A multi-criteria optimization and decision-making approach for improvement of food engineering processes

Abstract

The objective of this study was to propose a multi-criteria optimization and decision-making technique to solve food engineering problems. This technique was demonstrated using experimental data obtained on osmotic dehydration of carrot cubes in a sodium chloride solution. The Aggregating Functions Approach, the Adaptive Random Search Algorithm, and the Penalty Functions Approach were used in this study to compute the initial set of non-dominated or Pareto-optimal solutions. Multiple non-linear regression analysis was performed on a set of experimental data in order to obtain particular multi-objective functions (responses), namely water loss, solute gain, rehydration ratio, three different colour criteria of rehydrated product, and sensory evaluation (organoleptic quality). Two multi-criteria decision-making approaches, the Analytic Hierarchy Process (AHP) and the Tabular Method (TM), were used simultaneously to choose the best alternative among the set of non-dominated solutions. The multi-criteria optimization and decision-making technique proposed in this study can facilitate the assessment of criteria weights, giving rise to a fairer, more consistent, and adequate final compromised solution or food process. This technique can be useful to food scientists in research and education, as well as to engineers involved in the improvement of a variety of food engineering processes.

Towards a holistic approach for multi-objective optimization of food processes: A critical review

“… Organoleptic quality is described either by denaturing kinetics (or conversely development kinetics) of compounds relating to organoleptic appraisal of a product (Gergely et al, 2003;Kahyaoglu, 2008;Yuen et al, 2000 …), or by sensory scores. These scores directly express the appraisal of product quality by the consumer, but they are based on a posteriori evaluation (Abakarov et al, 2013;Singh et al, 2010 …). Sensory scores can be aggregated to produce integrative indicators of overall appraisal, provided they have been evaluated on a common scale (e.g.…”

“…Black box models are based on experimental or compiled data, and require approaches which employ model parameter identification algorithms to be determined once the mathematical structure has been chosen. There are countless examples of modelling approaches in the literature; Response Surface Methodology (RSM) is the most common in food processing, particularly for modelling osmotic dehydration (Abakarov et al, 2013;Arballo et al, 2012;Corzo and Gomez, 2004;Eren and Kaymak-Ertekin, 2007;Singh et al, 2010;Themelin et al, 1997;Yuan et al, 2018), in which the complex mechanisms involved (transfer through vegetable cell membranes) are well-suited to the black box approach. The field of possible modelling approaches is wide, also encompassing Artificial Neural Networks (Asgari et al, 2017;Chen and Ramaswamy, 2002;Izadifar and Jahromi, 2007;Karimi et al, 2012), gene expression programming (Kahyaoglu, 2008), fuzzy logic, pure algorithms, etc.…”

“…If the decision-maker is faced with a multitude of objectives, it may be difficult to rationally and consistently attribute the weights. That is why there are methods to help the decision-maker to prioritize the objectives: the AHP method (Analytic Hierarchy process -Saaty (1990)) for example, which designates a method even capable of ranking the solutions, includes a step of defining the weights by comparing the objectives (or indicators) in pairs, is used in Abakarov et al (2013). A score of between 1 and 9 is attributed to each objective depending on its significance compared to every other objective, and the results are aggregated using a given formula to provide a numerical value for the weight of each objective.…”

“…Some authors (Alam et al, 2010;Arballo et al, 2012) use in addition an MADM based on desirability functions to select the best solution from those filtered. Another method, this time based on using tables (known as the "Tabular method"), is used in Abakarov et al (2013). Its principle is to rank the values adopted by each indicator according to whether they must be maximized or minimized.…”

“…However, it makes it possible to eliminate solutions where an indicator adopts a very low value or zero, under an "aggressive" strategy as mentioned above. Besides these "primary" aggregation functions (as per Marler and Arora (2004)), it is possible to adopt more complex aggregation strategies, at least two of which have been identified within this work:  Integration strategy within a more complex decision-making aid framework, such as the AHP method employed in Abakarov et al (2013): the steps for determining the weights, set out in section 4, lead to a weighted sum aggregation.  "Mathematical" strategy, aimed at increasing the complexity of the aggregation functions.…”