Bayesian Machine Learning Approach for Evaluating the Effectiveness of an Order Fulfilment Reengineering Project in the Downstream Oil and Gas Supply Chain

Coşkun S. S.

in: Artificial Intelligence for Business An Implementation Guide Containing Practical and Industry-Specific Case Studies, Hemachandran K Raul V. Rodriguez, Editor, CRC, New York , New York, pp.0-1, 2023

  • Publication Type: Book Chapter / Chapter Research Book
  • Publication Date: 2023
  • Publisher: CRC, New York 
  • City: New York
  • Page Numbers: pp.0-1
  • Editors: Hemachandran K Raul V. Rodriguez, Editor
  • Istanbul University Affiliated: Yes


The oil and gas trade involves highly complex procedures and relations through the downstream supply chains. To ensure the real-time data flow, distributors, vendors, and customers need to be well integrated. Coordination defects and disruptions occurring in data flow lead to direct and indirect costs while reducing the supply chain performance. The fourth industrial revolution (industry 4.0) promises new technologies to diminish complexity and improve efficiency. However, it becomes difficult to evaluate the effectiveness of these implementations when the uncertainties arise regarding the performance outcomes. This study introduces a business process reengineering (BPR) project that aims to improve the order fulfillment performance of a large-scale oil and gas distribution network. We introduce the shadow sell concept as a novel performance metric and use it to evaluate the project success. In the case study, we assume that the daily count data of the shadow sales convey a Poisson process. An eight-step Bayesian methodology describes the performance density and explores a potential breakpoint across the project timeline. The results provide insights into a successful real-world industry 4.0 implementation and emphasize the merits of the Bayesian methodology to overcome uncertainties when deciding on the success of large-scale projects.