Integration of expert knowledge in milk microfiltration optimization

Skim milk 0.1 µm microfiltration, used to fractionate dairy proteins, can be carried out with different types of membranes (ceramic or polymeric) in various operating modes. Despite the dairy sector’s interest in microfiltration, this operation is not formally optimized. Formulating the microfiltration as an optimization problem, which is the preliminary step of the optimization process, is essential. It requires knowledge from both scientists and experts (users and equipment providers) about food processing, dairy product production and equipment manufacturing.

We formulated the multi-objective optimization problem of milk microfiltration, in an innovative way, by integrating expert knowledge. We proposed a methodology, adapted from Hobballah et al. (2018), to establish the optimization objectives and graphically represent the relations between variables and optimisation objectives. This methodology is based on 4 iterative steps.

Methodology proposed to formulate multi-objective optimization of 0.1 µm microfiltration of skim milk

For the specific case of optimization of skim milk microfiltration, five conflicting objectives covering economic and technical aspects as well as fraction characteristics, were defined (maximisation of casein concentration in retentate, maximisation of whey protein concentration in permeate, maximisation of whey protein recovery ratio in permeate, minimisation of investment cost, minimisation of production cost). They were influenced by 36 variables, including five decision variables used to control the microfiltration process (membrane type, feed flow, permeation flux, volume reduction ratio and number of stages).

This approach opens new perspectives for optimizing microfiltration and, more generally, food processes lacking scientific knowledge by integrating expert knowledge.

Collaborations

• Membrane Processing Team UMR 1253 Milk and egg Science and Technology laboratory (STLO), INRAE-Institut AgroRennes;

• Knowledge assembly team (AsCo), JRU 5295 Institute of Mechanical Engineering (I2M); Bordeaux University - Arts et Métiers Institute of Technology - Bordeaux INP - CNRS - INRAE;
• R&D, Boccard, Montauban-de-Bretagne.

This study was supported by a grant from the Brittany Region (contract no. 16006734, INRA convention 30001292) and from FEDER (contract no. EU000171, INRA convention 30001293).

Read more

Belna, M., Ndiaye, A., Taillandier, F., Agabriel, L., Marie, A.-L., Gésan-Guiziou, G., 2020. Formulating multiobjective optimization of 0.1 μm microfiltration of skim milk. Food and Bioproducts Processing 124, 244–257. https://doi.org/10.1016/j.fbp.2020.09.002

Contacts

Geneviève Gésan-Guiziou • scientist UMR STLO

Amadou Ndiaye • scientist UMR I2M Bordeaux