International Journal of Experimental Design and Process Optimisation 

About: International Journal of Experimental Design and Process Optimisation is an academic journal published by Inderscience Enterprises Ltd.. The journal publishes majorly in the area(s): Response surface methodology & Taguchi methods. It has an ISSN identifier of 2040-2252. Over the lifetime, 86 publications have been published receiving 345 citations.

Alternatives to resolution IV screening designs in 16 runs

Abstract: The resolution IV regular fractional factorial designs in 16 runs for six, seven, and eight factors are in standard use. They are economical and provide clear estimates of main effects when three-factor and higher-order interactions are negligible. However, because the two-factor interactions are completely confounded, experimenters are frequently required to augment the original fraction with new runs to resolve ambiguities in interpretation. We identify non-regular orthogonal fractions in 16 runs for these situations that have no complete confounding of two-factor interactions. These designs allow for the unambiguous estimation of models containing both main effects and a few two-factor interactions. We present the rationale behind the selection of these designs from the non-isomorphic 16-run fractions and illustrate how to use them with an example from the literature.

An investigation of surface texture parameters during turning of a reinforced polymer composite using design of experiments and analysis

Abstract: The influence of cutting speed and feed rate during turning on the arithmetic mean roughness (Ra), the maximum peak to valley (Rt), and the fractal dimension (D) of a glass fibre polymer composite (Ertalon 66 GF-30) was experimentally investigated. Test specimens in the form of bars and a P20 cemented carbide cutting tool were used with the cutting depth kept constant during the experiment. Robust design using an orthogonal matrix experiment was conducted and the experimental results were analysed using an ANOM and an ANOVA analysis approach. Based on the statistical analysis of the experimental results it was found that the arithmetic mean roughness, the maximum peak to valley and the fractal dimension depend mainly on the feed rate parameter. Also, based on the interaction charts and evaluation experiments it was found that regression modelling applies only for the arithmetic mean roughness.

Choice of second-order response surface designs for logistic and Poisson regression models

Abstract: Response surface methodology is widely used for process development and optimisation, product design, and as part of the modern framework for robust parameter design. For normally distributed responses, the standard second-order designs such as the central composite design and the Box-Behnken design have relatively high D and G efficiencies. In situations where these designs are inappropriate, standard computer software can be used to construct D-optimal and I-optimal designs for fitting second-order models. When the response distribution is either binomial or Poisson, the choice of an appropriate design is not as straightforward. We illustrate the construction of D-optimal second-order designs for these situations and show that they are considerably better choices than the standard designs. We present an example applying this approach to optimisation of an etching process.

Rethinking the truncated normal distribution

Abstract: A truncated distribution is a conditional distribution which is restricted by the domain of a random variable in special situations. Truncated distributions were introduced more than one hundred years ago, but only a few papers have examined the standardisation of a truncated distribution, especially the standardisation of a truncated normal distribution. However, the traditional truncated standard normal distribution, derived from the truncation of a standard normal distribution, has varying mean and variance, depending on the location of truncation points. As a result, its statistical analysis may not be done on a consistent basis. The contribution of this paper is three-fold. First, we develop a standard truncated normal distribution by standardising a truncated normal distribution. The truncated mean and variance of this distribution are zero and one, respectively, regardless of the location of the truncation points, thereby being more consistent with the well-known standard normal distribution. Second, we develop the cumulative probability table of the standard doubly truncated normal distribution as a set of guidelines for engineers and scientists. Finally, we prove that the variance of the truncated distribution is always smaller than the variance of the original distribution.

A response surface methodology based approach to machining processes: modelling and quality of the models

Abstract: Various techniques for developing prediction models for various machining performance measures such as surface roughness/surface integrity, cutting force, tool life/tool wear etc in machining processes are available. These methods include, but are not limited to, analytical, numerical, empirical, and artificial intelligence (AI) based methods. While empirical modelling often employs the use of response surface methodology (RSM), however, proper understanding must be established regarding RSM-based models with respect to their development, validation and acceptability. Therefore, the general framework for developing RSM-based prediction models and testing their quality are discussed in this paper. This is followed by a practical surface roughness (Rt) model developed for precision grinding of silicon, a machining process that is very difficult to model. The result shows that the procedural modelling frameworks work well for the Rt developed model.