SENS: Semantic Synthetic Benchmarking Model for integrated supply chain simulation and analysis

TL;DR

SENS is a semantic, ontology-based benchmarking model that enables integrated simulation, analysis, and synthetic data generation for supply chains, improving resilience understanding and performance benchmarking.

Contribution

The paper introduces SENS, an ontology-driven knowledge graph and SENS-GEN, a configurable data generator for comprehensive supply chain analysis and benchmarking.

Findings

Enhanced simulation and analysis capabilities for supply chains.

Improved understanding of supply chain resilience during disruptions.

Synthetic data generation supports benchmarking across scenarios.

Abstract

Supply Chain (SC) modeling is essential to understand and influence SC behavior, especially for increasingly globalized and complex SCs. Existing models address various SC notions, e.g., processes, tiers and production, in an isolated manner limiting enriched analysis granted by integrated information systems. Moreover, the scarcity of real-world data prevents the benchmarking of the overall SC performance in different circumstances, especially wrt. resilience during disruption. We present SENS, an ontology-based Knowlegde-Graph (KG) equipped with SPARQL implementations of KPIs to incorporate an end-to-end perspective of the SC including standardized SCOR processes and metrics. Further, we propose SENS-GEN, a highly configurable data generator that leverages SENS to create synthetic semantic SC data under multiple scenario configurations for comprehensive analysis and benchmarking applications. The evaluation shows that the significantly improved simulation and analysis capabilities, enabled by SENS, facilitate grasping, controlling and ultimately enhancing SC behavior and increasing resilience in disruptive scenarios.