Abstract:
One standard in measuring supply chain management success is that established by the SCOR model. The SCOR model was created by a management consulting firm of the Supply Chain Council which relies on specific performance measures that are related to the five-core process building blocks: Plan, Source, Make, Deliver, and Return with a sixth block of “Enable” added later. With its origins in Western/Developed Countries, there is some question about the applicability of the same metric system in Low- and Middle-Income Countries. This thesis relies on a survey methodology to explore the extent to which companies across multiple industries are measuring the SCOR Level 1 and Level 2 metrics in Lebanon and the MENA region. The results of the survey are analyzed via two machine learning techniques – an unsupervised clustering technique (kMeans) to identify companies with similar behavior relative to the SCOR metrics and a supervised learning technique (Classification Trees) to ascertain which company demographics (ie industry, age, size, age of employees, and SCOR familiarity) dictate cluster membership.
