- DMAIC
- The five-phase Six Sigma improvement cycle for existing processes: Define, Measure, Analyze, Improve, and Control.
- Six Sigma
- A data-driven methodology aimed at reducing variation and defects to achieve 3.4 defects per million opportunities (a 6σ process level).
- DPMO
- Defects Per Million Opportunities = (defects / (units × opportunities)) × 1,000,000.
- Cp
- Process capability (potential): Cp = (USL − LSL) / (6σ); it ignores centering and assumes the process is on target.
- Cpk
- Process capability index accounting for centering: Cpk = min[(USL − x̄)/(3σ), (x̄ − LSL)/(3σ)].
- FMEA
- Failure Mode and Effects Analysis: a structured tool to identify potential failures and prioritize them by Risk Priority Number (RPN = Severity × Occurrence × Detection).
- Pareto chart
- A bar chart ordering causes from most to least frequent, illustrating the 80/20 rule that ~80% of effects come from ~20% of causes.
- Control chart
- A time-ordered plot of a process statistic with a center line and control limits used to distinguish common-cause from special-cause variation.
- DOE
- Design of Experiments: a structured method of varying input factors simultaneously to identify their effects and interactions on a response.
- ANOVA
- Analysis of Variance: a hypothesis test comparing the means of three or more groups by partitioning total variation into between-group and within-group components.
- Gauge R&R
- A measurement system study quantifying repeatability (equipment variation) and reproducibility (appraiser variation) as a percentage of total variation.
- Voice of the Customer (VOC)
- The process of capturing customers' stated and latent needs, expectations, and preferences to drive design and improvement.
- CTQ
- Critical to Quality: a measurable product or process characteristic whose performance standard must be met to satisfy the customer.
- Six Sigma roadmap
- An organization's structured plan defining the vision, infrastructure, training, and project pipeline for deploying Six Sigma enterprise-wide.
- Champion
- A senior leader who sponsors Six Sigma projects, removes barriers, secures resources, and links projects to business strategy.
- Master Black Belt (MBB)
- An expert who trains and mentors Black Belts, leads deployment strategy, and provides technical and statistical guidance across projects.
- Black Belt
- A full-time Six Sigma leader who manages complex cross-functional improvement projects and coaches Green Belts.
- Green Belt
- A part-time practitioner who leads smaller projects and supports Black Belt projects while performing a regular job.
- Yellow Belt
- A team member with basic Six Sigma awareness who participates on project teams and supports data collection.
- Executive leadership role
- Top management sets the Six Sigma vision, allocates resources, and establishes accountability and alignment with strategic goals.
- Hoshin Kanri
- Strategic policy deployment that cascades a few breakthrough objectives from top leadership down through the organization with catchball alignment.
- Balanced scorecard
- A strategic management tool tracking performance across four perspectives: financial, customer, internal process, and learning/growth.
- SWOT analysis
- A planning tool that evaluates internal Strengths and Weaknesses and external Opportunities and Threats.
- Strategic project selection
- Choosing projects aligned with organizational goals, customer impact, and financial return rather than convenience.
- Project prioritization matrix
- A tool that ranks candidate projects against weighted criteria such as benefit, effort, risk, and strategic fit.
- Critical to satisfaction (CTS)
- Top-level customer needs that branch into critical-to-quality, cost, delivery, and safety requirements.
- Critical success factors
- The few key areas where satisfactory results are essential for the organization to achieve its mission.
- Key performance indicators (KPIs)
- Quantifiable measures used to evaluate progress toward strategic and operational objectives.
- Organizational drivers
- Factors such as customers, employees, suppliers, and shareholders that shape an organization's goals and metrics.
- Benchmarking
- Comparing processes and performance metrics to industry best practices or best-in-class organizations to identify improvement opportunities.
- Process benchmarking
- Comparing specific work processes against superior performers to adopt better methods.
- Voice of the business (VOB)
- The needs and requirements of the business, including profitability, growth, and shareholder value.
- Voice of the process (VOP)
- What the process is actually delivering, expressed through its statistical performance and capability.
- Change management
- A structured approach to transitioning individuals, teams, and organizations to a desired future state while minimizing resistance.
- Kotter's 8-step change model
- A change framework: create urgency, build a coalition, form a vision, communicate it, empower action, generate short-term wins, consolidate gains, and anchor change.
- ROI of Six Sigma
- The financial return from improvement projects, typically measured as hard savings, cost avoidance, and revenue gains versus deployment cost.
- Deployment infrastructure
- The roles, training, governance, and tracking systems established to sustain a Six Sigma initiative.
- Project tracking system
- A repository that monitors project status, financial validation, and benefits across the deployment portfolio.
- Stakeholder analysis
- Identifying parties affected by a project and assessing their interest, influence, and likely support or resistance.
- Process management
- The practice of defining, measuring, controlling, and improving processes to consistently meet customer requirements.
- Process owner
- The person accountable for a process's performance, documentation, and ongoing improvement.
- SIPOC
- A high-level process map listing Suppliers, Inputs, Process, Outputs, and Customers to scope a project.
- Process map
- A visual diagram of the sequence of steps, decisions, and flows within a process.
- Cross-functional process
- A process whose steps span multiple departments or functions within the organization.
- Business process metrics
- Measures such as cycle time, cost, quality, and customer satisfaction used to gauge process health.
- Cost of Poor Quality (COPQ)
- The costs incurred from defects: internal failure, external failure, appraisal, and prevention costs.
- Cost of Quality (COQ)
- Total cost of achieving quality, comprising prevention, appraisal, internal failure, and external failure costs.
- Prevention costs
- Costs of activities that prevent defects, such as training, planning, and process design.
- Appraisal costs
- Costs of inspecting and testing to detect defects, such as audits and measurement.
- Internal failure costs
- Costs from defects found before delivery, such as scrap and rework.
- External failure costs
- Costs from defects found after delivery, such as warranty claims, returns, and lost goodwill.
- First Pass Yield (FPY)
- The proportion of units completing a step correctly the first time without rework: FPY = good units / total units in.
- Rolled Throughput Yield (RTY)
- The probability a unit passes all process steps defect-free: RTY = product of each step's first-pass yield.
- Normalized yield
- The geometric average yield per process step: normalized yield = RTY raised to the power 1/(number of steps).
- Throughput
- The rate at which a process produces completed units over a period of time.
- Cycle time
- The total time from the start to the completion of a process or task, including wait time.
- Lead time
- The total elapsed time from a customer order or request to delivery of the product or service.
- Little's Law
- Work in process = throughput rate × lead time; relates inventory, flow rate, and time in a stable system.
- Process performance index
- A summary measure of how well a process meets requirements over the long term, such as Pp and Ppk.
- Sigma level
- The number of standard deviations between the process mean and the nearest specification limit; higher sigma means fewer defects.
- 1.5 sigma shift
- A long-term mean drift assumption added to short-term capability, so a 6σ process yields 3.4 DPMO rather than ~2 ppb.
- Defect
- Any instance where a product or service fails to meet a specification or customer requirement.
- Defective
- A unit that contains one or more defects, regardless of how many.
- Opportunity
- Any characteristic of a unit that could be measured and could be defective; used in DPMO calculations.
- Yield
- The percentage of units that meet requirements out of the total produced.
- Process capability vs performance
- Capability (Cp/Cpk) uses short-term within-subgroup variation; performance (Pp/Ppk) uses long-term overall variation.
- Tuckman's stages of team development
- Teams progress through Forming, Storming, Norming, Performing, and Adjourning.
- Forming stage
- The initial team stage marked by orientation, politeness, and dependence on the leader for direction.
- Storming stage
- The team stage characterized by conflict, competition, and challenges to authority and roles.
- Norming stage
- The team stage where members resolve differences, establish norms, and build cohesion.
- Performing stage
- The mature team stage where members work interdependently and productively toward goals.
- Adjourning stage
- The final team stage involving completion, disengagement, and recognition of accomplishments.
- Team roles
- Defined responsibilities such as sponsor, leader, facilitator, timekeeper, scribe, and member.
- RACI matrix
- A responsibility chart identifying who is Responsible, Accountable, Consulted, and Informed for each task.
- Team facilitation
- Guiding group interaction so meetings stay focused, inclusive, and productive without dominating decisions.
- Nominal Group Technique (NGT)
- A structured method where members silently generate ideas, share them round-robin, then rank or vote to prioritize.
- Multivoting
- A technique to narrow a large list of options through successive rounds of voting.
- Brainstorming
- A group method for rapidly generating many ideas while deferring judgment.
- Affinity diagram
- A tool that organizes large numbers of ideas into natural groupings based on relationships.
- Conflict resolution
- Methods for managing disagreements, including collaborating, compromising, accommodating, competing, and avoiding.
- Thomas-Kilmann conflict modes
- Five conflict-handling styles based on assertiveness and cooperativeness: competing, collaborating, compromising, avoiding, and accommodating.
- Consensus
- A decision that all team members can support and commit to, even if it is not everyone's first choice.
- Groupthink
- A dysfunction where the desire for harmony suppresses dissent and critical evaluation of alternatives.
- Coaching
- Helping team members develop skills and performance through guidance, feedback, and questioning.
- Mentoring
- A developmental relationship where an experienced person guides a less-experienced one over time.
- Team motivation
- Using recognition, autonomy, purpose, and goal alignment to sustain member engagement and effort.
- Maslow's hierarchy of needs
- A motivation theory ordering needs from physiological, safety, social, and esteem to self-actualization.
- Herzberg's two-factor theory
- A motivation theory distinguishing hygiene factors (which prevent dissatisfaction) from motivators (which create satisfaction).
- Team performance evaluation
- Assessing team effectiveness using metrics, milestones, and member feedback against project goals.
- Negotiation
- A discussion aimed at reaching mutually acceptable agreement among parties with differing interests.
- Communication plan
- A document specifying what information will be shared with which stakeholders, how, and how often.
- Active listening
- Fully concentrating on, understanding, and responding to a speaker to ensure accurate communication.
- Meeting management
- Practices such as agendas, ground rules, and minutes that make meetings efficient and effective.
- Team launch
- The kickoff activity that aligns members on purpose, roles, charter, and ground rules.
- Ground rules
- Agreed behavioral norms that govern how team members interact during the project.
- Force field analysis
- A tool listing driving and restraining forces affecting a change to plan how to strengthen or weaken them.
- Stakeholder buy-in
- Securing support and commitment from those affected by or influential over the project.
- Define phase
- The first DMAIC phase that scopes the problem, goals, customer requirements, and project boundaries.
- Project charter
- A document defining the problem statement, goal, scope, business case, team, and timeline for a project.
- Problem statement
- A concise, factual description of the gap between current and desired performance, without assigning cause or solution.
- Goal statement
- A SMART target for the project's outcome, specifying the metric, baseline, target, and timeframe.
- Business case
- The justification for a project, linking it to strategic goals and quantifying expected benefits.
- Project scope
- The defined boundaries of what is and is not included in the project.
- Scope creep
- The uncontrolled expansion of project scope beyond its original boundaries.
- SMART goals
- Objectives that are Specific, Measurable, Achievable, Relevant, and Time-bound.
- Critical-to-quality tree (CTQ tree)
- A diagram translating broad customer needs into specific, measurable quality requirements.
- Kano model
- A model classifying customer requirements as must-be (basic), one-dimensional (performance), and delighters (excitement).
- Must-be quality
- Basic Kano requirements that cause dissatisfaction if absent but do not delight when present.
- Delighter (excitement) quality
- Unexpected Kano features that greatly increase satisfaction when present but are not missed if absent.
- Customer segmentation
- Dividing customers into groups with similar needs to better target requirements and solutions.
- VOC data collection
- Gathering customer needs through surveys, interviews, focus groups, complaints, and observation.
- Affinity diagram in Define
- Used to group raw VOC statements into themes that reveal underlying customer needs.
- Quality Function Deployment (QFD)
- A method that translates customer requirements into technical specifications using the House of Quality matrix.
- House of Quality
- The primary QFD matrix relating customer wants (whats) to technical requirements (hows) with a correlation roof.
- Project metrics
- The primary, secondary, and consequential measures used to judge project success.
- Primary metric
- The main measure directly tied to the project goal that defines success.
- Secondary metric
- A measure that ensures improving the primary metric does not harm another important outcome.
- Project timeline
- A schedule of milestones and deliverables across the DMAIC phases.
- Gantt chart
- A bar chart showing project tasks along a timeline with start dates, durations, and dependencies.
- Critical path
- The longest sequence of dependent tasks that determines the minimum project duration.
- Work breakdown structure (WBS)
- A hierarchical decomposition of project deliverables into smaller, manageable work packages.
- Project risk analysis
- Identifying and assessing risks that could affect project schedule, cost, or outcome.
- Cost-benefit analysis
- Comparing the expected costs and benefits of a project to justify investment.
- Elevator speech
- A brief, compelling summary of a project's purpose and value for quick stakeholder buy-in.
- Project closure
- Formal completion activities including documentation, validation of benefits, and handoff to the process owner.
- Measure phase
- The DMAIC phase that quantifies the current process performance and validates the measurement system.
- Data collection plan
- A document specifying what data to collect, how, by whom, when, and from where.
- Operational definition
- A precise, agreed description of what is measured and how, ensuring consistent data collection.
- Continuous data
- Measurable data on a continuous scale, such as time, length, or weight.
- Discrete data
- Count or category data, such as number of defects or pass/fail outcomes.
- Population
- The complete set of items or events of interest in a study.
- Sample
- A subset of the population selected to represent it for analysis.
- Random sampling
- Selecting samples so every member of the population has an equal chance of selection.
- Stratified sampling
- Dividing the population into homogeneous strata and sampling from each to ensure representation.
- Sampling error
- The difference between a sample statistic and the true population parameter due to sampling.
- Sample size
- The number of observations collected, balancing precision, confidence, and cost.
- Mean
- The arithmetic average of a data set: sum of values divided by the count.
- Median
- The middle value of an ordered data set, robust to outliers.
- Mode
- The most frequently occurring value in a data set.
- Range
- The difference between the maximum and minimum values: range = max − min.
- Variance
- The average squared deviation from the mean: σ² for a population, s² for a sample.
- Standard deviation
- A measure of dispersion in the same units as the data, equal to √(variance): σ = √(variance). It is the most common measure of spread.
- Coefficient of variation
- A relative measure of dispersion: CV = (σ / mean) × 100%.
- Normal distribution
- A symmetric, bell-shaped distribution defined by its mean µ and standard deviation σ.
- Empirical rule (68-95-99.7)
- For a normal distribution, ~68%, ~95%, and ~99.7% of data fall within 1, 2, and 3σ of the mean.
- Z-score
- The number of standard deviations a value lies from the mean: z = (x − µ) / σ.
- Binomial distribution
- A discrete distribution of the number of successes in n independent trials with constant probability p.
- Poisson distribution
- A discrete distribution modeling the count of events in a fixed interval given a constant mean rate.
- Central Limit Theorem
- The distribution of sample means approaches normal as sample size increases, regardless of the population's shape.
- Measurement System Analysis (MSA)
- The study of measurement variation to ensure data are accurate, precise, and adequate for decisions.
- Accuracy
- How close measurements are to the true value, related to bias.
- Precision
- How close repeated measurements are to one another, related to repeatability and reproducibility.
- Bias (measurement)
- The systematic difference between the average measured value and a reference (true) value.
- Linearity (MSA)
- How consistent the measurement bias is across the operating range of the gauge.
- Stability (MSA)
- The consistency of a measurement system over time when measuring the same item.
- Repeatability
- Variation when one appraiser measures the same item multiple times with the same gauge (equipment variation).
- Reproducibility
- Variation among different appraisers measuring the same item with the same gauge (appraiser variation).
- %Study variation (%GRR)
- Gauge R&R as a percent of total study variation; under 10% is acceptable and over 30% is unacceptable.
- Number of distinct categories (ndc)
- The number of distinct groups a measurement system can reliably distinguish; an ndc of 5 or more is desired.
- Attribute agreement analysis
- An MSA for pass/fail or categorical data assessing within-appraiser, between-appraiser, and vs-standard agreement.
- Kappa statistic
- A measure of categorical measurement agreement beyond chance; values above 0.75 indicate good agreement.
- Histogram
- A bar chart showing the frequency distribution of continuous data across intervals.
- Box plot
- A graphical summary of data showing median, quartiles, and outliers.
- Run chart
- A line plot of data over time used to spot trends, shifts, and patterns before adding control limits.
- Scatter diagram
- A plot of paired data points used to reveal the relationship between two variables.
- Process baseline
- The measured current performance level used as the reference for improvement.
- Pp
- Process performance using long-term overall variation: Pp = (USL − LSL) / (6σ_overall).
- Ppk
- Process performance index for centering using overall variation: Ppk = min[(USL − x̄)/(3σ_overall), (x̄ − LSL)/(3σ_overall)].
- Cpm
- Taguchi capability index penalizing deviation from a target T: Cpm = (USL − LSL) / (6√(σ² + (x̄ − T)²)).
- Confidence interval
- A range, computed from sample data, expected to contain the population parameter with a stated confidence level.
- Standard error
- The standard deviation of a sampling distribution: SE of the mean = σ / √n.
- Analyze phase
- The DMAIC phase that identifies and verifies the root causes of the problem using data.
- Root cause analysis
- The systematic process of identifying the fundamental causes of a problem rather than its symptoms.
- 5 Whys
- An iterative questioning technique that asks 'why' repeatedly to drill down to a root cause.
- Fishbone (Ishikawa) diagram
- A cause-and-effect diagram organizing potential causes into categories such as the 6 Ms.
- 6 Ms
- Common cause categories: Man, Machine, Method, Material, Measurement, and Mother Nature (environment).
- Cause-and-effect matrix
- A tool relating process inputs to customer-weighted outputs to prioritize key inputs (X's).
- Hypothesis test
- A statistical procedure that uses sample data to decide between a null and an alternative hypothesis.
- Null hypothesis (H0)
- The default claim of no effect or no difference that the test seeks to disprove.
- Alternative hypothesis (Ha)
- The claim of an effect or difference that the test seeks to support.
- Type I error
- Rejecting a true null hypothesis (a false positive); its probability is alpha (α).
- Type II error
- Failing to reject a false null hypothesis (a false negative); its probability is beta (β).
- Alpha (significance level)
- The maximum acceptable probability of a Type I error, commonly set at 0.05.
- Beta (β)
- The probability of a Type II error; power equals 1 − β.
- Power of a test
- The probability of correctly rejecting a false null hypothesis: power = 1 − β.
- p-value
- The probability of observing data as extreme as the sample if the null hypothesis is true; reject H0 when p < α.
- Critical value
- The threshold a test statistic must exceed to reject the null hypothesis at a given alpha.
- Z-test
- A hypothesis test for means when the population standard deviation is known and the sample is large.
- t-test
- A hypothesis test for means when the population standard deviation is unknown, using the t-distribution.
- Paired t-test
- A test comparing means of two related (paired) measurements, such as before-and-after on the same units.
- Two-sample t-test
- A test comparing the means of two independent groups.
- F-test
- A test comparing two variances or assessing overall significance in ANOVA using the ratio of variances.
- One-way ANOVA
- A test comparing the means of three or more groups based on a single factor.
- Two-way ANOVA
- An analysis of variance examining the effects of two factors and their interaction on a response.
- Degrees of freedom
- The number of values free to vary in a calculation, used to select the correct reference distribution.
- Chi-square test
- A test for categorical data assessing goodness-of-fit or independence using observed vs expected counts.
- Chi-square goodness-of-fit
- A χ² test comparing an observed frequency distribution to an expected distribution.
- Chi-square test of independence
- A χ² test assessing whether two categorical variables in a contingency table are associated.
- Contingency table
- A cross-tabulation of counts for two categorical variables used in chi-square analysis.
- Nonparametric tests
- Distribution-free tests used when data are ordinal or not normally distributed.
- Mann-Whitney U test
- A nonparametric test comparing two independent groups' medians or distributions.
- Kruskal-Wallis test
- A nonparametric alternative to one-way ANOVA comparing three or more independent groups.
- Levene's test
- A test for equality of variances across groups that is robust to non-normality.
- Correlation
- A measure of the strength and direction of a linear relationship between two continuous variables.
- Correlation coefficient (r)
- Pearson's r ranges from −1 to +1, indicating the strength and direction of linear association.
- Coefficient of determination (R²)
- The proportion of variation in the response explained by the model: R² = r² for simple regression.
- Simple linear regression
- A model fitting a straight line ŷ = b0 + b1x to predict a response from one predictor.
- Multiple regression
- A model predicting a response from two or more predictor variables.
- Logistic regression
- A regression for binary or categorical outcomes that models the probability of an event.
- Residual
- The difference between an observed value and the value predicted by the model: residual = y − ŷ.
- Residual analysis
- Examining residuals for randomness, normality, and constant variance to validate a model.
- Least squares method
- Fitting a regression line by minimizing the sum of squared residuals.
- Multicollinearity
- A condition where predictors in a regression are highly correlated, inflating coefficient variance.
- Multi-vari study
- A graphical analysis separating variation into positional, cyclical, and temporal families to focus root-cause search.
- Positional variation
- Within-unit or within-piece variation across locations, one of the multi-vari families.
- Cyclical variation
- Piece-to-piece variation among consecutive units, one of the multi-vari families.
- Temporal variation
- Time-to-time variation across hours, shifts, or days, one of the multi-vari families.
- Confounding (analysis)
- A situation where the effect of one variable is mixed with another, obscuring the true cause.
- Improve phase
- The DMAIC phase that develops, tests, and implements solutions to address verified root causes.
- Design of Experiments (DOE)
- A systematic method to plan tests that efficiently estimate factor effects and interactions on a response.
- Factor
- An independent input variable deliberately varied in an experiment.
- Level
- A specific setting or value of a factor in an experiment.
- Response variable
- The measured output (dependent variable) studied in an experiment.
- Main effect
- The average change in the response produced by changing one factor from low to high.
- Interaction effect
- When the effect of one factor on the response depends on the level of another factor.
- Full factorial design
- An experiment testing every combination of factor levels, e.g., 2ᵏ runs for k two-level factors. It estimates all main effects and interactions.
- Fractional factorial design
- An experiment testing a carefully chosen subset of combinations to reduce runs while estimating key effects.
- Confounding (DOE)
- Aliasing where two or more effects cannot be separately estimated in a fractional design.
- Resolution (DOE)
- A measure of how badly effects are confounded; Resolution III, IV, and V indicate increasing clarity of effects.
- Replication
- Repeating experimental runs to estimate experimental error and improve precision.
- Randomization
- Running experimental trials in random order to spread the effect of unknown variables.
- Blocking
- Grouping experimental runs to remove the effect of a known nuisance variable from the analysis.
- Center points
- Runs at the midlevel of all factors used to detect curvature in the response.
- Response surface methodology (RSM)
- A set of designs and models used to optimize a response and locate the best factor settings.
- Central composite design (CCD)
- An RSM design adding axial and center points to a factorial to fit a second-order model.
- Box-Behnken design
- A three-level RSM design for fitting quadratic models without extreme corner runs.
- Screening design
- A small experiment, often fractional factorial, used to identify the vital few significant factors.
- Effect plot
- A graph such as a main-effects or interaction plot used to interpret DOE results.
- Pareto of effects
- A bar chart ranking the magnitude of factor effects to identify significant ones in a DOE.
- Solution selection matrix
- A weighted tool ranking candidate solutions against criteria such as cost, impact, and feasibility.
- Pilot study
- A small-scale trial of a solution to validate effectiveness and uncover issues before full rollout.
- Implementation plan
- A detailed plan specifying tasks, owners, timelines, and resources to deploy a solution.
- Lean
- A methodology focused on eliminating waste and maximizing customer value through flow and pull.
- Seven wastes (muda)
- Defects, Overproduction, Waiting, Non-utilized talent, Transportation, Inventory, Motion, and Excess processing (DOWNTIME).
- Value-added activity
- A step that transforms the product or service in a way the customer is willing to pay for.
- Non-value-added activity
- A step that consumes resources but adds no value the customer would pay for.
- Value stream map (VSM)
- A diagram of material and information flow used to identify waste and design a leaner future state.
- Takt time
- The pace of production needed to meet demand: takt time = available production time / customer demand.
- Kanban
- A visual signaling system that controls production and inventory using pull based on actual demand.
- 5S
- A workplace organization method: Sort, Set in order, Shine, Standardize, and Sustain.
- Kaizen
- A philosophy of continuous, incremental improvement involving everyone in the organization.
- Kaizen event
- A focused, short-duration team workshop to rapidly improve a specific process.
- Poka-yoke
- Mistake-proofing devices or methods that prevent or immediately detect errors.
- SMED
- Single-Minute Exchange of Dies: techniques to reduce setup and changeover time to under ten minutes.
- Total Productive Maintenance (TPM)
- A program that maximizes equipment effectiveness through operator-led preventive and autonomous maintenance.
- Overall Equipment Effectiveness (OEE)
- A productivity metric: OEE = Availability × Performance × Quality.
- Heijunka
- Production leveling that smooths volume and mix to reduce variation and waste.
- Single-piece flow
- Producing and moving one unit at a time through the process to reduce work in process.
- Pull system
- A system that produces only in response to downstream demand rather than forecasts.
- Theory of Constraints (TOC)
- A method that improves throughput by identifying and managing the system's bottleneck constraint.
- Spaghetti diagram
- A drawing of the physical flow of people or material used to reveal excess motion and transport.
- Standard work
- Documented best-known method specifying sequence, timing, and steps for a task.
- Cellular manufacturing
- Arranging workstations in close sequence to enable smooth flow and reduce transport and inventory.
- Risk assessment of solutions
- Evaluating potential failures and unintended consequences of a proposed solution before implementation.
- Control phase
- The final DMAIC phase that sustains gains through monitoring, documentation, and control plans.
- Control plan
- A document detailing how each key input and output will be monitored, measured, and responded to.
- Statistical Process Control (SPC)
- Using control charts and statistical methods to monitor and control process variation over time.
- Common cause variation
- Inherent, random variation present in a stable process from many small sources.
- Special cause variation
- Variation from an identifiable, non-random source that signals the process is out of control.
- Control limits
- Statistically derived boundaries, typically at ±3σ from the center line, signaling special-cause variation.
- Specification limits
- Customer- or engineering-defined boundaries for acceptable product; distinct from control limits.
- Center line
- The average of a control chart statistic, plotted between the control limits.
- Variable control chart
- A control chart for continuous measurement data, such as X̄-R or I-MR.
- Attribute control chart
- A control chart for count or classification data, such as p, np, c, or u charts.
- X̄-R chart
- A pair of charts monitoring the subgroup average (X̄) and range (R) for continuous data, typically subgroups of 2 to 9.
- X̄-S chart
- A pair of charts monitoring the subgroup average (X̄) and standard deviation (S), used for larger subgroups (n ≥ 10).
- I-MR chart
- An individuals and moving-range chart used when data are collected one at a time (subgroup size of 1).
- p-chart
- An attribute chart tracking the proportion defective with variable sample sizes.
- np-chart
- An attribute chart tracking the number defective with a constant sample size.
- c-chart
- An attribute chart tracking the count of defects per unit with a constant inspection area.
- u-chart
- An attribute chart tracking defects per unit when the sample size varies.
- Rational subgrouping
- Selecting subgroups so within-subgroup variation reflects only common cause and shifts appear between subgroups.
- Western Electric rules
- A set of run rules that flag out-of-control patterns, such as a point beyond 3σ or 2 of 3 beyond 2σ.
- Run rules
- Patterns such as trends, runs, and zone violations used to detect special causes on a control chart.
- Out-of-control signal
- A point or pattern indicating special-cause variation that requires investigation.
- EWMA chart
- An exponentially weighted moving-average chart sensitive to small, sustained process shifts.
- CUSUM chart
- A cumulative-sum chart that accumulates deviations to detect small shifts quickly.
- Pre-control
- A simple chart-free method using zones relative to specification limits to control a centered, capable process.
- Control chart selection
- Choosing the chart type based on data type (variable vs attribute) and subgroup size.
- Mistake-proofing (control)
- Embedding poka-yoke devices into the process to sustain error-free performance.
- Standard operating procedure (SOP)
- A documented, step-by-step instruction ensuring consistent execution of a process.
- Visual management
- Using visual cues such as boards and color codes to make process status and abnormalities obvious.
- Response plan
- Predefined corrective actions to take when a control chart or metric signals a problem.
- Process control system
- The combination of charts, procedures, and responses that maintains a process in control.
- Audit (control)
- A periodic review verifying that controls and procedures are being followed and remain effective.
- Sustaining the gains
- Practices that lock in improvements through documentation, training, monitoring, and ownership transfer.
- Control chart capability link
- A process must be in statistical control before its capability indices (Cp, Cpk) are meaningful.
- Project handoff
- Transferring responsibility for the improved process to the process owner with documentation and training.
- Lessons learned
- Documented insights from a project used to improve future projects and share knowledge.
- Total Quality Management (TQM)
- An organization-wide philosophy of continuous improvement and customer focus that predates Six Sigma.
- PDCA cycle
- Plan-Do-Check-Act: an iterative four-step cycle for continuous improvement and control.
- Design for Six Sigma (DFSS)
- A methodology for designing new products or processes to meet Six Sigma quality from the start.
- DMADV
- A DFSS roadmap: Define, Measure, Analyze, Design, and Verify.
- DFSS vs DMAIC
- DFSS creates new designs to prevent defects, while DMAIC improves existing processes by reducing defects.
- Robust design
- Designing products and processes to perform consistently despite variation in noise factors.
- Taguchi methods
- Robust design techniques using orthogonal arrays and the signal-to-noise ratio to minimize variation.
- Taguchi loss function
- A model expressing quality loss as proportional to the squared deviation from target: L = k(y − T)².
- Signal-to-noise ratio
- A Taguchi metric measuring robustness; higher values indicate performance less sensitive to noise.
- Noise factors
- Uncontrollable or costly-to-control variables that cause variation, addressed through robust design.
- Control factors
- Design parameters that can be set to make a product robust against noise factors.
- Orthogonal array
- A balanced experimental matrix used in Taguchi methods to study many factors with few runs.
- Tolerance design
- Setting component tolerances to balance quality and cost after parameter design.
- TRIZ
- A theory of inventive problem solving using principles and patterns abstracted from patents to resolve contradictions.
- QFD in DFSS
- Quality Function Deployment cascades customer needs into design requirements across multiple houses of quality.
- Design FMEA (DFMEA)
- An FMEA focused on potential failure modes in a product or system design.
- Process FMEA (PFMEA)
- An FMEA focused on potential failure modes in a manufacturing or service process.
- Design scorecard
- A DFSS tool that predicts and tracks the capability of a new design against its requirements.
- Reliability (DFSS)
- The probability a product performs its intended function without failure over a specified time and conditions.
- Pugh matrix
- A concept-selection tool scoring design alternatives against a datum on weighted criteria.
- Critical parameter management
- Identifying and controlling the few design parameters that most affect customer-critical performance.
