The rise of intelligence: How AI is redefining the manufacturing execution system

Artificial intelligence is now regarded as a general-purpose technology, such as electricity or the internal combustion engine. It is here to stay — and it will also change production sustainably. A Forbes article describes why manufacturing execution systems are not the right basis for this and what characterizes the alternatives. In this article, the argument is supplemented with solutions with AI support.

Manufacturing execution systems are a product of the 80s and 90s that have stuck surprisingly well in the thinking of even the current generation of management. You understandably focus on what you know from the past.

Why implementing a manufacturing execution system in the 2020s is no longer a good idea has now even Forbes magazine addressed thematically. Even though the article is not easy to read, it is very valuable for all managers who are in the process of selecting shop floor software, i.e. who are making a strategic, long-term decision.

In essence, the article says that a modern digital strategy cannot be built on outdated technology and architecture. The future of software support in production consists of composable, app-based solutions.

We passionately share this position at oee.ai and want to add a perspective to the Forbes article: Artificial Intelligence.

If you ask an AI which top 10 use cases their technology can use to support the shop floor of the future, you get the following list — the first words of each bullet line are sufficient for the quick reader:

• Predictive maintenance: AI algorithms can analyze sensor data from machines to predict when devices are likely to fail. By identifying issues before they occur, maintenance can be proactively planned, minimizing downtime and preventing costly outages.
• Optimizing production planning: AI can analyze historical production data, market demand, and resource availability to optimize production plans. This ensures that resources are used efficiently, reduces idle time, and maximizes performance.
• Improving quality control: AI-powered computer vision systems can check products in real time as they move along the production line. This allows faults to be identified more precisely and faster than through manual inspection, reducing waste and improving product quality.

• Energy management: AI algorithms can analyze energy consumption patterns within the factory and identify optimization opportunities. This may include adjusting equipment settings, planning off-peak production, or implementing energy-efficient technologies to reduce costs and environmental impact.
• Inventory optimization: AI can analyze demand forecasts, production plans, and supplier lead times to optimize inventory levels. By ensuring that the right amount of inventory is available at the right time, companies can reduce inventory costs while avoiding shortages and reorders.
• Process automation: AI-powered robotic process automation (RPA) can automate repetitive and everyday tasks on the factory floor, such as data entry, assembly, and packaging. This allows human workers to focus on more complex and value-adding tasks, improving overall productivity.
• Workforce Management: AI algorithms can analyze employee planning data, skills, and key performance indicators to optimize staffing. By ensuring that the right people are tasked with the right tasks at the right time, companies can improve efficiency and work ethic.
• Demand forecast: AI algorithms can analyze historical sales data, market trends, and external factors (such as weather or economic indicators) to more accurately predict future demand. This helps manufacturers optimize production plans and inventory levels to meet customer demand while minimizing excess inventory or shortages.
• Simulation and modeling: AI-driven simulation and modeling tools can create virtual replicas of the factory environment to test various scenarios and optimize processes. Manufacturers can use these simulations to identify bottlenecks, experiment with process improvements, and make data-driven decisions to improve productivity.
• Manufacturing Analytics: AI can analyze large amounts of production data to identify opportunities for process optimization and innovation. By continuously monitoring and analyzing key performance indicators, companies can identify trends, patterns, and anomalies that can lead to productivity gains over time.

Development in AI fields of large language models has only just begun in November 2023 with the release of ChatGPT and development — including and especially among open source models — is rolling over. In spring 2024, Open Source Time Series Foundational Models will be published, which are specially trained to analyze time series data and thus open up even more possibilities for the above use cases — what a time to be alive.

And now, as a decision maker, you have to ask yourself whether you believe that A software provider based on its old technology and architecture (see Forbes above) can offer the best — or even just a good and future-oriented — solution in all these fields.

As already stated in the Forbes article, the alternative architecture consists of a selection of specifically the best Manufacturing Execution Support Apps, which comprise what has been called a Manufacturing Execution System in the past.

This means that alternative providers are those who dig deep into one of the solution spaces and drive progress there — including with the help of current and future AI opportunities. This is what oee.ai does in the last enumeration field — “Manufacturing Analytics.”

In the target architecture, your software landscape then has an app for manufacturing analytics, an app for predictive maintenance, an app for inventory optimization, etc. — each from a provider who really focuses on the solution space and its current and future options.

Transformer technology, on which a large part of the AI solutions currently under discussion is based, is still very young. Providers such as OpenAI, Google, Meta and Mistral have made huge progress. If you would like to discuss how this technology is incorporated into the topic of “Manufacturing Analytics,” feel free to contact us at info@oee.ai.

Author: Linus Steinbeck