 |
Statistica Process Analysis is comprised of two modules which include comprehensive implementations
of process capability analysis, gage repeatability and reproducibility analysis, Weibull analysis,
sampling plans, and variance components for random effects, each of which is described in the
following sections.
Statistica Process Analysis is compatible with Windows 95, Windows 98, Windows NT, Windows 2000,
Windows XP, Windows Me.
|
Statistica Process Analysis includes a comprehensive selection of options for computing process
capability indices for grouped and ungrouped data (e.g., Cp, Cr, Cpk, Cpl, Cpu, K, Cpm, Pp, Pr,
Ppk, Ppl, Ppu), normal/distribution-free tolerance limits, and corresponding process capability
plots (histogram with process ranges, specification limits, normal curve). In addition, instead
of these normal distribution indices and statistics, the user can choose estimates (e.g., Cpk,
Cpl, Cpu based on the percentile method) based on general non-normal distributions (Johnson and
Pearson curve fitting by moments), as well as all other common continuous distributions including
the Beta, Exponential, Extreme Value (Type I, Gumbel), Gamma, Log-Normal, Rayleigh, and Weibull
distributions.
|
|
The program will compute maximum-likelihood parameter estimates for those distributions, and it
provides numerous options for evaluating the fit of the respective distribution to the data,
including the frequency distribution with observed and expected frequencies, the Kolmogorov-Smirnov
d statistic, histograms, Probability-Probability (P-P) plots, and Quantile-Quantile (Q-Q) plots.
An option is also available for automatically fitting all distributions, and choosing the
distribution that best fits the data.
|
Repeatability/reproducibility experiments with single or multiple trials can be generated and analyzed.
The data for the R&R analysis can be arranged in raw-data format, or tabulated in a standard R&R data
sheet format (as used in many publications of the American Society for Quality Control, and manuals
of the Automotive Action Group). Results include estimates of the components of variance (repeatability
or equipment variation, operator or appraiser variation, part variation, operator-by-part variation,
operators-by-trials, parts-by-trials, operators-by-parts-by-trials), which can be computed based on
the range method, or the ANOVA table. If based on the ANOVA table, confidence intervals for the
variance components will also be estimated.
|
|
Additional statistics for the variance components include the percent-of-tolerance, process variation,
and total variation. The program will also compute descriptive statistics by operator/part, range and
sigma charts by operators/parts, box-and-whisker plots, and the summary R&R plot. Comprehensive
selections of methods for estimating variance components for random effects are also available in
the designated Statistica Variance Components module (included in this application, see below), and
the General Linear Models module available in Statistica Advanced Linear/Non-Linear Models.
|
The Weibull analysis options provide powerful graphical techniques for exploiting the power and
generalizability of the Weibull distribution. The user can produce Weibull probability plots and
estimate the parameters of the distribution, along with confidence intervals for reliability.
Probability plots can be computed for complete, single-censored, and multiple-censored data, and
parameters can be estimated from hazard plots of failure orders. Estimation methods include Maximum
Likelihood (for complete and censored data), weighting factors based on linear estimation techniques
for complete and single-censored data, and Modified Moment Estimators which are unbiased with respect
to both the mean and variance.
|
|
Confidence intervals are computed for the shape, scale, and location parameters, as well as for the
percentiles.The program includes graphical goodness of fit tests, and the Hollander-Proschan,
Mann-Scheuer-Fertig, and Anderson-Darling tests of goodness of fit. Note that the Generalized
Linear Models module of STATISTICA Advanced Linear/Non-Linear Models provides options for fitting
generalized linear models from the exponential family of distributions to normal and non-normal
data.
|
Fixed and sequential sampling plans can be generated for normal and binomial means, or Poisson
frequencies. Results include the sample sizes, operating characteristic (OC) curves, plots of the
sequential plans with or without data, expected (H0/H1) run lengths, etc. Note that Statistica
Power Analysis also provides options for computing required sample sizes and power estimates for
a large number of research designs (e.g, ANOVA) and data types (e.g., for binary counts, censored
failure time data, etc.).
STATISTICA Process Analysis is an add-on package that requires a base product such as Statistica
Base or Statistica Quality Control Charts.
|
|
|
|
|
|
