Analysis of the effect of process variables on the osmotic dehydration of mango and process optimisation using a genetic algorithm approach

Intermediate moisture fruit products obtained by using osmotic dehydration are gaining increasing commercial interest. The lower moisture content increases the product stability, yet the higher sugar content may impair consumer acceptability. Therefore, process conditions should be defined so that a given moisture content is reached while minimising sugar gain. The objective of this work was to study the effect of process variables on the moisture and soluble solids contents of mango pieces osmotically dehydrated in a sucrose solution. Experiments were performed at 30, 40 and 50 °C in 45, 55 and 65 °Brix sucrose solutions, from 20 to 240 minutes. A Pareto analysis showed that moisture and soluble solids contents were affected by all the variables tested, with time having the major effect, followed by the solution concentration. The interactive effects solution concentration/time and temperature/time were also significant. Optimization of the process was performed using a Genetic Algorithm. The dependence of moisture and soluble solids contents on the process variables was described by 2^nd order polynomial equations considering only the terms with significant effects. The optimisation problem was defined as the minimisation of the sum of moisture and soluble solids contents, each divided by the respective maximum value in the range tested. The best results were obtained using a 45 °Brix solution. The optimal temperature increased with decreasing product moisture content, varying from 33.3 °C to 42.1 °C when the samples were dehydrated during 20 and 240 minutes, respectively; the product moisture content in these conditions was decreased by 7 and 36%, respectively.