A Two-stage Discriminating Framework for Making Supply Chain Operation Decisions under Uncertainties

H. Gu and R. Gang

Abstract
This paper addresses the problem of making supply chain operation decisions for refineries under two types of uncertainties: demand uncertainty and incomplete information shared with suppliers and transport companies. Most of the literature only focus on one uncertainty or treat more uncertainties identically. However, we note that refineries have more power to control uncertainties in procurement and transportation than in demand in the real world. Thus, a two-stage framework for dealing with the considered problem is proposed, which discriminates the two types of uncertainties for decision-making. This framework introduces a new and complete workflow to decision makers. The trade-off between economy and expected value of customer satisfaction level (CSL) under uncertainties is realized by managing the safety stock levels. At the first stage, a new simulation-based optimization approach is introduced to cope with demand uncertainty, where an outer loop for large adjustment and an inner loop for tiny adjustment are integrated. Incomplete information will be revealed gradually and overcome by negotiation loops in the second stage. The target CSL can be achieved or approached when executing the final decisions under considered uncertainties. In addition, a combination of hierarchical supply-chain optimization models and if-then rules based simulator are described for this framework. The performances of this two-stage framework are proved by the case studies.

This work is licensed under a Creative Commons Attribution 4.0 International License

Keywords
Supply chain, simulation based optimization, uncertainty, customer satisfaction level, safety stock level