Process optimisation and development of non-invasive monitoring approaches for the drying of apples

Drying is of great importance particularly in the field of preserving agricultural products. It plays a very critical role by delivering nutrient rich products with a long shelf-life as well as help to combat food loss. Drying processing techniques have changed and improved significantly over time, helping to ensure food abundance in both aesthetical and nutritional regards. Convective hot air drying is one of the most common technique practiced by 85 % of industries. It can be easily applied in the remote areas of developing countries, such as in the Jumla and Mustang districts of Nepal where apple losses can be reduced. Additionally, drying of apples could help to overcome several constraints, such as poor transportation infrastructure, which causes apple to spoil before they reach the market, packaging and storage challenges; seasonality of apples; weather changes; and perishability. Moreover, dried apples as an example can improve the nutritional quality of diets of infants and young children that is of utmost importance in the quest to reduce undernutrition. The production of high-quality dried apples/apple products could be highly relevant to small and medium scale farmers looking for new markets. Therefore, the thesis seeks to demonstrate how this scientific knowledge contained herein can be useful in the development of high-quality dried apple using low-cost convective drying, focusing on decentralised applications for farmers in Jumla and Mustang district in Nepal. The novelty of this research is represented by the framework in the application of hyperspectral imaging to estimate quality metrics such as moisture content, shrinkage during the apple drying process independent of applied treatments and different drying settings (i.e. drying temperature, slice thickness and cultivars) but also evaluation on in combination with the applied pre-treatment. In the practical application of visualising the apple slices during drying, developed algorithms proved to have a potential feasibility in investigating spatial and spectral features of the quality attributes irrespective of all applied pre-treatments, drying settings, cultivars, slice thickness. These findings could find potential utilisation in the commercial drying of apple produce detecting in real time and controlling the characteristics of apple slices in- online process system. Visualising the colour changes and targeted levels of moisture content through online monitoring could inform when to end or stop the drying process. This research has demonstrated that non-invasive optical systems and the related algorithms for detection of quality features in apple slices are feasible and their approaches are essential for the optimisation of processes and the development of new strategies in the long term. Implementation of these strategies could be helpful in allowing more substantial variations of quality of supplied raw material in the establishment of a control process driven by product quality and multivariate control systems. The assessment of food quality in a fast and reliable approach is a vital factor in all aspects of business from the farmer to the industry and the consumers. Therefore, this approach could help to significantly reduce on-farm postharvest losses which are caused by several factors such as rigorous pre-sorting and improper processing.