PurposeThe purpose of this paper is propose a methodology for monitoring industrial processes that cannot be stabilized, but are nevertheless capable. Design/methodology/approach The proposed procedure uses the C P (u,v) family of capability indices proposed by Vännman (including the indices C PK , C PM , C PMK ) combined with one-sided two-out-of-three and three-out-of-four run rules strategies. FindingsThis paper introduces a new strategy, where capability indices are monitored in place of the classical sample statistics like the mean, median, standard deviation or range. Practical implicationsWhen doing a capability analysis it is recommended to first check that the process is stable, e.g. by using control charts. However, there are occasions when a process cannot be stabilized, but is nevertheless capable. Then the classical control charts fail to efficiently monitor the process position and variability. The approach suggested in this paper overcomes this problem. Originality/valueThe experimental results presented in this paper demonstrate how the new proposed approach efficiently monitors capable processes by detecting decreases or increases of capability level.

In order to monitor unstable but capable processes Castagliola and Vännman have recently suggested a procedure based on an EWMA approach, called EWMA capability chart, for monitoring Vännman's Cp(u,v)-family of capability indices and showed how their proposed approach efficiently monitors capable processes by detecting a decrease or increase in the capability level. The goal of this paper is to investigate the efficiency of this capability chart in terms of ARL. The procedure used for computing this ARL is presented and simple guidelines for obtaining approximations to the optimal EWMA parameters are proposed.

When performing a capability analysis it is recommended to first check that the process is stable, for example, by using control charts. However, there are occasions when a process cannot be stabilized, but it is nevertheless capable. Then the classical control charts fail to efficiently monitor the process position and variability. In this paper we propose a new strategy to solve this problem, where capability indices are monitored in place of the classical sample statistics such as the mean, median, standard deviation, or range. The proposed procedure uses the Cp(u,v) family of capability indices proposed by Vännman combined with a logarithmic transformation and an EWMA approach. One important property of the procedure presented here is that the control limits used for the monitoring of capability indices only depend on the capability level assumed for the process. The experimental results presented in this paper demonstrates how this new approach efficiently monitors capable processes by detecting changes in the capability level.

In order to monitor unstable but capable processes Castagliola & Vännman have recently suggested a procedure based on an EWMA approach, called EWMA capability chart, for monitoring Vännman's Cp(u, v)-family of capability indices and showed how their proposed approach efficiently monitors capable processes by detecting a decrease or increase of the capability level. The goal of this paper is to investigate the efficiency of this capability chart by computing optimal EWMA parameters yielding the smallest possible ARL's. The method for computing the ARL is explained and a table summarizing optimal values for the EWMA parameters (λ*,K*) is given for the Cpk chart.