DoEStat

🧪 DoEStat – Statistical Design of Experiments (DoE) for professional use

DoEStat is the complete DoE suite for development, formulation and process optimisation: every relevant design type, modern analysis and multi-response optimisation in one Windows application – no add-on modules, no forced cloud. Developed in Germany by statisticians for use in regulated industries.

✅ Why DoEStat?

🧩 One program, start to finish – planning, data entry, analysis, optimisation and report in a single application, with no separately licensed extensions
📚 Scientifically grounded – every method follows a published standard reference (Montgomery, Myers, Cornell, Box & Hunter, Scheffé, Jones & Nachtsheim …); numerical kernels are checked against R and NIST reference data
🔒 Your data stays local – native .NET 9 Windows application, projects are files on your machine or network share, no user account, no cloud
🔄 Switch without losing data – existing Design-Expert projects (.dxpx) import directly, with a preview and a list of notes before anything is taken over
🌐 Bilingual – German and English interface with contextual help and result interpretation
🏭 GxP-ready – validation support with a categorised changelog (Impact on Validation / Numerical Impact / Non-Validation-Relevant)

📐 Designs – 19 design types, no add-on module

🔹 Screening: full factorial (2^k and mixed-level), fractional factorial 2^k−p with alias structure and resolution III–V, Plackett-Burman, definitive screening (DSD), min-run resolution IV and V
🔹 Response surfaces: central composite (CCD, rotatable / face-centred / inscribed), Box-Behnken
🔹 Optimal designs: D-, A-, I- and G-optimal by coordinate exchange – able to hold already-executed runs fixed
🔹 Mixtures: simplex-lattice, simplex-centroid, extreme vertices (constrained), optimal mixture designs, mixture × process, mixture-amount
🔹 Robust & special: Taguchi arrays (L4 … L36), robust parameter design (inner/outer), nested (hierarchical), Latin hypercube
🔹 Split-plot / split-split-plot: hard-to-change factors automatically get their own randomisation stratum
🔹 Augment instead of starting over: fold-over, axial points, lack-of-fit points or a criterion-driven top-up of an existing design

📊 Analysis – what becomes of the measurements

🔹 Model & fit: regression with ANOVA (R², adjusted R², predicted R²), lack-of-fit test, model selection by AIC/BIC with term hierarchy, Box-Cox transformation, generalised regression (ridge)
🔹 Separating effects: half-normal plot, Pareto chart with t and Bonferroni limits, Lenth PSE, two-stage DSD analysis
🔹 Diagnostics & prediction: studentised residuals, normal Q-Q, leverage, Cook's D, DFFITS, VIF, confidence and prediction intervals, confirmation runs
🔹 Structure & distribution: split-plot and split-split-plot (REML), nested ANOVA, mixed models, binomial & Poisson GLM, mixture ANOVA (Scheffé)
🔹 Optimisation & robustness: desirability (Derringer-Suich), Pareto front, profiler & ramps, overlay / sweet spot, propagation of error (POE), Monte Carlo with defect rate and Cpk, robust parameter design (S/N)
🔹 Evaluating the design: power per effect, replicates from power, D/A/G efficiency, FDS curves, prediction variance, strategy comparison

📈 18 charts – from design to optimum

FDS curves, prediction variance, correlation matrix, power curves, half-normal, Pareto, interaction, residuals, normal Q-Q, Box-Cox, contour, 3-D surface, profiler, overlay (sweet spot), ramps, mixture trace, ternary contour, Monte Carlo – all interactive, all produced by DoEStat itself.

🌍 Typical applications

✔️ Pharma & biotechnology (formulation, process validation, QbD)
✔️ Chemistry & process engineering
✔️ Materials science & materials development
✔️ Food & cosmetics formulations (mixture designs)
✔️ Manufacturing & process optimisation (robust design, Six Sigma)
✔️ Universities & research institutions

🔑 Licence models

🔹 Commercial – named-user annual licence, full feature set, e-mail support (EN/DE); volume pricing from 5 licences
🔹 Academic – identical feature set for teaching and non-commercial research (proof of institution required)
🔹 Free trial – 30 days, full feature set, no payment details needed
🔹 After expiry the installation stays usable as a free viewer: open projects, inspect results, export reports (watermarked) – colleagues, reviewers and auditors need no licence of their own

All licences and the trial are available through STATCON – use the request form or write to vertrieb@statcon.de. We are happy to advise on the right licence model and prepare a quote.

💬 In short: DoEStat – Design of Experiments without compromise

DoEStat brings the complete DoE workflow – from planning to multi-response optimisation – into a modern, local Windows application with a clear licence model. If you work with Design-Expert you will feel at home right away and can bring your projects with you. STATCON supports you as authorised partner in Germany, Austria and Switzerland with sales, training and DoE consulting.

📥 Try it for free or request a quote:
➡ Visit the DoEStat product page
➡ Quote, trial, academic and volume licences: vertrieb@statcon.de or via the request form