Cutting your Cloth

01 Nov 2005
Oslos Ormsund Terminal - 9 wide (+1 lane) RTGs

Oslos Ormsund Terminal - 9 wide (+1 lane) RTGs

Many new container terminals are built to a tried and tested formula, particularly by the big global operators. Benedict Young talks to two experts on the different rationales and approaches.

While P&O Ports historically have a preference for straddle carrier operations, Hutchison seems to prefer trucks and yard gantry cranes. "The manner in which containers are handled is crucial to their success and they're not in the business of carrying out a raft of experiments, " says Jonathan Tyler, director of Royal Haskoning regional maritime office.

Extending older facilities and developing brownfield port areas is perhaps a greater challenge. With a traditional quay, the usual problem is lack of space both behind and in front of the berth. "Many traditional cargo ships have never got longer than about two hundred metres and a lot of container ships are a lot bigger than that, " says Richard Clarke, market sector director for containers, at Halcrow.

"You always have to turn the ship either in the approach to the berth or coming off the berth and just to find those turning circles can be quite a major challenge within an existing port."

Traditional cargo berths had a shed close to the quay and space for trucks behind the shed with a typical depth back from the berth of about two hundred metres. "When we're starting container terminals, we like to have 500 metres back from the berth, " says Clarke. "You can manage with about 350 metres but anything less than that and you're constraining the capacity of the berth."

Exposure to wind and waves is generally not a problem with traditional berths. However, converting bulk berths can be a problem as the loading or discharging of bulks can tolerate ship movement of up to three quarters of a metre. "With a container terminal, anything more than 100mm of ship movement is pretty difficult, " says Clarke.

"Generally you can shelter part of a berth by putting a stud breakwater out at right angles from the end of it but that tends to give you access problems. Very often the only answer is to say that they'll have to build somewhere else or accept high downtime. You can tolerate high downtime on a feeder berth but if you have mainlines coming through, you can't accept anything more than five or six days downtime a year."

Even if a terminal doesn't face downtime, wave action can reduce productivity which can be a particular problem where lines are demanding guaranteed throughputs. A PIANC working group is now beginning to research how ship response to waves translates into loss of productivity at the berth.

YARD ISSUES Once the berth and the other main parameters of a terminal are in place, the layout can be decided. With an existing site, the shape of the yard may dictate the solution. "In Oslo we had a long thin yard and ended up with just two stacks with the only nine-wide RTGs in the world because we couldn't quite fit a third stack in, " says Clarke.

"That was an example of tailoring the machinery and the operating system just to fit the shape of the yard. There was no way we could get more space." (Interestingly, the machines shown in the photo are all electrically powered Kalmar RTGs to avoid noise and pollution in the city location. ) In some cases, operators may prefer RTGs or RMGs but the only solution is straddle carriers which can tolerate odd shaped yards much better. They can work a series of short stacks whereas there's no substitute for having long parallel stacks with RTGs.

"A lot of the conversions that we get involved in are for fairly low density yards. In which case you can use reachstackers which is a system that a lot of people decry, " says Clarke. "But at sensible costs and to fit things into an existing situation with lots of flexibility, it's a pretty good way to operate a yard up to say 150,000TEUs a year."

Building infrastructure to handle the projected growth of a container terminal has many inherent risks. Throughputs in ten or fifteen years may be far more or far less than anticipated but phasing development can offer safeguards.

"If a port is built where there is quite a lot of reclamation, our usual approach is to make sure there is a large amount of reclamation for the future but within sensible limits, " says Tyler. "You only pave what you think you've going to need for seven to ten years into the future."

In this situation, provided the statutory approvals to build have been obtained, yard paving can be built relatively easily within two years of the need being identified. In some cases, part of a site may be more desirable to build on because the reclamation will settle less in the early days compared with another part of the site. Phasing the construction lends itself very well to this and Tyler also recommends careful phasing of the equipment purchasing to match expenditure and income as closely as possible.

SOFT VS HARD CHANGES As well as modelling changes in the physical yard, which Tyler calls the hard engineering, he also recommends analysis of the soft issues such as operational changes. "Opening your gate an extra half day a week does not require extra infrastructure, " he argues. "We usually carry out sensitivity analysis with regard to dwell times as well. Some ports historically have a reticence to get tough with customers and offer a lot of long free storage and it's not surprising that those yards are completely full. We work on the premise that a soft change can be much cheaper to implement than hard change.

Computer modelling has become a popular tool for terminal planning. Tyler argues these techniques can very rapidly show bottlenecks that arise or why it might be better to reroute a terminal road or change traffic flows. "We've developed software which models the whole terminal from the quay, through the yard to the gate and the rail head, " he explains. "It models every container and equipment which carries it. There are lots of variables and the skill is in varying the parameters in a controlled manner so that the results actually tell you something."

Clarke and Tyler both warn that it is possible to use sophisticated modelling techniques which you think are providing meaningful insights when in fact too many and varied parameters are set. The consensus is that the inputs must be varied to make sure that the solution is robust so that if any one of the variables changes slightly, it doesn't completely destroy the productivity of the operation.

"Simulation is fine for deciding whether you need 16 or 17 RTGs but it shouldn't be used to decide how many RTG stacks there should be as that should have been worked out manually beforehand, " argues Clarke. "When you're doing the optimisation, you should be pretty cynical. It's quite useful to have an experienced operations manager to hand because he will usually tell you if the drivers would or wouldn't act in a certain way."

An example of this is a straddle operation where three or four units work with one quay crane. In theory, greater efficiencies can be achieved by pooling the straddles so that when one finishes a move, it is given instructions on where to go next. However, drivers don't like being run by the machines and productivity can reduce due to lack of motivation.

Clarke concludes, "I often joke that the biggest single benefit of simulation is the questions that the owner has to ask himself to set up the parameters. Very often that shows him what the problems are with the existing operation."

Links to related companies and recent articles ...