Whether you like it or not, artificial intelligence (AI) is here to stay. While some have their concerns, the rise of the machines is not all bad, not all "Terminator 2" territory; it's hard to argue against the need for improving efficiency in how we do things. With the exit of Audi from the FIA World Endurance Championship, they've given us a peek into how they see the future of building cars, and it is radical stuff.
Utilizing the power of "big data" married to AI, the German giant intends to throw out the constraints of the traditional assembly line and become much more flexible, and up to 20 percent more efficient.
The Strain of Tradition
At the beginning of 1914, Henry Ford introduced the first full assembly line production in the automobile industry in his newly built factory in Dearborn, Michigan. Since then, this principle has formed the basis of large-series production in the automobile industry. Audi is now working on a vision for the era following the assembly line, which involves the development of a completely new principle called "modular assembly."
If it ain't broke, why fix it?
Well, if the traditional production line is holding up, it is under strain, at least. Analysis shows that as model diversity grows, the more complicated it becomes to master the complexity in a rigid sequential process while integrating more and more new working routines. The fixed tempo leads to inactivity on many sections of the line — for example: the installation of optional extras such as auxiliary heating systems, which only a small proportion of the cars are fitted with. These losses accumulate with an increasingly heterogeneous model mix on the line.
With typical German efficiency, Audi gives a clear example of the problems they face continuing with traditional methods of production. When sharply differing versions are on the same assembly line, it becomes even more difficult. Put simply, the line moves at the pace of the slowest car assembly process:
"One example is the assembly of the Audi A3 Sportback e-tron in Ingolstadt. The plug-in hybrid model, which accounts for only a relatively small percentage of the overall Audi A3 production, passes through seven separate workstations, where it receives a large proportion of its electrical equipment. While this is going on, its sister models with conventional drive move along the conveyor belt suspended below the ceiling; they are not worked on during this time, so the time until completion becomes longer for all the cars on the line."
New and Improved
Audi's solution is to develop a modular production line with an increased number of assembly stations to take advantage of the new flexibility. To optimize this new philosophy in the struggle for maximum efficiency and productivity, a number of new tools are also being developed and refined, undergoing trials at the Audi engine plant in Győr, Hungary, as well as the Lamborghini factory in Sant'Agata Bolognese, Italy.
Autonomous forklifts will replace the 500+ manned units currently used at the Ingolstadt factory. The drivers will be reassigned to more productive roles, which is said to also boost efficiency and safety.
There will also be the introduction of Driverless Transport Systems or DTS. The aim of the DTS is to pick up parts from a remote warehouse location and deliver them to the modular assembly station with complete accuracy.
An element of these autonomous parts bins is that they are powered electrically and will serve multiple workstations with a single set of parts and components. Both of these driverless floor conveyors use laser scanners to monitor their routes ahead and evaluate the safety systems of the trailers they are towing. This avoids potential collisions with persons or objects. Unlike conventional driverless transport systems, driverless floor conveyors have the advantage, as they are developed from existing vehicles. Standard equipment such as trailers, existing structures and transport relations can generally be taken over without any major adaptations.
Humans are not excluded from the production process; "human-robot co-operation" is how Audi describes this aspect of the new systems. Pictured we see a worker operate a FlexShapeGripper that can grip and hold objects, passing them to an employee or placing them into a workpiece carrier to ensure the correct choice and fitting of, for example, an A3 roof aerial.
CleverKlaus helps the workers on the pre-assembly of Audi A4 doors with the complex cabling. The model range includes several hundred different variations; in the top versions, the door trim integrates up to 14 connectors for the electric windows, loudspeakers, central locking, mirror adjusters and other optional equipment. Two high-resolution 2D cameras above the table check whether all the cables have been connected properly. The sequence of work is irrelevant.
The LBRInLine is a robotised production station for fixing undertrays to A3 and Q2 models, there being a great variety of parts and fixing depending on the individual car. Under the control of the worker, the robot tools performs the fixing operation, increasing accuracy and cutting down time.
And straight into the space age, Audi is experimenting with drones to deliver lightweight components such as steering wheels. This not only ensures the correct part is where it needs to be, when it needs to be available, but it also helps the narrow spaces in the assembly lines avoid clogging up.
3D printing is also in extensive use in the prototype stage of producing parts; Audi is working to accelerate the pace at which these tools work. The parts are stronger and lighter than more traditional prototyping processes, but currently, the 3D printing is too slow.
The first step in car assembly is the production of sheet-metal parts. Audi believes that AI and big data can fine-tune the accuracy of this fundamental process by comparing the CAD design with a detailed analysis of the finished product. In this way, the press shops are becoming part of the smart factory, too.
Currently, Audi finishes around 2,300 cars per day. The logistical management of moving cars once they're made is another major potential roadblock, taking 1.6 days to clear that number of cars. Predictive Yard Management is the proposed solution to smooth the flow of cars and transporters to maximize efficiency, extrapolating the big data already used in the Smart Factory initiative.
However, it is not just internally that Audi is looking to use AI to solve everyday, real world problems. Examples of this revolution are already with us; there was a recent announcement that Audis can network with U.S. traffic lights:
"If you know in advance when a traffic light will switch from red to green, your driving is more relaxed and efficient. Audi is the first automobile brand to connect the car to the city infrastructure — an important step towards autonomous driving. In Las Vegas, starting immediately, the Audi A4 and Q7 models can display traffic-light phases directly in the car."
This, like autonomous cars, is part of the trend to take control away from drivers and presumably to hand it over to the authorities. Handing power of choice and freedom over to our rulers has always worked so well in the past — what could possibly go wrong?
Parking in cities is another aspect of motoring that can be challenging to say the least. Last year, Audi announced a partnership with the city of Somerville, Massachusetts, just outside Boston:
"Thanks to its central location in the Boston metropolitan area, Somerville is a role model for 'smart cities.' After Silicon Valley this region has the fastest-growing economy in North America. In the coming years, Union Square in Somerville will be transformed into a flourishing city centre. By means of urban redevelopment, new dwellings, offices and commercial real estate will be built. More people on the same surface area also means, however, that the existing mobility infrastructure reaches its limits. Therefore Audi is supporting the project at Union Square with innovations such as the traffic-light assistant. This will help traffic to flow faster.
"In addition to networked infrastructure, Audi is bringing automated parking to the project. Self-parking cars result in three different benefits. Parking garages can be relocated from the city center to less attractive places. At the same time the parking area required per car is reduced by approximately two square meters. The cars park closer together and need fewer, much narrower lanes in garages, where pedestrian paths, elevators and stairs are no longer required. A parking garage of the same size can then take up to 60 percent more vehicles — sufficient to end curbside parking. Finally, there are fewer cars on the roads searching for a place to park."
The whole motor industry is getting behind this sea of change, another recent proclamation:
"BMW Group, Daimler AG, Ford Motor Company and Volkswagen Group with Audi and Porsche have signed a Memorandum of Understanding to create the highest-powered charging network in Europe. The goal is the quick build-up of a sizable number of stations in order to enable long-range travel for battery electric vehicle drivers. This will be an important step towards facilitating mass-market BEV adoption.
"'We intend to create a network that allows our customers on long-distance trips to use a coffee break for recharging,' says Rupert Stadler, Chairman of the Board of Management of AUDI AG. 'Reliable fast charging services are a key factor for drivers to choose an electric vehicle. With this cooperation we want to boost a broader market adoption of e-mobility and speed up the shift towards emission-free driving.'"
This transformation is the real reason behind the financial sacrifices that are under consideration in the VW Group — similar programs are, no doubt, underway at all the mass production manufacturers in the automotive sector.
It is going to require imagination and courage to bring into existence this revolution to enable us to have some form of personal transportation in the coming decades. And money, lots and lots of money. "May you live in interesting times" is supposed to be an ancient Chinese curse; I suspect we are all going to test the validity of that malediction sooner than later.