Production planning: it works better with AI

The latest trends in factory optimization

There's no doubt about it: artificial intelligence has arrived in industrial production – and that's a good thing. Anyone who wants to build a solid foundation for future competition is well advised to use AI-supported solutions. This recommendation applies all the more as the user does not necessarily have to have or learn their own AI expertise. For example, they can also use software with “built-in” artificial intelligence and benefit from the many advantages without having to be an expert themselves.

By: Gerald Scheffels for ifm

A good start: production planning

When it comes to the question of where to start such a “trial run” with AI, production planning is an obvious choice for a manufacturing company. In many companies, there is a wide range of different orders that pass-through production at different stations and take different routes.

Quite “classic” solutions, e.g. from the SAP/supply chain environment, can already ensure order, shorten lead times and take bottlenecks into account or, better yet, eliminate them. But with artificial intelligence as an acceleration and improvement factor, much more can be achieved here. The potential is huge, and the responsible operations managers, production planners, intralogistics experts and supply chain managers can and should try it out.

Production planning across multiple work centres – with Factory Optimization Excellence (FOX)

The question of how the effects of AI in production planning can be tested in concrete terms can be answered clearly and simply: by using the optimizer “Factory Optimization Excellence”, or FOX for short, from ifm. The software improves production planning using AI and opens up the possibility of broadening production planning by enabling valid and realistic planning across multiple work centres.

Pegging: It's the relationship that matter

FOX was developed by the SCM branch of ifm in cooperation with ifm statmath gmbh. statmath contributed the algorithms and AI functions, while the ifm supply chain team provided the platform and practical expertise for IT-supported production planning and optimization.

The “game changer” of the jointly developed solution is the pegging function, with which production is checked as a linked process with all predecessor and successor relationships – as a whole and not step by step. This function is stored in the SAP core for FOX users. It utilises the existing data and – unlike previous software systems – the familiar terminology used in the SAP world. The user “only” has to feed the FOX algorithm with data, including scheduling, delivery strategy and adherence to delivery dates, shift times and the required machines and tools. This is how the AI works and delivers a valid result – and that is only a slight simplification.

Just the right AI

Ultimately, what is needed here is not rocket science and it is not a more or less creative fireworks display as with some “use cases” from ChatGPT and other tools that use generative AI. The AI does not sift through the world's knowledge here, nor does it paint pictures. Rather, it combines existing data according to predefined algorithms and thus enables the optimization of the production sequence. This is not referred to as generative (i.e. creative) AI, but as “weak AI”, which is not weak, but does exactly what it is supposed to do: select the best combination from a multitude of combinations and, thanks to its intelligence, becomes better and better at this clearly defined task.

Optimum result in minutes

How long does such a calculation take? You have to plan a few minutes – then you have a first solution.

Whether faster or slower, he can trust the result 100%. And he can profit from an additional benefit: The production solution that is optimized in terms of time is usually also the one that consumes less energy and resources. The use of AI therefore also contributes to sustainability.

The development continues

Previous users of FOX are convinced of the benefits of AI: this is not (or not only) hype, but real added value. This is another reason why ifm and statmath are driving the development of AI-supported production planning forward.