The process control charts created by Shewhart in the 20 s and still in use today are efficient in signaling large shifts in the quality characteristics of a process (e.g. shifts greater than two standard deviations, in the case of the chart for means); they are however slower in the case of small and moderate shifts, in which case CUSUM and EWMA schemes are known to be more efficient, as are the recently developed adaptive charts, also called variable parameter charts because some or all of their design parameters (sample size, sampling interval and control limits) are allowed to vary during the operation, according to the information of the latest sample. In this thesis, looking for an enhancement in the performance of control charts for attributes, the strategy of adaptive charts (using a variable sampling interval) is incorporated to the EWMA scheme. The proposed scheme is applied to c charts for detecting small shifts in the number of nonconformities in a production process. A mathematical model is developed for calculation of the performance measures of the chart, and a performance analysis is carried out for several values of c0 and c1 (in- and out-of-control number of nonconformities), together with a comparison with other control charts for nonconformities. The results show the advantage of the proposed scheme in the majority of the analyzed situations, through faster detection of a range of shifts.