An expert system for transport mode selection for US medical supply chains based on data mining

Wang, Chenyu (2020) An expert system for transport mode selection for US medical supply chains based on data mining. [Dissertation (University of Nottingham only)]

[img] PDF - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (1MB)


It is easy to notice that the exposure of polluted environments, chronic diseases, and rising injury accidents contribute to the disease diversity and frequency among people, and the demands of medical supplies keep increasing around the world. Therefore, it is necessary to standardize and improve the medical supply chain system continuously. The transport mode selection, as a critical aspect in the medical supply chain system, has been concerns of many. There is evidence that the transport mode often influences the quality of products in supply chains, and the limited transport selection can lead to some potential risks. To guarantee serving customers with the high quality of medical supplies, the scientific transport mode selection system is necessary to be developed to contribute to providing decision-makers reasonable transport mode recommendations.

In this paper, we propose a transport mode selection expert system to help managers select the best transport mode based on the analysis of the known features data of the US medical supply chains. The expert system contains two main parts, the Neural Network model and the association rules learning model. Initially, we collect the full US medical supply chains records in 2012 and perform a series of data preprocessing toward these raw data to change them into the input data, which can be processed in our model. Then, we develop a Neural Network model to recommend transport mode for decision-makers based on the known features of medical supply chains as input data. The Neural Network model perform well on the test set, with 0.92 accuracy, which means it has a high level of guiding significance and can be used in the real decision process. In addition, we also conduct the association rules learning based on the known features, and generate several interesting rules as supplements of decision making, which can contribute to making prompt transport mode decision. Compare the transport recommend results between these two techniques, and the managers can have sufficient decision support evidence to help them determine the best transport mode, which can contribute to ensure the high quality of medical supplies and reduce the potential risks in the supply chains.

Item Type: Dissertation (University of Nottingham only)
Depositing User: Wang, Chenyu
Date Deposited: 21 Dec 2022 14:41
Last Modified: 21 Dec 2022 14:41

Actions (Archive Staff Only)

Edit View Edit View