CausalPulse: An Industrial-Grade Neurosymbolic Multi-Agent Copilot for Causal Diagnostics in Smart Manufacturing

Chathurangi Shyalika; Utkarshani Jaimini; Cory Henson; Amit Sheth

cs.AIMar 31, 2026

CausalPulse: An Industrial-Grade Neurosymbolic Multi-Agent Copilot for Causal Diagnostics in Smart Manufacturing

Authors: Chathurangi Shyalika, Utkarshani Jaimini, Cory Henson, Amit Sheth

Organizations: Artificial Intelligence Institute, University of South Carolina · Stride Lab, University of Michigan, Dearborn · Bosch Center for Artificial Intelligence, Pittsburgh

Abstract

Modern manufacturing environments demand real-time, trustworthy, and interpretable root-cause insights to sustain productivity and quality. Traditional analytics pipelines often treat anomaly detection, causal inference, and root-cause analysis as isolated stages, limiting scalability and explainability. In this work, we present CausalPulse, an industry-grade multi-agent copilot that automates causal diagnostics in smart manufacturing. It unifies anomaly detection, causal discovery, and reasoning through a neurosymbolic architecture built on standardized agentic protocols. CausalPulse is being deployed in a Robert Bosch manufacturing plant, integrating seamlessly with existing monitoring workflows and supporting real-time operation at production scale. Evaluations on both public (Future Factories) and proprietary (Planar Sensor Element) datasets show high reliability, achieving overall success rates of 98.0% and 98.73%. Per-criterion success rates reached 98.75% for planning and tool use, 97.3% for self-reflection, and 99.2% for collaboration. Runtime experiments report end-to-end latency of 50-60s per diagnostic workflow with near-linear scalability (R^2=0.97), confirming real-time readiness. Comparison with existing industrial copilots highlights distinct advantages in modularity, extensibility, and deployment maturity. These results demonstrate how CausalPulse's modular, human-in-the-loop design enables reliable, interpretable, and production-ready automation for next-generation manufacturing.

Figures & tables

Figure 1: High-level overview of CausalPulse system architecture and components.

Figure 2: CausalPulse four-layer framework: user-facing, agent, utility, and data, connected through standardized protocols (MCP, A2A, and LangGraph).

Panel of Figure 3 — Figure 3: Illustration of domain-specific rules for a grinding process. Control parameters (rectangles) may cause observation parameters (circles) at the same or later grinding stage (Rules R1, R2, R5), observation parameters may only cause later observations (Rules R3, R4), and edges violating type or temporal order are explicitly disallowed (dashed red arrows).

Figure 4: Agentic task orchestration and workflow execution in CausalPulse: (a) Query interpretation and planning by the LangGraph-based workflow engine, (b-I) Agent card registration and central server coordination, (b-II) Agent interactions and data flow during execution, and (b-III) Postprocess chaining with the Recommender Agent.

Figure 5: Root cause analysis results: (a) ranked root-cause paths ordered by causal significance for the anomaly in E_8 in the PSE dataset (refer to ”Datasets Descriptions” Section), and (b) visual representation of the corresponding causal graph, highlighting identified root causes and propagation paths. The legend on the right lists variable names and their hierarchical levels, where levels represent the sequential order of grinding steps in the PSE use case.

Panel of Figure 6 — Figure 6: Canonical agent pipeline in CausalPulse. Predefined workflows correspond to subpaths of this pipeline with optional reuse of cached intermediate results.

ID	Workflow path (agents in order)
W1	Preprocessing $\rightarrow$ Recommender
W2	Preprocessing $\rightarrow$ Background Info $\rightarrow$ Recommender
W3	Background Info $\rightarrow$ Recommender
W4	Preprocessing ∗ $\rightarrow$ Anomaly Detection
	$\rightarrow$ Recommender
W5	Preprocessing ∗ $\rightarrow$ Background Info $\rightarrow$

Table 1: Predefined workflows in CausalPulse as paths through the canonical agent pipeline in Figure 6 . Steps marked ∗ may reuse cached results if available.

Dataset	Stage	Criteria	No.of Queries	Expected (steps/agents)	Predicted (steps/agents)	Successful attempts	Failed attempts	Criterion success (%)	Aggregate criterion success (%)
PSE	Preprocessing	Planning	20	3	3	20	0	100.0	98.73
		Tool use	20	2	2	20	0	100.0	98.73
		Self-reflection	20	3	2	19	1	95.0	98.73
		Multi-agent collab.	20	3	3	19	1	95.0	98.73
	Anomaly detection	Planning	50	3	3	50	0	100.0	98.73
		Tool use	50	3	3	50	0	100.0	98.73

Table 2: Criterion success and aggregated criterion success per stage across datasets, workflow stages, and agentic criteria

Panel of Figure 7 — Figure 7: Aggregated success rate per agentic criterion for PSE and FF datasets.

Anom.	Ground-truth causes	PRORCA top predictions	Hits@k	Prec.	Rec.	F1
F4	{ SetAngle_3 , GrindDepth_3 , GrindDepth_4 }	{ SetAngle_4 , GrindDepth_4 }	hits@1 = 0, hits@2 = 1	0.50	0.33	0.40
E12	{ GrindDepth_E_9 , GrindDepth_E_12 }	{ SetAngle_3 , GrindDepth_E_12 , GrindDepth_E_9 , SetAngle_4 , SetAngle_2 }	hits@1 = 0, hits@2 = 1, hits@3 = 1, hits@4 = 1, hits@5 = 1	0.40	1.00	0.57
E3	{ SetAngle_3 , GrindDepth_3 }	{ GrindDepth_3 , SetAngle_3 , SetAngle_4 , GrindDepth_4 }	hits@1 = 1, hits@2 = 1, hits@3 = 1, hits@4 = 1	0.50	1.00	0.67

Table 3: Representative PRORCA root-cause attribution outcomes on the PSE dataset, illustrating (top) a partial match where only one of three ground-truth causes appears in the top-2 positions, (middle) a high-recall but noisy case where all ground-truth causes are recovered along with several false positives, and (bottom) a strong match where both ground-truth causes are ranked in the top-2 positions.

	UCSD	MATMCD	AssetOps	SmartPilot	CausalPulse
1. System type	Single agent	Multi-agent discovery	Multi-agent simulation	Multi-agent copilot	Industry-grade multi-agent copilot
2. End-to-end
causal diagnostics
pipeline	✗	✗	✗	✗	✓
3. Dynamic
multi-agent

Table 4: Summary comparison of CausalPulse with representative industrial copilots.

Figure 8: End-to-end deployment pipeline of CausalPulse

Smart Manufacturing

See all 10 papers

Agentic Root Cause Analysis through Evidence-Grounded Reasoning
Jul 24, 2026Amaury Wei, Olga FinkIndustrial Anomaly Detection Agentic Reasoning
EPFL - IMOS Laboratory CH-1015 Lausanne Switzerland
ORCA: An End-to-End Interactive Copilot for Optimized Root Cause Analysis
May 26, 2026Phi Nguyen Xuan, Nicholas Tagliapietra, Lavdim Halilaj +2AI-Assisted Decision Making Causal Discovery
Bosch Global Software Technologies Company Limited, Vietnam · Robert Bosch GmbH, Germany · Computer Science Department, TU Darmstadt, Germany +2
LLM-Driven Multi-Agent Control for Skill-Based Smart Manufacturing
Oct 1, 2026Kay Köhle, Darko Anicic, Thomas A. Runkler +1Manufacturing Multi-Agent LLM Systems
Technical University of Munich, Munich, Germany · Siemens AG, Foundational Technologies, Munich/Garching, Germany · Siemens AG, Digital Industries, Fürth, Germany

Jul 24, 2026J/K move · Enter open · S save

Agentic Root Cause Analysis through Evidence-Grounded Reasoning

Amaury Wei, Olga Fink

EPFL - IMOS Laboratory CH-1015 Lausanne Switzerland

Industrial Anomaly Detection Agentic Reasoning Root Cause Analysis

Open paper AskPDF arXiv

Diagnosing the root cause of anomalies is essential for safe industrial operation. Despite extensive sensor instrumentation, formulating hypotheses and gathering evidence remains a manual process, creating a major operational bottleneck. While existing data-driven approaches aim to automate this, two critical limitations restrict their deployment: their operate as black boxes unable to justify their diagnosis, and they require scarce labeled examples of faulty operation. To address this gap, we introduce AgentRCA, a zero-shot agentic framework for evidence-grounded root cause analysis. Rather than learning fault-specific mappings, AgentRCA performs inference-time reasoning by combining a data-driven digital twin (modeling normal system dynamics) with a tool-augmented large language model. The agent iteratively gathers statistical evidence, evaluates competing hypotheses, and identifies the physical fault that best explains the observed behavior. Evaluated on a real-world multiphase-flow facility and a large-scale chemical plant, AgentRCA achieves diagnostic performance competitive with fully supervised baselines without relying on fault-specific training. Crucially, it produces transparent reasoning traces that explicitly link observed symptoms to their underlying physical causes. These results establish autonomous hypothesis-driven reasoning as a practical foundation for scalable industrial root cause analysis.

Abstract

Smart Manufacturing

Explore similar work

Agentic Root Cause Analysis through Evidence-Grounded Reasoning

ORCA: An End-to-End Interactive Copilot for Optimized Root Cause Analysis

LLM-Driven Multi-Agent Control for Skill-Based Smart Manufacturing