loading the future of Autonomous Mobile Robots


Reach out and we will get back to you!

    Modular Production: a Story About arculus Founding Vision

    Modular Production has recently made the headlines after Audi made public its plans to revolutionize assembly lines in manufacturing. As the founding vision of arculus, it’s time to shed light on this futuristic manufacturing technique. For that, we’ve invited Darshan Patel, Software Developer.

    First things first: what is Modular Production?

    You may have heard of Modular Production, a concept that promises a new take on manufacturing. Although previously an idea existing only in engineers’ futuristic dreams, it is actually about to become a reality, thanks to the ever-evolving AMRs. While this modern approach promises a lot, it does have to compete against the universally known – and widely proven – assembly lines.

    Established more than 100 years ago by Ford Motor Company, the linear production model offered the perfect solution for a market in need of producing uniform products in high quantities. In this configuration, assembling workstations are positioned along a linear path linked by a fixed transportation system.

    Nowadays, however, with consumer needs becoming more complex, conventional assembly lines are increasingly struggling with the market’s new dynamic. That’s precisely where Modular Production comes in: as an adaption of traditional manufacturing techniques that “ensures both an efficient large-scale assembly of mixed-model production programs and the needed level of flexibility and changeability to react to variations, disturbances and changes.” 1

    The Journey

    Replacing a century-old system is challenging and requires a considerable amount of tools. As Darshan Patel (Software Developer) explains, “in a linear assembly line, you might have the fixed transportation system, usually a conveyor belt. But in modular systems, we have Autonomous Mobile Robots (AMRs), like the arculee, that transport parts and products from station to station to push the assembly forward”.

    The Prequel

    But implementing an optimal modular assembly system requires much more than working robots. As a system that understands manufacturing as a holistic process, it requires a high volume of input data for the planning, such as1:

    • Product portfolio (volume and product mix)
    • Assembly steps per order/product variant
    • Standard assembly time for each step
    • Needed components and equipment for each assembly step
    • Work area at the product for each assembly step
    • Group of assembly steps resulting in one assembly task
    • Assembly priority chart (APC) on the level of assembly tasks for each product

    With these answers at hand, the developers can feed the data into software that uses genetic algorithms to provide a set of possible layout plans for the manufacturing facility. For that, it considers variables like codependency relationships between assembly tasks, field restrictions, and fleet size. “The system then shows us, for example, that station x has a heavy workload and should be split into two. And then also, because there is a huge load density in this specific point, we need a better driving network there to avoid bottlenecking.”, explains Darshan.

    Modular Production assembly chart after using the planning software
    Chart showing the modular assembly system after using the planning software 1

    Crossing the Threshold

    Only with the layout planning at hand can the Modular Production system be put to work. “It’s like if Modular Production were a movie, it would be parts two and three, and the planning would be the prequel”, explains Darshan.

    The holistic manufacturing model itself consists of three different components:

    • Main Modular Production system → controls the assembly of any product in the pipeline. It knows how, when, and where to do any task within the production process;
    • Logistics system → fetches the necessary parts from the storage and brings them to the proper assembly station, at the right time and in the correct order;
    • Worker Management System → assigns the workforce to certain stations according to the production demand, e.g. more workers in stations with a higher volume of goods.

    The cooperation between the three systems thrives through simultaneous and mutual optimization. “It’s a delicate dance between these components, which work together to improve the production rate of the whole system”, describes Darshan. “Increasing the speed of transporting components between stations would not work without the simultaneous delivery of equipment to assemble them. And even if you have the right components and pieces of equipment, you still need a suitable amount of workers”, he continues.

    The Ordeal

    When reading the description of such a well-coordinated system, one question does come to mind: does the whole thing actually work in the real world? The short answer comes from experience – yes, it does, which is currently demonstrated in the pilot project with Audi.

    For the pilot implementation, the chosen focus was a door panel pre-assembly. At the Audi Ingolstadt plant, our arculees bring these door panels directly to the station where the components are assembled. “By reducing production time through an orientation toward value creation and self-guidance, we can increase productivity by up to 20 per cent in some cases,” explains Wolfgang Kern, project manager at Audi Production Lab.

    Two arculees transporting Audi car doors at a modular production system in an Ingolstadt factory
    arculees transporting doors for assembly at Audi’s Ingolstadt plant (Source: Audi Media Center)

    The Cliff Hanger

    Despite the current success of the pilot, it will still take time until we see a big roll-out in the industry. The reason is simple: “The market is simply not yet ready to accept such a complex and sophisticated system”, clarifies Darshan. But the time and resources invested in the mechanism definitely paid off.

    Modular Production is the very root of our current solutions. “Our software ecosystem, for example, is one of our base modules, which controls how we drive and the coordination between robots. And that’s also the base for Modular Production – seamless coordination of AMRs for production tasks”, clarifies Darshan.

    Software view of the Fleet Manager

    However, one significant outcome of arculus efforts towards Modular Production is definitely the arculee. These autonomous mobile robots (AMRs) navigate freely within a defined area, avoid obstacles and orient themselves when picking up a load carrier by its position and alignment. Although originally developed with manufacturing tasks in mind, such features also fit in perfectly with intralogistics tasks.

    As Jungheinrich explains, “the small and manoeuvrable autonomous robot offers a ready-to-use solution for horizontal transport over large areas. This versatile assistant increases throughput and interacts seamlessly with automatic components in the warehouse, immediately improving efficiency in numerous industries and fields of application”.

    Two arculus mechanical engineers in front of a computer working on the arculee
    arculus engineers at work on the arculee

    With these two solutions combined, a third one also comes to life: warehouse automation. This goods-to-person system reverses the traditional picking process by bringing the goods to the employee on autonomous mobile robots, which use the arculus software ecosystem as their central control.

    The Moral Of The Story

    Although Modular Production is a solution for which the market still needs to be ready, its original concept served as the base for arculus current core products. Namely: the arculee, the fleet manager, and the warehouse management system. Now, with solid solutions distributed and integrated by Jungheinrich, we are ready to help reshape and modernize internal logistics across the globe while also laying the foundations for long-term opportunities with Modular Production.

    1 Kern, W., Rusitschka, F., & Bauernhansl, T. (2016). Planning of workstations in a modular automotive assembly system. Procedia CIRP, 57, 327-332.