Home / Lean Six Sigma
Graphical summary report (descriptive statistics, confidence intervals and Anderson-Darling normality test), hypothesis tests (1-sample, 2-sample and paired t, proportions, variances, Poisson rates and A × B comparison), power and sample size (t tests, proportions, variances, Poisson, ANOVA, 2^k factorial, equivalence and margin of error), one-way and multi-factor ANOVA (general linear model, multiple comparisons and equal variances), chi-square (association and goodness of fit), nonparametric tests (sign, Wilcoxon, Mann-Whitney, Kruskal-Wallis, Mood, Friedman and runs), Pearson and Spearman correlation, simple and multiple linear regression, tolerance intervals (normal and nonparametric) and Monte Carlo simulation with tolerance stack-up (worst case, RSS); shortcuts to process capability, control charts, the multi-vari chart, attribute capability and Weibull analysis.
Paste one or several columns (machines, shifts, lots) and get, for each, the summary report with histogram and normal curve, boxplot, confidence intervals for the mean and median, Anderson-Darling test and full descriptive statistics, plus boxplots and a probability plot comparing the groups.
Descriptives, normality, CICompare supplier A × B, old × new method, shift × shift or before × after: the page picks the test from the type of result, shows the difference with its confidence interval, the no-real-difference band, the power and how many more samples are needed, and translates it into parts and money per year. It also runs 1- and 2-sample and paired t-tests, 1 and 2 proportions, 1 and 2 variances and Poisson rates.
t, proportion, variance, A × BFind out how many parts, items or runs to measure before comparing suppliers, machines, shifts or methods: sample size for the desired power, power of a given sample or the smallest detectable difference, with power curves, a scenario table and sampling cost.
How many parts to measureCompare means of suppliers, machines, shifts or process settings: one-way ANOVA (with Welch) and a general linear model with up to 4 factors, interactions and covariates, Tukey, Fisher, Dunnett, Hsu MCB and Games-Howell comparisons, equal-variance test, effects, interaction and residual plots.
Tukey, Welch, interactionPaste a count table (press × shift, supplier × inspection) or the raw data and see whether there is an association, with expected counts, contributions, adjusted residuals, Cramér’s V and Fisher for 2×2; or test whether defect types follow historical proportions and whether defects per part follow the Poisson.
Association, goodness of fit, FisherCompare medians without assuming normality: repair times by shift, bearing life from two suppliers, setup against the target, inserts tested on the same machines, and check whether a production sequence is random. With median CIs, ranks, tie adjustment and plots by group.
Medians, ranks, runsPaste a table with two or more variables and get the Pearson and Spearman correlation of every pair, with confidence intervals and p-values, an interactive scatterplot with line, ellipse, LOWESS and histograms, a matrix plot, an ordered heat map and a report card.
Pearson, Spearman, matrix plotPaste a table with the response and the process factors and get the equation, coefficients with p-values and VIF, ANOVA, adjusted and predicted R², stepwise selection, residual, effects, contour and surface plots, a what-if simulator, predictions with CI and PI and the optimal factor settings.
Equation, R², residuals, what-ifPaste the measurements of one or more samples and get the limits that, at the chosen confidence, contain the requested percentage of parts: normal method with exact or Howe k factor, two-sided or one-sided, and nonparametric method with achieved confidence, plus the Anderson-Darling test, histogram, probability plot and the smallest percentage of parts within the specification.
Range of 95%/99% of partsWrite the equation of the result (flow, gap, strength, cost) or the dimension chain of an assembly, choose the distribution of each input and simulate 100,000 parts: histogram with the specification, percent out and PPM, Pp and Ppk, contribution of each input to the variation, tornado chart, sensitivity curves, convergence and how much to tighten (or loosen) each tolerance to reach the target Ppk; in a stack-up, comparison with worst case and RSS.
Equation, worst case, RSSPaste measurements from Excel and get a histogram with automatic or adjustable bins, normal curve, mean and ±3s, specification limits with % out, a dot plot for small samples, overlaid, stacked or paneled groups, a plain-language reading of the shape and a frequency table.
Shape, bins, % outPaste one or several columns (machines, shifts, suppliers) and get the box plot in vertical or horizontal orientation, with Minitab-method quartiles, whiskers to 1.5 × IQR, outliers labeled with value and row, the mean, individual points, a median-CI notch, ordering by median and specification limits.
Quartiles, outliers, groupsBasic quality tool to see whether two variables move together: paste X and Y (and, if you like, a group column) and the page draws the scatter diagram with the line, r and R² on the figure itself, a plain-language strength statement, the points far from the line and the quadrant reading, ready to download as PNG, PDF or spreadsheet.
X × Y, r, R², groupsPaste measurements (single column, subgroups or ID) or enter mean and standard deviation: Cp, CPL, CPU, Cpk, Pp, Ppk, Cpm with 95% CI, observed and expected PPM, Z.Bench, control charts, Box-Cox, Johnson, Weibull and lognormal, and the payoff of reducing variation.
Cp, Cpk, PPM, sigma levelPaste measurements to build Xbar-R, Xbar-S, I-MR, I-MR-R/S, P, NP, C, U, Laney P′/U′, EWMA or CUSUM charts with tests 1 to 8, stages, historical limits, unequal subgroups and the within sigma for Cpk.
X̄-R, I-MR, P, U, EWMA, tests 1–8Paste measurements with up to 4 factors (position on the part, part, hour, shift) and see the multi-vari chart with means connected by level and each family’s share of the variation; or analyze data order on the run chart, with clustering, mixtures, trend and oscillation tests.
Families of variation, run chartFor pass/fail characteristics or defect counts: paste the samples and get % defective or DPU with an exact confidence interval, PPM, process Z, P or U chart with tests, cumulative plot with confidence bounds, rate × sample size, histogram, report card and how many more parts to inspect to prove the target.
% defective, Z, DPUPaste times to failure and the units still running (censored) to get the Weibull, lognormal, exponential or normal fit by maximum likelihood or LSXY, the probability plot with CI, distribution identification, B10 life and MTTF with intervals, Kaplan-Meier, supplier or lot comparison, expected warranty failures and a demonstration test plan.
Censored Weibull, B10, warrantyImportant notice: Indus Metrics tools produce estimates and pre-sizing results from the data you enter and from published engineering methods. They do not replace design, calculations or reports signed by a licensed professional engineer (PE), nor verification against the applicable codes, standards and regulations, such as OSHA rules and the ASME codes. Material and fuel properties vary: check manufacturer data or your own process analyses before deciding. By using this site you agree to the Terms of Use.
© 2026 Indus Metrics. Found an error? Write to contato@seudominio.com.