Browsing by Author "Diem, Michael"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Implementation of machine learning to improve the decision-making process of end-of-usage products in a circular economy
    (Stellenbosch : Stellenbosch University, 2020-03) Diem, Michael; Louw, Louis; Braun, Anja; Stellenbosch University. Faculty of Industrial Engineering. Dept. of Industrial Engineering.
    ENGLISH ABSTRACT: Rising consumption due to growing world population and increasing prosperity, combined with a linear economic system have led to a sharp increase in garbage production, general pollution of the environment and the threat of resource scarcity. At the same time, the perception of environmental protection becomes evident. The Circular Economy (CE) could reduce waste production and decouple economic growth from resource consumption, but most of the products currently in use are not designed for the recovery options of the CE. In addition, the decision-making process regarding following the steps of End-of-Usage (EoU) products has further weaknesses in terms of economic attractiveness for the participants, which leads to low return rates. This work proposes a model of the decision-making process for laptops, which is divided into two parts. In the first part, the condition of the product on component level is determined by the use of Machine Learning (ML). For this purpose stress factors are developed, which have an impact on the condition of the product. Furthermore, ways are elaborated to capture them, as the product is not physically present. A ML method is selected to process this information. A suitable software application is selected on the basis of defined criteria. In the second part, an economic and ecological evaluation is conducted based on the conditions delivered by the ML process. A possible purchase price is determined on the basis of the costs incurred and the expected selling price. In addition, the emissions saved as a result of the recovery are calculated. In order to demonstrate the potentials of the developed processes and thus validate them, comprehensive data is simulated and a prototype developed. The data is used to train the Artificial Neural Networks (ANNs) and as test cases. This work will contribute to carrying out more advanced decision-making and thereby increase the attractiveness, which should lead to higher return rates of EoU products.