A DATA-DRIVEN RISK CASCADING EFFECT EVALUATION FOR SUPPLY AND PROCUREMENT IN THE CONSTRUCTION INDUSTRY

In the procurement and supply of a large commodity trading company in the construction industry, there are interrelated risk sources in the cooperative work between the project department, commodity trading company and supplier. These interrelated risk sources adversely affect the realization of procurement, supply and collaborative goals and have cascading effects on different collaborative networks in the supply chain. To control the risk in the process, this paper proposes a methodology for analysing and modelling the interrelationships and for evaluating the risk levels and risk cascading effect. First, key processes and risk knowledge are extracted through an analysis of enterprise documents and the deconstruction and organization of data in reports and files related to the procurement and supply processes. Second, a data-driven risk (DDR) ontology is constructed based on the danger/risk/consequence (DRC) chain and scenario-risk-accident chain (SRAC) ontology. Third, data-driven risk-level-inference rules (RLIRs) are formulated by analysing the associations between knowledge, including functions, risk, information, and attributes. Finally, an example is used to verify the model and reasoning of the process. This methodology has considerable implications for risk control in the construction industry.