Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Proposing a Framework for Machine Learning Adoption on Legacy Systems

Ashiqur Rahman, Hamed Alhoori

TL;DR

The paper addresses the economic and operational barriers to adopting machine learning in safety-critical, legacy manufacturing environments. It proposes an API-centric framework that decouples the ML lifecycle from production, hosting heavy computations externally (cloud or on-premises) and delivering results through a browser-based, human-in-the-loop interface. A central API gateway coordinates secure, low-latency inference and lifecycle management, enabling zero-downtime model updates and scalable deployment without hardware upgrades. The framework is demonstrated as generalizable beyond non-destructive inspection, with strong implications for production quality, safety, and cost reduction through improved trust and agility in industrial workflows.

Abstract

The integration of machine learning (ML) is critical for industrial competitiveness, yet its adoption is frequently stalled by the prohibitive costs and operational disruptions of upgrading legacy systems. The financial and logistical overhead required to support the full ML lifecycle presents a formidable barrier to widespread implementation, particularly for small and medium-sized enterprises. This paper introduces a pragmatic, API-based framework designed to overcome these challenges by strategically decoupling the ML model lifecycle from the production environment. Our solution delivers the analytical power of ML to domain experts through a lightweight, browser-based interface, eliminating the need for local hardware upgrades and ensuring model maintenance can occur with zero production downtime. This human-in-the-loop approach empowers experts with interactive control over model parameters, fostering trust and facilitating seamless integration into existing workflows. By mitigating the primary financial and operational risks, this framework offers a scalable and accessible pathway to enhance production quality and safety, thereby strengthening the competitive advantage of the manufacturing sector.

Proposing a Framework for Machine Learning Adoption on Legacy Systems

TL;DR

The paper addresses the economic and operational barriers to adopting machine learning in safety-critical, legacy manufacturing environments. It proposes an API-centric framework that decouples the ML lifecycle from production, hosting heavy computations externally (cloud or on-premises) and delivering results through a browser-based, human-in-the-loop interface. A central API gateway coordinates secure, low-latency inference and lifecycle management, enabling zero-downtime model updates and scalable deployment without hardware upgrades. The framework is demonstrated as generalizable beyond non-destructive inspection, with strong implications for production quality, safety, and cost reduction through improved trust and agility in industrial workflows.

Abstract

The integration of machine learning (ML) is critical for industrial competitiveness, yet its adoption is frequently stalled by the prohibitive costs and operational disruptions of upgrading legacy systems. The financial and logistical overhead required to support the full ML lifecycle presents a formidable barrier to widespread implementation, particularly for small and medium-sized enterprises. This paper introduces a pragmatic, API-based framework designed to overcome these challenges by strategically decoupling the ML model lifecycle from the production environment. Our solution delivers the analytical power of ML to domain experts through a lightweight, browser-based interface, eliminating the need for local hardware upgrades and ensuring model maintenance can occur with zero production downtime. This human-in-the-loop approach empowers experts with interactive control over model parameters, fostering trust and facilitating seamless integration into existing workflows. By mitigating the primary financial and operational risks, this framework offers a scalable and accessible pathway to enhance production quality and safety, thereby strengthening the competitive advantage of the manufacturing sector.

Paper Structure

This paper contains 12 sections, 2 figures.

Table of Contents

  1. Introduction
  2. Related Works
  3. Use and Challenges of AI and ML in Production
  4. ML in the Industry
  5. Challenges for Legacy Systems
  6. Method
  7. ML Model Hosting and Management
  8. User-Centric Interface Design
  9. The API Gateway
  10. Discussion
  11. Limitations and Future Work
  12. Conclusion

Figures (2)

  • Figure 1: The components of the proposed method. Each component is surrounded by dotted lines. The one at the top represents the user interface on a legacy system, then the one in the middle represents the API, and the one at the bottom represents the ML model maintenance.
  • Figure 2: Prototype user interface for the framework. The white dots on the black scan images represent regions of interest identified by the model.