Quality and Performance - Chapter 3

Chapter 3: Quality and Performance

• This chapter covers definitions related to quality, methods for calculating Upper Limit (UL) and Lower Limit (LL) for both variables and attributes and determining if a process meets customer requirements using Cp and Cpk.

Quality

• Quality is how customers describe their satisfaction with a service or product.
• A defect is any instance where a process fails to satisfy its customer.
• It's important to consider quality from both internal and external customer perspectives.

Quality in Manufacturing Environment

• Various steps are involved in ensuring quality in manufacturing, from material storage to rework, including material kitting, PCB coding, SMT, AOI, X-ray inspection, functional testing, in-circuit testing, mechanical assembly, wave soldering, and manual insertion.
• 100% inspection with no sampling is possible with quality SMT and AOI equipment.

Costs of Quality

• Aiming for too high quality might be cost prohibitive with experts agreeing that cost of quality range 20 to 30% of sales.
• There are four major categories of costs associated with quality: prevention, appraisal, internal failure, and external failure.

1. Prevention Costs

• Costs are associated with preventing defects before they happen, including redesigning processes or products to make them simpler to produce.
• Training employees in continuous improvement and working with suppliers is a part of prevention.
• Investment of time, effort and money is required for prevention.

2. Appraisal Costs

• Incurred when the firm assesses the performance level of its processes, appraisal cost decreases.
• Appraisal include: receiving inspection.

3. Internal Failure Costs

• Costs resulting from defects are discovered during the production of a service or product.
• Two types of internal failure costs: rework (service must be performed again/corrected) and scrap (item is unfit for further processing).

4. External Failure Costs

• Costs arise when a defect is discovered after the customer receives the service or product.
• Dissatisfied customers may talk about bad service to their friends with consumer protection group alerting media.
• External failure costs impact future profits and include warranty service and litigation.
• Warranty costs must be considered in the design of new services or products.

5. Ethical Failure Costs

• Encompass societal and monetary costs when defective services/products are deceptively passed to customers, jeopardizing stakeholders' well-being.
• Deceptive business practice has three elements: intentional exploitation of the customer, concealed truth, and disproportionate economic benefit at the customer's expense.
• Deception undermines employee competence, demean their relationships with external customers.
• Ethical behavior is essential for all employees within an organization.

Examples of Ethical Failure Costs

• Enron Scandal (2001): Shareholders lost $74 billion, with thousands of employees and investors losing retirement accounts.
• WorldCom Scandal (2002): Inflated assets by $11 billion, resulting in 30,000 job losses and $180 billion in losses for investors.
• Lehman Brothers Scandal (2008): Hid over $50 billion in loans disguised as sales.
• Bernie Madoff Scandal (2008): Tricked investors out of $64.8 billion through a Ponzi scheme.

Total Quality Management and Six Sigma

• Total Quality Management (TQM): A philosophy that stresses customer satisfaction, employee involvement, and continuous improvement.
• Six Sigma: A flexible system for maximizing business success by minimizing defects and variability in processes.

Total Quality Management – 3 Principles

• Customer Satisfaction: Expectations of customers have been met or exceeded- internal or external.

  • Conformance to Specifications: It is measuring processes really being judged, on-time consistency and speed.
  • Value: How well the service or product serves its intended purpose at a price customers are willing to pay.
  • Fitness for Use: Convenience of service or mechanical features/appearance.
  • Support: Service/product support matters and employee affects the customers' perception.

• Employee Involvement:

  • Challenge is to define customer for each employee.
  • Quality at the source is essential; quality can't be inspected into a product, it must be built into it.
  • Employee involvement is a key tactic with small groups of people with a common purpose.
  • Problem-solving teams (quality circles), special-purpose or self-managing teams.

• Continuous Improvement: Seek to improve processes - kaizen.

  • Getting Started: SPC training, build work teams and employee involvement, utilize problem-solving tools, develop operator ownership.
    • Problem-solving using The Deming Wheel (PDCA):
      • Plan-Improvement, document, analyze, set goals, discuss benefits and costs.
      • Do-Implement plan, monitor improvements.
      • Study-Analyze data.
      • Act-Document improved process as standard procedure.

Six Sigma

• The goal is achieving low rates of defective output; the mean output for a performance measure is +/- six standard deviations (sigma) from the limits of the design specifications for the service or product.
• It is driven by close understanding of customer needs, the diligent use of facts, data, statistical analysis.
• Variability causes customer dissatisfaction.

Acceptance Sampling

• Acceptance sampling is a statistical technique used to determine whether to accept or reject a material based on inspecting a sample.
  • Acceptable Quality Level (AQL) is a statement of the proportion of defective items that will be accepted.
• In a supply chain, a company can be both a producer and a consumer, with buyer's specifications becoming the targets a supplier shoots for.
• The buyer's sampling plan will provide a high probability of accepting AQL or better.
• Example: A random sample taken from a large quantity of items and tested measured relative to specification.
  • If the sample passes the test, then the entire quantity of items is accepted.
  • If a sample fails the test, either the entire quantity of items is subjected to 100% inspection, all defective items repaired.
• Firm A (as manufacturer) uses Total Quality Management (TQM) and Six Sigma to achieve internal process performance to meets buyer specs via motor sampling.
• Supplier manufactures fan blades for the target of firm A spec via blade samplings.

Statistical Process Control (SPC)

• A method using statistical techniques to determine if a process delivers what the customer wants and is used to detect defects.
• Control charts are used to detect defects, can detect a decrease in complaints, increase in defective gear boxes, increase in time to process applications, decline in scrapped units, and increase in late payments.
• Manager should identify root causes to take corrective actions.

Variation of Outputs

• No two services/products are exactly alike, and variations in output must be minimized
• Performance Measurements are evaluated in 2 ways
  • Variables: characteristics measured on continuous scale, knowing how much corrections are indicated
  • Attributes: characteristic counted in discrete units (yes-no, integer number)
• Sampling can be done via complete inspection or sampling plans.

Key Concepts, Sample Average and Distribution

• A sample mean calculates a number of observations/total observations
• Standard deviation is the average measure of distribution within the data
• The most important property of all of the distributions.

Statistical Process Control (SPC)

• Categories of Variation
  • Common cause from unidentifiable sources of variation.
  • Assignable cause from variation-causing factors that can be identified and eliminated and detected by SPC.
• Control Chart: Time-ordered diagram to determine observed variations and has a value, or center line.
• In using a control chart, take a random sample, and calculate a performance measure to eliminate cause and improve performance, repeat the procedure,

Control Charts

• The control chart has various components including using upper control limit (UCL) and Lower Control Limit (LCL) where assignable causes are found in samples.

• Three ways in which a process can be out of control:

  • Assignable causes

• Using upper control limit (UCL) and Lower Control Limit (LCL)

• Sudden changed or Monitor action

• Type I eror is pure randomness where fact it was due.

• Type II error process is controlled and randomness is present when actually out of control.

Types of Control Charts

• R Chart measures process variability / generates output and target value.Attribute control is where a process that is being or products will be generated by
• p Charts measure proportion of effective services and by process, one defect in service
• c chart measures number of defects can be in place or with in the process

Control Charts for Variables

• R-Chart : σ NOT known

  • UCLR=DAR
  • LCLR = D3R
  • R = average of several past R values and the central line of the control chart
  • D3, D4 = constants that provide three standard deviation (three-sigma) limits for the given sample size

• X-Chart : σ NOT known

  • UCL = + A₂
  • LCL = - A₂

Steps to Compute Control Charts:

1. Collect data.
2. Compute the range.
3. Use Table 3.1 to determine R-chart control limits.
4. Plot the sample ranges. If all are in control, proceed to step 5. Otherwise, find the assignable causes, correct them, and return to step 1.
5. Calculate for each sample and determine the central line of the chart,
6. Use Table 3.1 to determine control limits Plot the sample means. If all are in control, the process is in statistical control. Continue to take samples and monitor the process. If any are out of control, find the assignable causes, correct them, and return to step 1.

Control Charts for Attributes

• UCL = + zo, where sigma = the center line on the chart p, and for The standard deviation is O=√p(1-p)/n, not known

Process Capability

• Process Capability is the ability of the process to meet the design specification (defined by customer requirement or market forces).
  • Nominal Value: A target for design spec.
  • Tolerance: An allowance above or below the nominal value.
• There are multiple process capabilty ratios that determine
  • Whether lower requirements of the specification come from
  • Whether there minimum sigma in extreme value -Process Capability Index (Cpk): An index that measures the potential for a process to generate defective outputs relative to either upper or lower specifications. : σ = standard deviation of the process distribution