe application of the simultanly out of control can increase the probability type I error. a: The total type I error k: the number of rules usedai: type I error for the rule number i.

Assumptions : the rules are independently exist.

The pin diameter produced by a machine is the quality characteristic that to be controlled . Based on the historical data the mean diameter of diameter is 15 mm with standard deviation equal to 0,8 mm. If the sample, with size 4, is taken randomly from the process :

Example 1

Determine the control limits for 1s, 2s, dan 3s for mean diameter of the pin.Determine the probability of false alarm?If the mean of the process shift to 14,5 mm, what is the probability that the shift cant be detected in the first sample plotted since the shift exist ?What is the probability failing to detect the shift for the second sample since the shift exist ?Make the OC curve for the control chart.

A quality engineering is asked to make a control chart to control a process with mean and standard deviation equal to 100 and 5 respectively. The control should use the control limit 3 sigma and has the ability to detect a shift from 100 to 105 with probability equal to 0,75. ( the rule number one is used to determine the out of control condition).Example 2

Determine CL, LCL, dan UCL for the control chart? If the control chart application used in (a) uses the rule number one and two simultaneously for the out-of-control condition, how is the effect in the capabiltiy of the control chart in detecting the shift the average from 100 to 105