STAVEX - 2026 Pricing, Features, Reviews & Alternatives

STAVEX is an on-premises, intranet-based Statistical Design of Experiments (DoE) software application designed for scientists, chemists, chemical engineers, and process engineers across pharmaceuticals, biotechnology, chemicals, food and beverage, automotive, industrial manufacturing, and academic research. Deployed as a continuously available server service on Windows environments and accessed through standard modern web browsers (Firefox, Edge, Chrome), STAVEX requires no external internet connection for day-to-day operation and no external database, supporting both single-user local and multi-user server installations under a named-user licensing model.

The platform guides users through the complete DoE workflow — project specification, experimental design selection, data collection, regression analysis, and result optimization — without requiring prior statistical knowledge. A wide design library covers screening designs (Plackett-Burman, Lin, Lin-Plackett-Burman, Desperado, 2- and 3-level full and fractional factorial), modelling and optimization designs (CCD, Box-Behnken, hexagon, pentagon, Doehlert, D-optimal), and a full suite of mixture designs (simplex-vertex, simplex-lattice, simplex-centroid, axial, projected, combined mixture-process). Flexible factor types include quantitative, qualitative, alternative, and mixture factors, with support for factor restrictions, blocking of noise factors, design adaptation for hard-to-change factors, and automatic completion of incomplete designs. Multi-response optimization uses a weighted desirability function, and Design Space calculation supports Quality by Design (QbD) workflows aligned with FDA PAT and ICH Q8 regulatory frameworks.

An integrated rule-based expert system automatically rates and ranks feasible experimental designs, performs consistency checks on factor restrictions, and generates plain-language analysis reports that flag outliers, disturbing factors, insufficient model fit, and confounding of effects. Regression analysis supports separate models per response variable, automatic transformation analysis, discriminant analysis for qualitative responses, and confidence intervals for predictions. A large graphics library includes 2- and 3-factor contour plots, response surface plots, ternary contour plots, half-normal plots, plot matrix representations for up to 4-5 factors simultaneously, and model diagnostic plots, all with interactive tooltips and export in PNG and SVG formats. Flexible PDF and HTML reporting with user-selected content and ordering supports both internal documentation and regulatory submissions. The interface is available in English, French, German, and Spanish, and 12 months of maintenance and helpdesk support are included after purchase.