How is every single link in the BMW supply chain being tightened?

Posted on 5 Mar 2019 by Jonny Williamson

Digital technologies unquestionably have the power to make supply chains more transparent and more efficient. They connect manufacturers to suppliers at one end and to customers at the other.

But how should manufacturers use digital technologies, in particular automation, to improve the internal supply chain – from Goods In to Goods Out? BMW has embarked on a programme that aims to tighten every single link in their supply chain, external and internal, to create a model of manufacturing efficiency.

Nick Peters visited the BMW plant in Regensburg to learn more.

The technologies and concepts BMW unveiled are already being assessed in a number of pilot projects across its global network – image courtesy of BMW Group.
Cutting-edge technologies and concepts are already being assessed in a number of pilot projects across BMW’s global network – image courtesy of BMW Group.

Large companies are no different from small ones in wanting to squeeze waste out of every process. It’s just that when it comes to the resources they can throw at the issue, large ones have a significant scale advantage.

But the processes they adopt and the mindset they bring to efficiency can offer companies of all sizes insights into how to create efficiencies of their own.

That is why I was intrigued to receive an invitation from BMW to visit their Regensburg plant, about 60 miles north-east of Munich, to be taken step-by-step through the progress they are making in creating the BMW Connected Supply Chain (CSC).

It was a programme they started in 2016 (and reported by The Manufacturer’s Jonny Williamson at the time), and while BMW aren’t far from completing the core platform, it is destined to be a process that will last for years – possibly forever – for as long as a new technology comes along that enables further refinement.

“We will always have some proof of technology or proof of concepts going on,” BMW Logistics Head Jürgen Maidl told me. “Yes, there will be a point in time when we say the base is set in terms of single technologies such as the Smart Transport Robots we use in our production halls, but in terms of connecting the whole environment with IoT sensors, well…I suspect I will have long retired by then!”

The scale of the logistic challenge at BMW will be familiar to anyone who has worked inside large OEMs with multi-continent manufacturing plants, but it bears spelling out:

The BMW Supply Chain

Number of plants worldwide: 30, including the US and China

Number of suppliers: 1,800 at 4,000 different locations

Number of parts delivered daily: 1,000,000


This article first appeared in the February issue of The Manufacturer magazine. To subscribe, please click here.

Managing volumes like that takes time and human involvement, and both cost money. How to save some of that?

Start-up agility

Jurgen Maidl, BMW’s Head of Logistics - image courtesy of BMW Group.
Jurgen Maidl, BMW’s Head of Logistics – image courtesy of BMW Group.

With this volume of traffic, squeezing even micro-inefficiencies from the system offers significant returns, which is why the CSC is so vital to the future of BMW, whose optimistic strapline is ‘The Next 100 Years’.

Since the CSC programme began, more and more suppliers and transportation providers have gradually been brought into the system. By the end of 2019 it will be several thousand, allowing BMW to create a seamless and transparent process from supplier through to dealer.

This will include each car being fitted with trackers that allow dealers to know precisely where their customer’s car is at any time on its journey from plant to forecourt (with a firm promise this tracker will be switched off for good on arrival).

It will even extend to fitting each car with a limited autonomous module that means it will drive itself from the final inspection bay to the car park to await loading onto a transporter.

Note, one more human interaction gone – this is about reducing as far as possible each non-value-add element of the production process.

The team charged with investigating how to do this is led by Marco Prüglmeier, Head of Innovations and Industry 4.0 for BMW Group Logistics.

Prüglmeier’s team has permission and scope to behave like a nimble start-up within the otherwise formal confines of the Regensburg plant and other BMW factories around the world where new ideas are being tested.

“We always have several pilot projects running at our locations worldwide,” he said. “We leverage various forms of international cooperation to make sure we always have access to the latest findings and technologies.

“We learn the most from these pilot projects. We already transferred a number of projects to series production, with further implementations planned for the future.”

Supermarket

BMW Supply Chain - Smart Transport Robots (STRs) use their hydraulic platform to lift parts trollies off the floor and move them to where they need to be - image courtesy of BMW Group.
Smart Transport Robots (STRs) use their hydraulic platform to lift parts trollies off the floor and move them to where they need to be – image courtesy of BMW Group.

BMW Plant Regensburg produces some of the 1, 3 and 4 series and Z4 BMWs. Every day, it receives its slice of those one million parts delivered to BMW plants around the world.

As Marcus Wollens, VP of Logistics at BMW says with commendable understatement, “It is about getting the right parts in the right order!”

Of course, that happens already. Marco Prüglmeier’s job is to use digital technologies to make it happen faster and cheaper.

The sequence starts with receiving parts in plastic crates mounted on pallets. The crates are taken off the pallets and the parts sorted into the correct storage bins for ultimate delivery to assembly teams at precisely the moment they need them.

What is intriguing is how the language of retail permeates the system. The parts hub is even described as the ‘supermarket’. But what really catches the eye are the robot transporters – autonomous ‘Tugger’ trains – carrying large loads around the plant.

These are supplemented inside the plant by Smart Transport Robots (STRs) which are essentially small flat-beds on wheels with hydraulic lift capability, which allows them to slide under shelf trollies, lift them a few centimetres off the ground then deliver them where they need to be.

They are self-navigating using SLAM (Simultaneous Localisation and Mapping) technology that allows them to operate independently of any internal networks. When something small needs to be collected and delivered urgently, miniSTRs, the ‘R2D2s’ of the operation, are deployed.

The transportation part of the process is relatively simple. Autonomous vehicles inside factories are not new, even if the STRs are a rather clever development. (They are powered by the same battery modules that power the BMW i3, by the way.)

There is an autonomous vehicle outside the factory too, the Autotrailer, which navigates by laser to shift large truck trailers up to 30 tons. And there is the variant known as the AutoBox, which can move trains of 20 large trollies, up to 25 tons at a time, between buildings.

Where the real smarts are being deployed is in the way Prüglmeier’s team are using digital technologies to streamline how parts are sorted and disseminated to where they need to be.

It has always been a human-intensive process, picking and packing parts and despatching them just-in-time to the assembly lines, even if that has been made easier over the years via the adoption of retail technologies such as barcodes and QR codes.

BMW Supply Chain - Pickbots will develop AI-driven memory to distinguish between 50,000 separate parts - image courtesy of BMW Group.
Pickbots will develop AI-driven memory to distinguish between 50,000 separate parts – image courtesy of BMW Group.

BMW believes each human element of that process can either be replaced by a machine or value can be added to the humans that remain via technologies such as virtual/augmented reality (VR/AR).

It is an intense, innovation-rich programme, with engineers supported by PhD students being given specific problems to resolve and all the tools they need to do it.

Watching these teams deploying elaborate arrangements of robotics, IoT sensors and VR/AR is a reminder that digital technologies may be capable of fantastic things, but it still needs human ingenuity to work out how to make the most of them. Otherwise they might as well stay in the box.

Pick ‘n’ pack

The job of managing inbound goods will in future be done by machines, some of which have already moved beyond the concept and trial stage to stage 2 testing deployment in manufacturing plants.

SplitBots take parts containers off inbound pallets and place them on conveyors for transmission to the next stage. Using AI, these bots can recognise 450 different containers.

PickBots take parts from containers using suction pads and place them in trays for carriage to the assembly line. These bots use AI to distinguish between different parts and learn how to pick them up – eventually, they will recognise 50,000 different items.

The PlaceBot will be stationed on assembly lines and unload parts from ‘Tugger’ trains and place them precisely where they need to be.

Finally, the SortBot takes care of empty containers, stacking them ready for despatch back to suppliers.

Also under trial is a 3D system that looks very similar to that developed by Tharsus for Ocado, whereby containers are stacked on top of each other in a grid, with robots whizzing around on top, shifting containers within the stack as they are needed and pulling parts out as needed.

BMW Supply Chain - Scanners mounted on gloves, or within smartphones, transmit comprehensive parts data without the need for paper – image courtesy of BMW Group.
Scanners mounted on gloves, or within smartphones, transmit comprehensive parts data without the need for paper – image courtesy of BMW Group.

The more minute sorting of parts will still be done by humans. What will take this process to the next level is the introduction of a unique part ID, a QR code, that will contain every bit of information about each part, including who supplied it, when it arrived, and where it is stored.

It is the ultimate iteration of the connected supply chain. This single ID will be applied at the point of origination and data will be added via the central server, so that every point in its journey is paperless.

That goes for the workers responsible for managing the parts in the plant. They will no longer have paper, instead wearing glove scanners that can read the parts’ IDs and display them on small screens on the gloves.

Similarly, researchers are testing AR goggles that speed up the process of slotting parts into the right bins, super-imposing images that guide the worker to the right bin every time by warning when a mistake is about to be made.

Smartphones are also part of the equation. QR code scanners will place all the information about a part directly into the hands of every worker who needs to know it.

BMW’s programme is the ultimate vindication of the power of digital technologies. The company also understands that much of the technology behind the Connected Supply Chain must be shared with other companies and trade bodies so that common standards emerge.

This is particularly applicable to autonomous transporters that must be able to talk to each other, even if they are deployed by different OEMs.

Equally, as BMW and other large companies begin to crack these problems, other companies will learn from them, adapt the techniques, in all probability commoditise them and thus they will eventually become mainstream, coming soon to a factory near you. To yours too, perhaps?