The increasing complexity of products and processes in modern production systems, together with the demand for high-quality products, has necessitated the use of methods such as design of experiments (DOE) and statistical process control (SPC) in various environments. This special issue features six papers dealing with the developments and industrial practices of these methods for process / product quality improvement and related issues. The papers have been contributed by members of ENBIS ( E uropean N etwork for B usiness and I ndustrial S tatistics) and some were presented at the ENBIS 8th Annual Meeting held in Athens, September 21–25 2008. The papers’ subject matter ranges from methodological developments to practical applications. Four are on DOE and two on SPC. Arnouts, Goos and Jones investigate strip-plot experimental designs, which are economically attractive in situations where the factors are hard to change and the process under investigation consists of two distinct stages. Dehlendorff, Kulahci and Anderson present a modeling framework for analyzing computer experiments with two types of variation. Based on a case study of an orthopedic surgical unit with both controllable and uncontrollable factors, it is shown that the structure of variation can be effectively modeled with linear mixed effects models and generalized additive models. Ginsburg and Ben-Gal explore a recent DOE alphabetic optimality criterion, the Vs-optimality, which seeks to minimize the variance of the optimal solution of an empirically fitted model. Viles and colleagues apply DOE to a lift test rig and determine the most important guiding system factors that affect a lift’s comfort. The papers on SPC tackle practical issues in quality control. Schoonhoven and Does consider different estimators of the standard deviation and investigate design schemes for the ¯ X control chart under non-normality. Pehlivan and Testik study exponential EWMA control charts, which utilize the time-between-events in the control statistics. Effects of departures from the assumed exponential model are described and design settings for robust performance are provided. of statistical quality and reliability engineering