Introduction

It is the cellular activities, such as those of enzymes or nucleic acids, that determine the performance of bioprocesses. Unfortunately, they are difficult or impossible to measure on-line. Consequently, we need to monitor or estimate variables such as the concentrations of biomass, metabolites, substrates, products or regulator molecules. These dependent variables often can represent the process objectives more or less directly. If not, models need to be constructed and exploited to link the variables that can be measured or monitored to the desired process objectives. We also need to know all the relevant operating variables that influence the desired objectives significantly and also how they influence them.

Terminology

Measuring means to describe a variable qualitatively and/or quantitatively. The variable can be a physical one, which is usually quite simple to measure, a chemical one or a biological one, which can be quite tricky to measure. It involves an analytical method and instrument(s). Measurements of variables of interest are normally indirect measuring methods. To measure glucose concentration, for example, we convert a known aliquot of β-d-glucose in the presence of (dissolved) oxygen using the enzyme glucose oxidase into gluconolactone and hydrogen peroxide, which we eventually oxidise using an amperometric electrode: the final measure is the electric current that we relate to the glucose concentration by using stoichiometric and kinetic models.

Monitoring means to obtain process information without manual or personnel intervention and without disturbing the process. This includes cases where samples must be withdrawn from the process.