Roberts (1959) first introduced the exponentially weighted moving average (EWMA) control scheme. Using simulation to evaluate its properties, he showed that the EWMA is useful for detecting small shifts in the mean of a process. The recognition that an EWMA control scheme can be represented as a Markov chain allows its properties to be evaluated more easily and completely than has previously been done. In this article, we evaluate the properties of an EWMA control scheme used to monitor the mean of a normally distributed process that may experience shifts away from the target value. A design procedure for EWMA control schemes is given. Parameter values not commonly used in the literature are shown to be useful for detecting small shifts in a process. In addition, several enhancements to EWMA control schemes are considered. These include a fast initial response feature that makes the EWMA control scheme more sensitive to start-up problems, a combined Shewhart EWMA that provides protection against both large and small shifts in a process, and a robust EWMA that provides protection against occasional outliers in the data that might otherwise cause an out-of-control signal. An extensive comparison reveals that EWMA control schemes have average run length properties similar to those for cumulative sum control schemes.

No abstract

No abstract

No abstract

The effects of dry heat setting conditions on streaks in finished carpets made from nylon 66 staple yams are considered. Two different polymers were subjected to controlled variation in heat-set temperature, residence time, and chamber steam pressure on a Suessen machine according to a factorial design. Yarns from some of the items were made into finished carpets, and other items were implanted into these carpets using two adjacent creel positions. Streak intensity grades were assigned to these implants, and then an empirical model was fit to the data relating streak intensity to the differences between the background and implant items in heat setting conditions. Streaks caused by variation in heat setting conditions are associated with differences in dye uptake rather than with physical or textural differences between tuftlines. If the Suessen is operated at a nominal temperature of 200°C, then surprisingly large differences in heat setting conditions are required in order to cause an objectionable streak. Furthermore, the effect of heat-set temperature differences on carpet streaks depends on the dyeing system used. For a continuous dyeing system, which has a dye uptake minimum at 200°C, any tuftline-to-tuftline heat-set temperature variation from 200°C causes dark streaks. On the other hand, the Beck system has a linear rather than a U-shaped dye response to heat-set temperature, and increases considerably above 200°C cause light streaks when that tuftline is implanted within a group of tuftlines having experienced a 200°C temperature.Prevention of streaks in finished carpets has long been a primary concern in the carpet industry as well as for nylon fiber producers. LeGault [ 8,9 ] discussed the wide variety of causes of streaks and concludes that differences in dry heat setting conditions are an important factor. Miller and Southern [ I I ] have recently shown that bulk-related streaks can be induced before heat setting. Several authors (Peters and White [ 13 ], Holfeld and Shepard [ 4 ] , Seu [ 14 ] , Kobsa el al. [ 7 ] , and Murthy et al. [ 12 ] ) attributed the effect of dry heat setting conditions to morphological changes in the fiber, but also suggested that increasing heat-set temperature causes an oxidative degradation of the dye sites. Ghosh [5] ] conducted a prior study of the effects of heat-set temperature differences on carpet streaks for both Suessen ( dry ) and Superba ( wet ) heat setting.He found that temperature differences of 4°C on the Superba caused severe streaks, but the Suessen required a larger temperature difference for streaking, and there was. less agreement between the panel members regarding streak intensity. At Monsanto (Gibson et al.[ 4 ] ), we conducted a factorial experiment to study the effects of Suessen heat setting conditions on fibcr properties under the standard process configuration and with an energy conservation package. We found effects of temperature and residence time on fiber structural and dyeing properties and observed more fiber vanability ...