چكيده به لاتين
Abstract
In several applications in real world, the quality of a process or a product is better characterized by a functional relationship, referred to as profile, between a response quality characteristic and one or more explanatory variables. Monitoring the quality profile is used to understand and to check the stability of this functional relationship over time. In some real applications, the profile can be represented adequately by a fuzzy linear regression model where the response quality characteristic is linguistic, imprecise, vague, and deficient.
The purpose of this research was to develop some novel approach for monitoring process/product profiles in fuzzy environment. A model in fuzzy linear regression was developed to construct the quality profiles by using multi-objective linear programming and then fuzzy individuals and moving-range (I-MR) control charts were developed to monitor both intercept and slope of fuzzy profiles to achieve an in-control process in phase I. Subsequently a multivariate approach was developed to monitor process/product fuzzy quality profiles in phase I. The multivariate approach includes three fuzzy multivariate control charts which were developed by using fuzzy set theory to monitor fuzzy profiles in order to achieve a stable process in phase I. Then, two univariate approach including F-IMR and F2-EWMA3 were developed for online monitoring of fuzzy profile in phase II. The F-IMR approach is comprised two individuals control chart and two moving range control chart which were developed through resolution identity technique. The F2-EWMA3 approach constitutes two fuzzy EWMA control chart and a conventional EWMA control chart. Furthermore, the influence of model parameters in constructing fuzzy regression were investigated at decision maker’s different preferences. The performance of developed approaches of phase I were investigated in the basis of signal probability in various out-of-control scenarios through a simulation study. Compared with univariate approach, the results indicated a good performance of multivariate approach in detecting all sized shifts in process profiles. The performance of developed approaches of phase II weere explored in the basis of Average Run Length in different shift scenarios throughout a simulation study. The results indicated superior performance of F2-EWMA3 approach. A case study in tourism industry and the other one in petrochemical industry were presented to show the applicability and validity of our approaches.
Keywords: Fuzzy Profile, Fuzzy Individuals Control Chart, Fuzzy Moving Range Control Chart, Fuzzy Hoteling Control Chart, Fuzzy Multivariate Exponentially Weighted Moving Average Control Chart, Fuzzy Multivariate Cumulative Sum Control Chart.
