Cookies

We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Department of Mathematical Sciences

Seminar Archives

On this page you can find information about seminars in this and previous academic years, where available on the database.

Statistics Seminars: "Design of Experiments in an Industrial Setting - Combining Recipe, Process and Storing Factors"

Presented by Froydis Bjerke,

13 March 2002 14:00 in CM221

"The most important objective in development or improvement of consumer (food) products it to achieve robust and stable products with the desired quality and characteristics. Type and amount of ingredients (recipe), processing steps, packaging and storage conditions will all affect the product properties through complex relations. In order to understand how these elements interact in their influence on product attributes, the product development team must take all elements into account simultaneously. The tools of experimental designs are vital to this approach. Shelf life studies are important for food manufacturers,and product robustness to environmental factors could be analysed by using methods of repeated measures - multiple outcomes in time or space measured on each subject. To be able to identify and understand possible effects from - and interactions between - production factors, storage factors and repeated measures on the response, the statistical model must describe a complete empirical functional relationship between the response and all factors and interactions. When product samples are made from a factorial design in several production factors, then split and stored for a time period under different conditions while repeated measures of the product during storage are taken, a split-plot-like situation is generated, which is also a multistratum unit structure. The general aspects of combining a response surface design to repeated measure-ments in a mulitistratum unit structure will be discussed and illustrated by a case study related to the production and storage of low-fat mayonnaise. Some issues of analysing and modelling such data will be discussed. These comprise graphical plots, choice of design and statistical model with emphasis on the general linear mixed model (SAS). "

"visiting Department of Statistics, Newcastle University"

Contact sunil.chhita@durham.ac.uk for more information