LPSCE cheat sheet

Five Lean principles in order: specify value, map the value stream, create flow, establish pull, pursue perfection.

TIMWOODS lists the eight wastes: Transportation, Inventory, Motion, Waiting, Overproduction, Overprocessing, Defects, Skills underutilized.

Overproduction is ranked as the most damaging waste because it hides and generates the other seven.

Value-added test: the customer must pay for it, it must transform the product or information, and it must be done right the first time.

Takt time = available production time / customer demand; cycle time above takt time means capacity is short.

Little's Law: WIP = throughput x flow time; reducing WIP shortens lead time at constant throughput.

5S sequence: Sort, Set in order, Shine, Standardize, Sustain; Sustain is the step that fails most often.

Pull systems authorize work with kanban; a kanban card is a replenishment signal, not a work order and not a schedule.

Value stream map measures both process time and lead time; PCE (process cycle efficiency) = value-added time / total lead time.

Six Sigma statistically targets 3.4 defects per million opportunities (DPMO) in the long term, allowing a 1.5 sigma process shift.

Sigma level from DPMO: 2 sigma ~308,537; 3 sigma ~66,807; 4 sigma ~6,210; 5 sigma ~233; 6 sigma ~3.4.

DPMO = (defects / (units x opportunities per unit)) x 1,000,000; yield = 1 - DPMO/1,000,000.

Rolled throughput yield multiplies the yields of each step; it is always less than or equal to the lowest single-step yield.

DMAIC phases: Define, Measure, Analyze, Improve, Control; each gate requires a tollgate review before proceeding.

A defect is any output outside specification; a unit can carry multiple defect opportunities, so count them explicitly.

Cp compares specification width to process spread (USL-LSL)/6sigma; Cpk additionally penalizes off-center processes.

Six Sigma uses data and statistics to reduce variation; Lean uses flow and waste removal to reduce lead time; Lean Six Sigma combines both.

DMADV (Design for Six Sigma) applies when no process exists or redesign is required; DMAIC applies to existing processes.

Charter elements: business case, problem statement, goal statement, scope, team, timeline, and expected benefit.

Problem statement rules: measurable, time-bounded, no cause and no solution stated.

SMART goal: Specific, Measurable, Achievable, Relevant, Time-bound, with baseline and target quantified.

VOC to CTQ flow: collect verbatim customer comments, group into needs (Kano or affinity), translate into measurable CTQs.

Kano model categories: must-be, performance, and delighter; must-be gaps cause dissatisfaction but do not delight.

SIPOC maps Suppliers, Inputs, Process, Outputs, Customers at a high level; it is a scope-setting tool, not a detailed flowchart.

Stakeholder analysis plots influence against interest; high influence and high interest require active management.

Project scope should be bounded by a clear start and stop point; scope creep is controlled through a change log.

Data types: continuous (variable) allows more powerful statistics; discrete (attribute) requires larger samples.

MSA components: accuracy (bias, linearity, stability) and precision (repeatability, reproducibility).

Gage R&R rule: under 10% of tolerance acceptable, 10-30% conditional, over 30% unacceptable; ndc should be 5 or more.

Repeatability is equipment variation from one appraiser; reproducibility is appraiser variation between operators.

Process capability requires a stable process in statistical control; computing Cpk on an unstable process is invalid.

Cp = (USL - LSL)/(6 sigma); Cpk = min(USL - mean, mean - LSL)/(3 sigma).

Ppk uses overall standard deviation and long-term data; Cpk uses within-subgroup sigma and short-term data.

Sample size for a mean: n = (z x sigma / margin of error)^2; doubling precision requires four times the sample.

Attribute agreement analysis and kappa measure how well appraisers agree beyond chance.

5 Whys drills to root cause by asking why repeatedly; stop when the cause is actionable and systemic.

Fishbone (Ishikawa) categories: Man, Machine, Method, Material, Measurement, Environment (6M).

FMEA: RPN = Severity x Occurrence x Detection; higher RPN gets priority, but severity alone can force action.

Hypothesis test structure: H0 states no difference or no effect; H1 states the effect being tested for.

Type I error (alpha) rejects a true null; Type II error (beta) fails to reject a false null; power = 1 - beta.

p-value is the probability of observing data this extreme if H0 is true; p < alpha leads to rejecting H0.

Test selection: continuous data, two groups use t-test; three or more groups use ANOVA; discrete data use chi-square.

Normality is checked first (Anderson-Darling or Shapiro-Wilk); non-normal continuous data call for transformation or nonparametric tests.

Correlation measures association only; causation requires designed experiments or strong process knowledge.

Solution selection matrix scores options against weighted criteria such as impact, cost, risk, and cycle time.

Poka-yoke prevents errors at the source; it is preferred over inspection because inspection is a non-value-added detection step.

Kaizen event is a focused, short-duration improvement burst, typically 3-5 days, on a bounded process.

SMED reduces changeover time by separating internal from external setup tasks.

DOE (Design of Experiments) varies factors together to estimate main effects and interactions with fewer runs.

Pilot testing uses a small-scale trial with defined success metrics before full rollout.

Standard work documents the current best method: takt time, work sequence, and standard in-process inventory.

Before/after comparison should use the same measurement system and statistically test the difference, not just compare averages.

Implementation plan elements: actions, owners, dates, resources, and a risk or failure-mode review.

Control charts separate common-cause from special-cause variation; only special causes justify local intervention.

Xbar-R chart for subgroups of 2-9; Xbar-S for larger subgroups; I-MR for individual measurements.

p-chart for proportion defective; np-chart for count defective with constant sample; c-chart for defect counts per unit; u-chart for defects per unit with varying sample.

Western Electric rules: one point beyond 3 sigma, 2 of 3 beyond 2 sigma, 4 of 5 beyond 1 sigma, 8 in a row on one side of center.

Control limits come from process data, not from specification limits; specification limits belong to the customer.

Process in control but not capable means reduce variation or center the process; out of control means find and remove the special cause.

Control plan lists the characteristic, specification, measurement method, sample size, frequency, and reaction plan.

Standard work and visual management sustain gains; response plans must name who acts when a chart signals.

Project closure includes documented savings, updated procedures, and handoff of ownership to the process owner.

Project phases: initiation, planning, execution, monitoring and control, closure; a tollgate review follows each DMAIC phase.

RACI matrix assigns Responsible, Accountable, Consulted, Informed; only one person should be Accountable per task.

Critical path is the longest sequence of dependent tasks; it determines the minimum project duration.

Team stages: Forming, Storming, Norming, Performing, Adjourning.

Change curve stages: denial, resistance, exploration, commitment; resistance is expected and should be surfaced, not suppressed.

Kotter's change steps begin with creating urgency and building a guiding coalition, and end with anchoring changes in culture.

Effective team size for a Six Sigma project is typically 4-8 members with a trained Green or Black Belt lead.

Risk management: identify, assess probability and impact, plan response, monitor; maintain a risk register throughout.

Communication plan defines audience, message, channel, and frequency; sponsors remove barriers and secure resources.