Towards Automated Solution Recipe Generation for Industrial Asset Management with LLM

TL;DR

The paper addresses the challenge of scaling Industrial Asset Management (IAM) with CBM by automating the generation of solution recipes using Large Language Models (LLMs) and a KPI-centric taxonomy. It introduces AutoRecipe, a framework that combines taxonomy-guided prompting (KPITaxo2Prompt) with multiple PromptSequence Execution Engine strategies to produce artifacts (knowledge documents, configurations, data, code, and wrappers) deployable in environments like MLFlow. Key contributions include a taxonomy-based prompting pipeline, a structured artifact set for IAM, a Reference Generator Pipeline for web-grounded factual validation, and iterative information-discovery methods to improve knowledge extraction quality. The work demonstrates the potential to reduce domain-knowledge requirements and accelerate CBM solution development across ten asset classes, while highlighting ongoing challenges in ensuring factual accuracy and relevance and outlining directions for future automation and deployment testing in real client contexts.

Abstract

This study introduces a novel approach to Industrial Asset Management (IAM) by incorporating Conditional-Based Management (CBM) principles with the latest advancements in Large Language Models (LLMs). Our research introduces an automated model-building process, traditionally reliant on intensive collaboration between data scientists and domain experts. We present two primary innovations: a taxonomy-guided prompting generation that facilitates the automatic creation of AI solution recipes and a set of LLM pipelines designed to produce a solution recipe containing a set of artifacts composed of documents, sample data, and models for IAM. These pipelines, guided by standardized principles, enable the generation of initial solution templates for heterogeneous asset classes without direct human input, reducing reliance on extensive domain knowledge and enhancing automation. We evaluate our methodology by assessing asset health and sustainability across a spectrum of ten asset classes. Our findings illustrate the potential of LLMs and taxonomy-based LLM prompting pipelines in transforming asset management, offering a blueprint for subsequent research and development initiatives to be integrated into a rapid client solution.

Figures (16)

• Figure 1: Condition-Based Maintenance Solution Dev. Process
• Figure 1: KPIs for Industrial Asset
• Figure 2: KPI-centric Taxonomy for Asset Health Score
• Figure 3: Architecture Diagram for AutoRecipe Generation
• Figure 4: Iterative Flow for Quality Refinement