04/13/2021 | Process Innovation

Digital twin makes for clean cosmetics tanks

Machines learn from humans

An EnviroChemie plant models processes in real time to tailor the treatment of waste water from cosmetics production at Kneipp GmbH.

At Kneipp GmbH in Würzburg, Germany, even the waste water smells good – of spruce needles, lavender flowers and lemon. Not a surprise for a producer of body care and health products. Nevertheless, the residues from production of bath oil, shower gel and skin cream have to be carefully removed from the waste water. The company pays a lot of attention to the sustainable use of resources, thus noteworthy amounts of waste water occur only at the cleaning station for the containers in which the raw materials are delivered: fragrant essential oils, but also vegetable oils and surfactants. Apart from the raw material residues, the waste water also contains cleaning agents that are used to clean the containers.

Only as much chemistry as necessary

The waste water is cleaned with a Split-O-Mat® system from EnviroChemie GmbH and can then be promptly be discharged into the sewage system. A digital twin is now being developed for this plant, which will in the future contain information about the water that accumulates in the cleaning station. Via a scanned barcode, the twin learns whether a container is currently being cleaned of shower gel, lavender oil or bubble bath, among others. Simultane- ously, it receives online measurement data from the EnviroChemie system on the quantity and typical parameters of the waste water such as redox potential, conductivity, turbidity and pH value. The system then calculates the amount and composition of the resulting waste water and models the chemical-physical processes of the waste water treatment plant in real time. With this information, the cleaning programme is selected so that only as few chemicals as necessary are used, and the process is as sustainable as possible. The goal is always to keep the waste water quality consistently high.

The waste water treatment plant is operated in batches; thus, the optimal treatment programme can be selected for each waste water batch – and the necessary reaction time can be maintained.

Learning from humans

The operators initially evaluate the recommendations of the digital twin. This evaluation is used by the digital twin to further improve the calculations and selection of the appropriate cleaning programme.


Dr Eva Bitter

is head of the Process Engineering, Research and Development department at EnviroChemie GmbH, a plant manufacturer of sustainable system solutions for all tasks in the field of industrial water.


Keywords in this article:

#water technology, #digitalisation

Find more contributions:

Detailed search in the magazine


Always up to date

With our newsletter you will receive current information on ACHEMA on a regular basis. You are guaranteed not to miss any important dates.

Subscribe now