Synthetic Scenario Generation for Evaluation of Industry 4.0 Agents
For researchers developing industrial agents, this work provides a scalable method to generate realistic evaluation scenarios, but the extension is incremental as it builds on an existing benchmark.
The authors extend the AssetOpsBench industrial agent benchmark with a Smart Grid Transformer asset class and four diagnostic tools, and introduce ScenarioGeneratorAgent, a pipeline for synthetic scenario generation. Optimizations reduce runtime by 8× for 50 scenarios while maintaining quality (composite score 74.2 vs 73.8).
Industrial agent benchmarks require realistic evaluation scenarios that integrate telemetry, failure modes, maintenance records, and domain standards. However, existing benchmarks such as AssetOpsBench rely on manually authored scenarios and cover a limited set of asset classes. We extend AssetOpsBench with a Smart Grid Transformer asset class and four IEC-grounded diagnostic tools for health-index prediction, dissolved-gas analysis, winding-temperature assessment, and load-profile assessment. We further introduce ScenarioGeneratorAgent, a pipeline for synthetic industrial-agent scenario generation. The pipeline constructs evidence-grounded asset profiles, allocates coverage-aware scenario budgets across operational domains, and generates candidates through a hybrid validation-and-repair loop that enforces schema validity, tool reachability, physical plausibility, standards alignment, and deduplication. To improve scalability, we apply two-level caching, parallel focus-group generation, thread-pool offloading, batched LLM calls, and early rejection filtering. On Smart Grid Transformer scenario generation, these optimizations reduce end-to-end runtime by $8\times$ for 50 scenarios while preserving quality, achieving a composite quality score of $74.2 \pm 1.9$ compared with $73.8 \pm 3.0$ for the unoptimized baseline. These results show that standards-grounded synthetic scenario generation can efficiently expand industrial-agent benchmarks without sacrificing scenario quality.