Accurate statistical methods
The food industry needs reliable decision-support information, which should be based on relevant summaries of information from several academic disciplines. Nofima’s number crunchers develop and use multi-block methods to explore connections in complex datasets and interpret the results.
In our blog you will find information about our publications, relevant seminars and software, visiting reseachers and others things of interest. The blogs is called nofimamodeling.org
We develop methods that contribute to:
- good and efficient product
- stable product quality
- better utilization of raw materials
- less waste
- less variation in a production process
- more environmentally friendly production
Access to information has never been greater. There is an enormous need both in industry and in research to be able to understand relationships between different types of information, available from many datasets. This is related to the complexity within any one academic discipline, and the fact that communication across traditional subject boundaries is necessary in order to solve large problems. Such an understanding of several datasets is vital in, for example, consumer research, process modelling, microbiology, packaging, bioinformatics, and research into food and health.
The development of relevant and user-friendly methods in multivariate statistics and mathematical methods, which are used to analyse and interpret relationships between different datasets, is an important mission for our scientists.
A long tradition in number crunching
One of Nofima’s goals is to be at the forefront of the development of statistical and mathematical methods, such that we can exploit the possibilities created by new and more powerful measurement techniques. We have a long tradition of data modelling and the use of statistics, and the subject has been under continuous development. Nofima has combined chemical ideas with statistics and mathematical calculation techniques since the 1970s. The discipline provides the background research that is necessary for other research fields, even though the customer may be unaware of this.
How statistics can be used – in product development
In order for food producers to achieve a higher success rate when developing new food products or improving existing products, they need answers to several types of question.
It is important during product development to take into consideration the nutritional composition of the product, its sensory properties, and consumer preferences. Knowledge about its shelf-life and effects on health are also needed. All of these disciplines produce large quantities of data, and product development may suffer if these quantities are not properly managed in concert. High-quality statistical methods are an absolute requirement in order to deal with management of the data. The methods make it possible to predict, interpret and understand relationships between the datasets.
How statistics can be used – stable food product quality
A food product is affected not only by the raw materials but also the production process. There are generally large variations in the quality of raw materials due to seasonal variation, climate, species variation, etc., which makes it very difficult to ensure that a product always tastes the same. Customers have a well-defined expectancy about the taste and consistency of a product, and it is important that the industry meets this. Guidance and control of the production contributes to a stable quality of the products, and profound knowledge about the raw materials, their processing, the production process, spectroscopy, chemical measurement and statistics is needed in order to achieve this.