Methodology for Monitoring Efficiency of Supply Chain Network Designs

Since performance cannot be gathered and assessed by a single indicator, this project develops an approach to effectively evaluate and monitor the realized efficiency of supply chain network designs based on multiple factors. More specifically, the methodology will focus on multi-factor, efficiency-based models (data envelopment analysis and related extensions) that compare the realized performance of a supply chain network against ideal targets. The methodology will assist in identifying any specific upward or downward trends in efficiency that may trigger the need for a network redesign to improve performance.

Most performance factors/metrics stand in relationship with other factors. Generally, these relationships are either conflicting or complementary; independence is the exception rather than the rule. Moreover, the factors may have unequal importance to decision makers.

To overcome these issues, the factors considered in the methodology and their relative importance will be identified through case studies, company interviews and extant literature in this domain. A multi-criteria relative prioritization analysis (analytic hierarchy process) will be utilized to identify the key success factors, relative importance (weights), and related targets that are expected. Factors such as supply chain costs, varying demand levels, inventory levels, customer service levels, quality rates, reliability scores, robustness to disruptions, and other critical measures identified through interviews and literature, can effectively be incorporated into the decision-making process.

The proposed approach will generate a supply chain efficiency score based on the aforementioned factors, overcoming the interdependency and varying importance issues, and will provide the decision maker with an index for monitoring performance over multiple time periods to determine if the supply chain meets the established goals. If specific downward trends or cycles in efficiency are observed, the model provides a trigger to assist managers in redesigning the network to counteract negative elements.