Consider conveying systems that combine material streams from multiple sources and then distribute those same materials to multiple workstations or machines, often with diversions between each main section to ensure overall material balance. Although the OT control system on each of these major conveyance sections is well equipped to tactically adjust speed within the section to ensure a constant flow of materials, it lacks awareness of the computer data that orchestrates the influx of materials in each section, thus staffing the workforce. or loading of machines fed by the conveyor. By combining this IT data with OT data regarding the current inventory level of each main convenience section, AI can prescribe efficient operation of workarounds between main sections to alleviate bottlenecks, create more consistent outputs and adjust for expected input imbalances.
SI: What advice can you offer businesses to manage the end-to-end ML model operations cycle (from model creation to maintenance)? Is it as tricky as it sounds?
MN: The growing prevalence of AI/ML technologies and the increasing complexity of integrating machine learning models into critical manufacturing processes have prompted the development of MLOps to effectively manage the entire lifecycle of these models, from their creation to their maintenance.
Capitalize on the strong foundations already established through experience within OT systems. OT systems in factories have often matured over long periods of time and have to a large extent organized the association and contextualization of information. Ensuring that the OT system’s I/O architecture is mapped to an ML model at build time initiates the path to value. Aligning the OT context with an ML model will enable economies to expand and maintain these capabilities post-deployment. This foundational practice will act as a catalyst, accelerating AI/ML initiatives.
Integrating MLOps methodology as a natural extension of well-established OT practices to monitor and maintain model and machine performance, aligning with well-established change management and standard operating procedures, is essential to the adoption of these practices. Scaling these capabilities should align with business value, encompassing the number of ML models deployed in the production environment, measuring their impact on business ROI, and evaluating the efforts of current maintenance.
SI: What are some tips for launching ML solutions with minimal effort from data scientists using pre-built ML libraries and BYOM models?
MN: Minimizing the efforts of data scientists involves important organizational and technological considerations. Organizationally, the most important aspect is to integrate process experts or operators into the overall modeling process, from creation to deployment. Often, for an ML solution to be used in a factory, the solution prescribed by the AI must be validated or approved by an expert in the operation of a process. Ensuring that a model has a chance of being deployed often involves understanding historical operating principles and ensuring that AI results can be explained in that context. Additionally, appropriate organizational involvement can ensure that ML activities are precisely targeted to prioritize decisions that generate tangible value while focusing on or ignoring less impactful process aspects.
Imagine renovating your home with guidance from a contractor who involves you from the initial planning stages to the finishing touches. Similarly, launching ML solutions with minimal effort involves collaboration between data scientists and process experts. This ensures that AI solutions integrate into operations.
When it comes to technologies, adopting platforms that seamlessly accommodate bring-your-own-models (BYOM) greatly simplifies deployment, especially OT models that have been developed and matured over time . These platforms provide the ability to fine-tune model parameters to match specific equipment and product characteristics, such as temperatures, pressures, motor speeds, etc. It is important to select the technology stack that provides easy-to-configure standard connectivity protocols to communicate with enterprise controllers and IT systems, the flexibility to package key functionality as microservices to decouple resilient components, and Built-in features to establish workflow pipelines streamlining deployment and maintenance at scale.
SI: What will this trend look like in five years? Twenty years?
MN: According to the latest market reports, the global industrial AI market was worth $16.9 billion in 2020 and is expected to reach $102.2 billion by 2026. As we consider the trajectory of this trend , it is becoming increasingly clear that AI/ML will revolutionize industrial manufacturing processes and drive unprecedented levels of business outcomes over the next five and ten years.
- Wide adoption of autonomous manufacturing: AI/ML technologies will drive widespread adoption of autonomous capabilities, where every automation controller will be equipped with intelligent AI agents to achieve higher levels of control and optimize manufacturing processes.
- Shaping the Next Generation Workforce: AI-enabled autonomous capabilities will become essential for businesses to retain decades of tribal knowledge from the retired workforce and shape the workforce of tomorrow.
- Empowering the 2.0 operator: AI will play a vital role in elevating the role of an operator from repetitive manipulations to managing machine performance.
- Improving quality control with machine vision: Combining advanced closed-loop control strategies with machine vision feedback at every critical step in the manufacturing process allows controllers to take automatic corrective action to minimize defects in real time, leading to higher levels of manufacturing quality. product that were previously inaccessible.
- Advancing vision-guided robotics: Advanced perception capabilities with stationary robotic systems (such as articulated robotic arms, delta robots, and gantry systems) and mobile robotic platforms (such as AGVs and AMRs) allow them to navigate environments complex applications, determine optimal paths for routing, handle delicate materials, and perform complex tasks. tasks accurately.
- Adoption of Generative AI to Accelerate Time to Value: This innovative form of AI will revolutionize manufacturing by generating synthetic data for data augmentation and training robust AI/ML models at a rapid pace, driving tremendous innovation.
- Reinforcement learning (RL) with human feedback: These capabilities combine the decision-making capabilities of RL algorithms with the expertise and intuition of human operators for a new era of intelligent systems that continuously learn and evolve under human guidance.
- Energy optimization: AI will play a central role in efficiently optimizing energy consumption and reducing costs, while maximizing the throughput and quality of energy-intensive processes.
The successful development and adoption of AI systems in manufacturing will depend on in-depth industry expertise and application-specific knowledge required. Companies that possess this expertise, combined with specialist knowledge in the application of AI technology, will become pioneers in innovation, unlocking the full potential of AI systems and delivering transformative results in operations Manufacturing.
How do these capabilities advance your manufacturing processes and help you develop strategic advantages over your competitors?
