13 research outputs found
Monitoring Multiple Linear Profile based on EWMA Control Charts by using Ridge Regression Estimators: an application to Wind Tunnel data of NASA Langley Research Centre
No full textIn many quality control studies the performance of a product or process is usually characterized by a single response variable However, in some applications of quality control, the performance of a product or a process can be best characterized by a linear relationship between a response variable and one or more explanatory variables (Noorossana et al., 2011). But, when more than one explanatory variables are involved in the profile it may indicate the presence of high collinearity among explanatory variables which is called multicollinearity (Gujarati et al., 2012). It should be noted that if the multicollinearity is neglected during the profile monitoring, then the designed control charts applied in phase II lack the sufficient effectiveness in detecting shifts or out of control signals. In this paper, the effect of the multicollinearity on the monitoring of multiple linear profiles has been studied and some new type of Exponentially Weighted Moving Average (EWMA) control charts for Intercept, Slopes and Mean Squared Error (MSE) by using Ridge Regression Estimators (RRE) have been proposed in order to provide the solution for multicollinearity. An application of wind tunnel data by NASA Langley Research Centre was used for monitoring profiles based on proposed EWMA control charts for Intercept, Slopes and MSE. The performance of the proposed EWMA control charts have been evaluated on the Average Run Length (ARL) criterion, the results indicated that the proposed EWMA control charts obtained from RRE for Intercept, Slopes and MSE outperform the existing control charts obtained from Ordinary Least Squared (OLS) estimator
New Shewhart and EWMA Type Control Charts using Exponential Type Estimator with Two Auxiliary Variables under Two Phase Sampling
No full textIn this paper, two new control charts have been proposed, one is shewhart-type and other one is EWMA-type control chart. The proposed control charts are based on the exponential type estimator for mean proposed by Noor-ul-Amin and Hanif (2012). We name them as DS-Shewhart control chart and DS-EWMA control chart. The results shows that the DS-Shewhart control chart shows more efficient results to traditional/simple Shewhart and EWMA control charts whereas, the DS-EWMA control chart shows more efficient results to traditional Exponentially Weighted Moving Average (EWMA) and Cumulative Sum (CUSUM) control charts because they uses the information from two phase sampling with two auxiliary variables. The proposed control charts can be used for efficient monitoring of the production process in manufacturing industries. The control limits of the proposed chart are based on estimator, its mean square errors and bias. A simulated example has been used to compare the proposed and traditional/simple EWMA and CUSUM control charts performance based on the average run length-out of control (ARL1). It is observed that the proposed chart performs better than existing EWMA and CUSUM control charts. At the end of the paper a real life implementation of the proposed control charts is also provided
Development and Regression Modeling of Dirt Resistive Latex Façade Paint
No full textA highly dirt-resistant paint for building façades without chemicals harmful to nature and the environment would resolve the unattractive disfigurement of building walls caused by dirt. The current ranking of Pakistan in terms of air pollution is 139th. A set of dirt-resistant paint formulae was constructed with the aid of computer programming. From this set, the best dirt-resistant paint formula was explored and identified. The final determination of the optimum formulation was based on statistically planned experiments conducted in the laboratory and in a natural environment. In order to achieve high-quality results, the best available laboratory equipment were used. The results obtained were analyzed and conclusions were drawn using appropriate statistical techniques. The procedure started with the selection of appropriate raw materials and generation of a target population of 543,143 paint formulations by adopting Basic Language computer programming. The average pigment volume concentration (PVC) percentage was computed using theory and found to be 54.98% for the target population paint formulations, verifying the literature results. Experimentation and statistical analysis were performed to compare the classical conventional agitator with the latest lab equipment such as a nano mill, and it was concluded that the nano mill performs better on average than the conventional agitator in the preparation of paint formulations. Hence, the sample of paint formulations was prepared on a nano mill and tested in the laboratory using advanced available technology for the analysis and comparison of paint properties to determine the best paint formulation. The results were analyzed using the Analysis of Variance (ANOVA) technique, and it was concluded that the newly developed paint has the highest dirt resistance on average. The final selected formula, No. 50 (the newly developed paint), was compared with the three best conventional paints available in the Pakistan market in a natural environment for a period of almost one year. A regression model was also constructed to study the effect of environmental factors like time, temperature, and humidity on the dirt resistance of paints. It was found that the newly developed paint formulation is the most environmentally friendly. It performs equally well as one conventional paint and has higher dirt resistance than two other conventional paint formulations containing harmful chemicals. The regression model of dirt resistance involving variables including time, temperature, and humidity shows that these factors significantly affect the dirt resistance of a given paint at a 5% level of significance. For a given paint, 95.34% of the variation in the dirt resistance is due to and explained by the given factors. The regression model is useful for predicting the average dirt resistance of a given paint with a certain level of confidence. The project exemplifies the work of applied research from conceptualization to successful commercialization in the paint industry
Development of Algae Guard Façade Paint with Statistical Modeling under Natural Phenomena
No full textAlgaecides are chemicals that cause serious health problems. Conventional paints contain algaecides to improve the algae resistance on the paint film. Present research has suggested an environment-friendly paint formulation that focuses on developing algae resistance without having algaecides. In this research, algae growth on newly developed paint is modeled by incorporating dirt resistance of paint and natural phenomena including humidity, temperature, and time, respectively. The fitted Model revealed explained variation of 59.65% in the average algae growth, of which, dirt resistance, humidity, temperature, and some of their interactions play significant role in this variation. The model suggests that the proposed newly developed paint without algaecides is more resistant to algae growth and significantly decreased the average algae growth rate by 0.53% as compared to conventional paints. Keeping the effect of all other factors constant, if dirt resistance of paint (Dc value) increases by one percent, average algae growth decreases by 12.98%; when temperature increases by 1 °C, average algae growth decreases by 22.4%; a positive unit change in the joint linear effect of dirt resistance, temperature, and humidity caused a decrease in average algae growth by 0.0031%. It was also observed that the individual effect of the humidity variable was inversely related with average algae growth. However, the combination of humidity and temperature, humidity and dirt resistance, humidity and time, and the quadratic effect of humidity were found to increase the average algae growth rate. The cubic effect of temperature variable by one degree centigrade resulted in decrease of average algae growth by 0.000907%
On exploring the generalized mixture estimators under simple random sampling and application in health and finance sector
No full textThere are numerous studies where data on population units’ auxiliary variables and attributes are simultaneously available. Therefore, due to cost-effectiveness and ease of recording, the study variable and several linearly related auxiliary variables are recorded. These auxiliary variables are commonly observed as quantitative and qualitative (attributes) variables and are jointly used to estimate the study variable’s population mean using a mixture estimator. In order to achieve this, a family of generalized mixture estimators was proposed under simple random sampling with the aim of improving performance under symmetrical and asymmetrical distributions. In addition, the estimator’s behavior for various sample sizes was examined with regard to its convergence to the normal distribution. The suggested generalized mixture estimator’s mean square error is deduced up to the first order of approximation. It is discovered that for the normal, uniform, Weibull, and gamma distributions, the suggested estimator estimates the population mean of the study variable with greater accuracy than the competing estimators. It is also revealed that when the proposed estimator converges to normality, the sample size is at least taken as 110, 1000, and 120. Furthermore, the implementation of three real-life datasets related to the finance and health sectors is also presented to support the proposed estimator’s significance
A comparison of some new and old robust ridge regression estimators
No full textRidge regression is used to circumvent the problem of multicollinearity among predictors and many estimators for ridge parameter k are available in the literature. However, if the level of collinearity among predictors is high, the existing estimators also have high mean square errors (MSE). In this paper, we consider some existing and propose new estimators for the estimation of ridge parameter k. Extensive Monte Carlo simulations as well as a real-life example are used to evaluate the performance of proposed estimators based on the MSE criterion. The results show the superiority of our proposed estimators compared to the existing estimators.</p
Efficient control charting methodology based on Distance Weighted Mean for normal distribution
No full textThis research suggests a Distance Weighted Mean (DWM) based control chart under normal distribution implementing Simple Random Sampling (SRS). The control limits are calculated using the quantile point method. The control chart\u27s performance is assessed using the Average Run Length (ARL) statistic. The numerical findings are illustrated using samples of sizes 3 and 5. The ARL1 values are determined using Monte Carlo Simulation for increasing and decreasing shifts in the location parameter ranging from 5% to 30%. Using the ARL1 measurement, the proposed DWM control charts are compared to the existing Shewhart control charts. According to the comparison analysis, the suggested DWM control chart surpasses the competing Shewhart control chart. The real-life application of the proposed DWM control chart is also shown by using the lifetime of the light bulb (in hours). The results suggest that the proposed DWM control chart can be a useful tool for monitoring process mean shifts, especially when the sample size is large, and the magnitude of the shift is significant
