Email email Print print

Get with the programme

04 Oct 2011
While AGVs can answer some efficiency questions, they leave others, literally, hanging

While AGVs can answer some efficiency questions, they leave others, literally, hanging

The interface between automated vehicles and non-automated vehicles is often the weak link in the chain, finds Stevie Knight

One of the big problems with Automatic Guided Vehicles isn’t actually the AGVs themselves. These usually keep running quite smoothly, come wind, rain, sun, fog and even in one notable case, ground collapse.

But you often find there’s a set of highly variable factors at the end of the process – driving the gantry and yard cranes. However, John Hunter of Armor Business Solutions explains it’s not just down to people’s inconsistencies: “All it takes is one twistlock to be jammed, and you have a delay. Yes, a smart computer speeds things up, but a delay can slow things all the way down the chain.”

Floating the automated vehicles between different quay cranes and working across the area helps, but it still leaves a complex efficiency decision about which AGV should do which job. Container terminals have highly variable routes with traffic running between multiple points, and one destination - the quay crane - shifts its own position. The control mechanism to deal with this kind of complexity has taken years to develop, explains Remmelt Thijs of TBA.

Further headway against this kind of hold-up has been made by “decoupling” the AGVs movement from the stacking side. Mr Thijs explains that while the most common type of AGV is the lift-on-lift-off variety, the lack of a “processing buffer” – in other words somewhere to put the box while it waits to be handled – means the vehicle is tied down to waiting around to be served.

Innovations easing this include Gottwald’s Lift AGV which puts the box into a rack while waiting for the yard crane, courtesy of a lifting mechanism on the AGVs platform. It’s a good answer since the largest slice of efficiency is gained by decoupling at the stacking cranes.

Mr Thijs adds that another option is given by the use of a robotic shuttle carrier. Typically less high than manual straddles, these straddle shuttles do, however, still need a different layout to AGVs as the machines are wider. However, they are by comparison very flexible: because they can pick up and put down on the ground, they don’t have to hang around at either end of their journey.

However, Mr Thijs explains that the decoupling move at the quay end doesn’t give the same advantage as firstly the container is not always at the same spot for pick up, and secondly, the quay crane tends to want to move on even in the short time between a box’s delivery by the shuttle and its loading; a known issue even in conventional straddle terminals.

Mr Thijs adds there is a third option of using an AGV in combination with ‘cassettes’, the AGV slots underneath the cassette for loading – then drives with cassette to destination so that the system can also decouple both the stack area and the quay crane operation. “The down side is that you sometimes have to move empty cassettes, adding to the complexity, and also to the weight being transported. Still, with the right set of control rules, such a system could be effective,” he says.

Needless to say any system requires very, very robust simulation testing to fit together not just the physical space and the vehicles, but also the terminal’s aspirations. Mr Thijs says: “Rather than looking at one piece of equipment you need to look at the system view, and having the most efficient vehicles might not be the most critical aspect.” Efficiency is often just as much to do with keeping enough slack. Like a good suit, you don’t want it to fit too tightly.

For example, it might be counter intuitive but simulations show that sometimes a net gain is achieved through utilising a higher number of the lower-efficiency AGVs, rather than the shuttle carriers. Taking just one instance, if a vessel comes in at the wrong berth it means longer-than-planned travel distances: this is something that a fleet of AGVs actually deal with better overall, just because the extra travel takes up the slack afforded by the waiting time.

Further, before any hardware is put in place, a “brain” robust enough not just to deal with the everyday challenges, but especially those unforeseen difficulties that could herald things going disastrously wrong needs to be found. Mr Thijs is obviously keen on systems that have a tried and tested maturity.

”It should be obvious that a software system has to have a number of fail-safe devices and be able to deal with unforeseen circumstances without being shutdown completely,” he says. Not investing in thorough emulation testing before go-live, he adds, “is shortsighted”, as beyond the safety aspect, it could easily mean more cost or underutilisation of expensive equipment.

In Rotterdam and Hamburg the AGVs are flexible enough to travel to multiple locations inside a bordered area. This in itself places strictures on the layout. “We strongly recommend no mixing with manned vehicles, AGVs need a completely separate area,” says Mr Thijs.

However, says Mr Hunter, if close interaction is needed, such as in actually loading and unloading a truck, then you have to be very careful. “This can be the most complex part of the development,” he adds.

For example, at the start of AGV operations in Brisbane, a robotic vehicle would line up the container, then a remote worker with a joystick would take control and guide it to the truck. When eventually the action became automated, there was still the problem of keeping truck drivers away from the automated straddles, and a complex ‘gate’ system was maintained.

Further, there are maintenance issues to deal with. “The Patrick operation in Brisbane is full of mobile straddle carriers, so there was quite a bit of work done on being able to isolate a path to a unit, for example, when one breaks down and you need to get engineers in,” says Mr Hunter.

There are other system requirements that aren’t obvious at first glance. Along with the first battery-powered AGVs that are now in pilot mode at the HHLA Container Terminal Altenwerder in Hamburg, there is a fully-automated battery changing station.

This battery power innovation means dropping the diesel generator entirely, so AGVs can now add to their ‘clean and green’ credentials: these weren’t entirely lacking because fuel useage tends to go down when you put a computer in charge. Mr Thijs explains: “Computers drive more consistently, although you can encourage certain behaviour from humans, you have to realise they are all individuals.” And of course, this adds to both the fuel and the maintenance bills.

Moreover, AGVs lower both noise and light pollution – both of which can be a problem when you have a terminal near a residential area. Mr Thijs says: “Robots put down the containers gently and consistently. Humans are sometimes driven by bonuses – and sometimes drop the containers on to the paving with a bang!”

Further, Mr Thijs adds that the AGV units are, by comparison, long lived when you consider the work schedule: Gottwald is now replacing a fleet of AGVs at Europe Container Terminals (ECT) in Rotterdam that has been operating round the clock since the early 1990s. However, the newer battery powered AGVs seem to have even lower ongoing costs than the diesel. “The system has fewer moving parts, so less maintenance is needed,” he says. “Overall the saving on operating costs in fuel and maintenance costs goes down significantly.”

He adds that in fog, low light – or no light – conditions, AGVs are actually much better than their manned counterparts. These terminals may find themselves not needing so much light across the apron for large parts of the day – which may be a bigger expense than many realise.

Obviously the costs are considerable, and have to be justified over the longer term. Still the maturity of the equipment and control systems has improved and costs are decreasing.

There are other downward pressures on the price. Silke Schael of Goetting explains that there's “growing interest” in the conversion of standard vehicles for automation and it’s usually a cheaper option than specially designed vehicles, because both initial cost as well as maintenance will be lower. Further, the increasing use of electronic controls in standard trucks (such as CAN bus systems) has meant the price of conversion itself is being brought down.

Ms Schael adds that there is also more interest in leaving the drivers cab intact, one good reason being that if a truck breaks down, you can go in and drive it out of the section to allow it to be serviced. But another more interesting point is that trucks could have ‘dual purpose’ controls, and actually still be used for normal manual operation for part of its operation.

However, it has to be said, labour issues are often behind any automation push, and this can be even more complex to evaluate than the mechanics. “It’s not just places like Dubai’s Jebel Ali terminal, which can, on the basis of cheap labour, afford to keep a number of terminal tractors hanging about just waiting to service each station,” explains Mr Hunter, “it’s places like Bremerhaven in Germany which have suffered from such high unemployment rates they simply may not ever consider automation.

"There are simply times its better to keep people employed, because it ties a port closer to both the local government and the community - you don’t make yourself popular by laying people off.”

Images for this article - click to enlarge

While AGVs can answer some efficiency questions, they leave others, literally, hangingGottwald’s Lift AGV uses yard crane waiting time to put boxes into a rack, courtesy of a lifting mechanism on the AGVs platform

Unless otherwise stated, all images copyright © Mercator Media 2012. This does not exclude the owner's assertion of copyright over the material.




Business News - Sign Up Today!

Email news News feeds
Magazines Networks