An Ontology for the Assessment of Procurement Risk Management

Today, a deep understanding of procurement risk management is of great importance for leading focal firms. This helps to avoid disruption along the value chain, to leverage the strength of their supply chains and improve their profit margins. Firms that use an advanced procurement performance and risk indicator system can evade significant upstream damages from procurement risks, react timely instead of delayed and can enhance expressively the robustness and, therefore, the profitability of their supply chain in the future.In this study, after having extensively overviewed literature about business risks today, supply chain risks, procurement risks, supply risk management processes and several interviews with supply risk experts, a prototype of the ontology and risk assessment based on the findings has been realized. Such a prototype, at hand, consists of a knowledge conceptualization about procurement risk management and a way to assess the risks based on information of differing natures such as events and company characteristics. This knowledge is represented in an ontology named APPRIS ontology which is about an enterprise and its relation to its suppliers. In this sense, the APPRIS ontology needs concepts representing relevant aspects of an enterprise. Thus, the approach in this work is to use an enterprise ontology and extend it. In parallel, a procurement risk assessment is proposed involving rules that are executed by a reasoner. However, in this work the risk assessment is also considered in terms of visibility, in the sense that allowing the risk manager to retrieve specific information about the procurement company enables him to assess this information. By means of a concrete example, it is shown how the procurement risk assessment may be executed.In order to reduce implementation efforts, the Protégé platform, a scalable and integrated framework for ontological engineering, has been used to construct the APPRIS ontology. The constructed ontology has been represented in Web Ontology Language format, which makes it more machine understandable.In the near future, knowledge of this ontology could be exploited through a proper user interface supporting the risk manager in his/her decisions making