Brussels, 22 Jul 2003
A EUREKA project consortium has designed a software system aimed at helping companies comply with chemical-limit regulations.
Companies involved in the food and pharmaceutical industry are particularly concerned with accurately measuring the chemical content of their products in order to assure traceability and avoid contamination. However, short-term fluctuations in temperature, humidity and air-pressure or variability in the performance of the measurer can give rise to uncertain measurement results, an issue which the international organisation of standardisation (ISO) says needs to be addressed.
The MUSAC project involved 20 consortium partners from Germany, Austria and Switzerland. With a total budget 0.7 million euro, the partners sought to tackle the issue of 'measurement uncertainty'. As Matthias Roesslein, MUSAC project leader from the Swiss-based EMPA research institute explained, 'In the past, companies have already been challenged for uncertainty statements. In food control, for example, it is important to show that pesticides are really below a limit defined by government.'
To ensure more reliable chemical measurements, the MUSAC software system, assesses 'measurement uncertainty' step-by-step. The system first asks for the quantities of each chemical and possible variables or 'uncertainties', such as changes in the laboratory temperature. It then calculates the margin of error and simulates uncertainty levels over the production base.
Partners expect that the software will help prevent wrong decisions being made as well as to eliminate cost intensive repetitions of a measurement series. As Mr Roesslein explained, the system is already in the process of being commercialised, with the help of Nestle. 'They are thinking about using this product in their labs worldwide,' he said. MUSAC partners have also created a spin-off company to market the new software.
Furthermore, the development of the chemical measurement software has led to the development of a related system to quantify uncertainty in physical measurements, such as component length and electric current.
For further information, please