Process Improvement: Quality Assurance, Control, and Management Miles Hamby, Ph.D. Copyright©2011 Miles M. Hamby

Quality Definitions Quality ~ conformance to requirements and “fitness for use” (Juran) Quality Management ~ “ensuring the process will consistently produce the desired product.” (Hamby) Philosophy ~ Gold plating is bad; prevention of over inspection is good

Quality Characteristics Grade vs. Quality Grade: meets spec requirements Quality: behaves as expected Prevention vs. Inspection Management Responsibility Processes Customer Satisfaction Keep customer informed Stick to requirements

Quality Processes Quality Planning Quality standards and methods to meet them Planning Quality Assurance Improvements, audits, measurement comparisons, considering standards appropriateness Executing Quality Control Measuring/testing errors, measuring schedule performance, comparing results to standard Controlling

Quality Planning Tools Standards Benchmarking (past analysis) Benefit/Cost Analysis (BCI) Flowchart (future analysis) Design of Experiments (what if?) Fishbone Diagram (also used in QC) Cost of Quality (costs of conformance & non-conformance)

Cost of Quality Cost of Conformance Prevention Appraisal Planning Training Auditing Controlling Cost of Nonconformance Failure Costs Internal (pre customer) External (post customer) Scrap Rework Expediting Warranty Service Recalls

Quality Assurance Tools Evaluation against standards on regular basis Re-evaluation of standards, methods, and procedures Quality Audit: structured review of quality activities that identifies lessons learned

Quality Control Tools Inspection Pareto Diagram Fishbone Diagram Checklists Statistical Sampling Control Charts Flow charting (also used in Quality Planning) Trend Analysis

Quality Techniques Continuous Process Improvement (Kaizen) Just in Time (JIT) Total Quality Management (TQM) (Juran)

Key Quality Gurus Walter Shewhart Edward Demming Joseph Juran Phillip Crosby Genichi Taguchi Kaoru Ishikawa

“Grandfather of Quality Management” Walter Shewhart (1891 – 1968) “Grandfather of Quality Management” “The object of industry is … to reduce everything possible to routines requiring a minimum amount of human effort.” Shewhart Cycle – PDCA “Deviations in the results of a routine process outside [statistical] limits indicate that the routine has broken down and will no longer be economical until the cause of trouble is removed.”

Shewhart Control Charts Product Variation – due to “assignable” causes and “chance” causes Objective – reduce variation from assignable causes Technique – use Statistical charts to measure variation UCL = Accepted value + k*process standard deviation LCL = Accepted value - k*process standard deviation

Edward Demming (1900 – 1993) 14 POINTS 1."Create constancy of purpose towards improvement". Replace short-term reaction with long-term planning. 2."Adopt the new philosophy". The implication is that management should actually adopt his philosophy, rather than merely expect the workforce to do so. 3."Cease dependence on inspection". If variation is reduced, there is no need to inspect manufactured items for defects, because there won't be any. 4."Move towards a single supplier for any one item." Multiple suppliers mean variation between feedstocks. 5."Improve constantly and forever". Constantly strive to reduce variation. 6."Institute training on the job". If people are inadequately trained, they will not all work the same way, and this will introduce variation. 7."Institute leadership". Deming makes a distinction between leadership and mere supervision. The latter is quota- and target-based.

Demming’S 14 Points (cont) 8."Drive out fear". Demming sees management by fear as counter- productive in the long term, because it prevents workers from acting in the organization's best interests. 9."Break down barriers between departments". Another idea central to TQM is the concept of the 'internal customer', that each department serves not the management, but the other departments that use its outputs. 10."Eliminate slogans". Another central TQM idea is that it's not people who make most mistakes - it's the process they are working within. Harassing the workforce without improving the processes they use is counter- productive. 11."Eliminate management by objectives". Demming saw production targets as encouraging the delivery of poor-quality goods. 12."Remove barriers to pride of workmanship". Many of the other problems outlined reduce worker satisfaction. 13."Institute education and self-improvement". 14."The transformation is everyone's job".

If answer to any is “no”, then Management controllable Joseph Juran (1904 – 200) First to think of the “cost of poor quality” “TQM” ~ Total Quality Management Juran’s Quality Process Measure your product Compare it to a standard Act on the difference Who has control over errors ~ 80% controllable by management; only 20% by operator Does the operator know what he is supposed to be doing? Does the operator know what he is doing? Does the operator have the tools to do what he is doing? If answer to any is “no”, then Management controllable

Phillip Crosby (1926 – 2001) “Zero Defects” Originated with Martin-Marietta, adopted by Dept of Defense in 60s Quality is conformance to requirements Defect prevention is preferable to quality inspection and correction “Zero” Defects is the quality standard Quality is measured in monetary terms – the Price of Nonconformance (PONC)

Kaoru Ishikawa (1915 – 1989) Cause and Effect - loss of quality in a product is and effect caused by something in the process Fishbone diagram - Product is the “head” of the fish; “bones” are the various process inputs; “sub-branches” within each bone are “causes” of errors in the process. Introduced Quality Circle concept to Japanese manufacturing

Genichi Taguchi (1924 - ) Quality Loss function Quality loss all the way through to the customer, including cost of scrap, rework, downtime, warranty claims and ultimately reduced market share. Costs of quality varies with product deviation from the mean. Loss (L) is a function of the variance (σ) from the target (m). L = k (x – t)2 Where: L = Loss in Dollars x = Quality Characteristic (diameter, no. of errors, etc) t = Target Value for x k = Constant

Optimal Quality and Responsibility Marginal Analysis Optimal quality is reached at point where incremental value from improvement = incremental cost to secure it. Responsibility for Quality Senior management is responsible for organizational quality PM has ultimate responsibility for quality of product of project Each team member is responsible for self inspection

Statistical Terms Mean ~ Arithmetic average obtained by dividing the sum of data by the number of data in the set Median ~ Middle value of a set of data in rank order Mode ~ most frequently occurring value (possible to have more than one mode) Standard Deviation (Sigma) ~ square root of the average sum of the squared deviations from the mean ~ indicates probability of occurrence in a distribution Six-Sigma ~ plus and minus 3 standards deviation from the mean of a distribution

Estimating the Standard Deviation PERT estimate (Project Evaluation and Review Technique) 1Sigma (σ)  range between pessimistic estimate (P) and optimistic estimate (O) divided by 6 _ X -1 σ 1σ 2 σ 3 σ -3 σ -2 σ  P O

Sigma - level Number of “Sigma” represents level of quality desired, ie., number of errors acceptable EG ~ 6 sigma (± 3σ) translates to 27 errors out of 10,000 units (ie, 99.73% probability error-free) 12 sigma (± 6σ) translates to 1 error out of 10,000 units (ie, 99.99% probability error-free) _ X -1 σ 1σ 2 σ 3 σ -3 σ -2 σ  -6σ 6σ

Statistical Sampling Quality Control Chart Graphic display of results, over time, of a process… used to determine if the process is “in control.” To create a control chart: Samples are taken Variables are measured Attributes are found and plotted on chart

Monitoring Project Results Variable - anything measured Attribute - binary value, ie, either right or wrong Probability - likelihood event will occur, usually expressed as %

Control Chart Features Control Limits Acceptable range of variation of a process often shown as 2 dashed lines on chart Upper and Lower Control Limits are determined by organization’s sigma quality standard Specification Limits Contractual requirements for performance and quality Not calculated based on control chart Outside chart control limits if project can meet Inside chart control limits if project cannot meet

Statistical Process Control Out of Statistical Process Control (“Out of control”) Errors exceed UCL or LCL Indicates lack of consistency or predictability in process

Rule of Seven Heuristic - non-random data points grouped together in a series that total 7 on one side of mean Assignable Cause - data point or Rule of Seven that requires investigation to determine cause of variation.

Process Improvement: Quality Assurance, Control, and Management Miles Hamby, Ph.D. Copyright©2011 Miles M. Hamby