What will it take?

In particular, the container terminal was thought to be among the most promising prospects. "An emerging application for AGVs that has tremendous potential for growth is the transportation of containers and cargo at ports. Rotterdam port?. is extensively using AGVs for such an application?. .Other ports are expected to use AGVs in large numbers to meet their rapidly growing container traffic?.

(and this is expected) to translate into huge growth opportunities for AGV manufacturers, " the report predicted.

But, the boom has not yet happened. So far the only terminals to have deployed AGVs in day-to-day operations are Europe Container Terminals (ECT) in Rotterdam and HHLA's Container Terminal Altenwerder (CTA), in Hamburg. These two remain big fans of the technology. "The high productivity in our container terminal is guaranteed to a large extent by Gottwald's AGVs, " says Ingo Witte, md of CTA. "The existing AGV system in our terminal has proven its reliability and performance."

But, considering the technology has been around for some time - ECT began working with Gottwald Port Technology on its AGV system 15 years ago - why are there not more such applications in existence?

The reasons are not hard to figure out. Many factors contribute to the average cost of moving a container around a terminal but labour remains the most significant. A report by the Center for Commercial Deployment of Transportation Technologies in the US found that a 1% increase in labour cost resulted in a 0.5% rise in average container cost. Therefore it is not surprising that AGVs have been deployed in Germany and the Netherlands, both countries with high labour costs.

"Labour cost is the most important factor influencing terminals to consider an AGV solution, " says Dr Armin Wieschemann, head of systems for automated port technology, at Gottwald Port Technology.

Second, an AGV system can operate 24 hours per day without any stoppages. "Like any other container handling system, a significant investment has to be made up front, but over time investing in AGVs will result in significant savings in operational costs."

Dr Wieschemann adds that the AGV solution will start paying for itself in roughly 3-5 years, depending on the size of the terminal, yearly throughput, required vessel handling performance and of course the hourly labour cost. Even this timescale could be reduced with the introduction of Gottwald's new environmentally and economically advanced E-AGV series, a diesel-electric drive AGV. Twenty-one EAGVs type CT 60-E will be supplied to CTA over the next year.

THE PROSPECTS?

So what are the prospects for further growth? Gottwald is convinced that AGVs will become more popular in container terminals. Dr Wieschemann sees the trend towards an increased acceptance of new technologies by labour unions, the growing concern about the environment and privatisation in the port industry as significant topics in this respect. Ports are becoming more disciplined in their management of costs and environmental issues. When combined with the more general phenomena of globalised markets and industrial automation, AGVs are expected to have a bright future.

Why would a terminal consider using AGVs, over more established handling equipment, such as straddle carriers, or truck-trailer combinations? Re-engineering the handling process has for some years been a critical issue for logistics and supply chain managers. Material handling is considered as a non-value-added activity, even though it contributes substantially to the final cost of goods. Yet conventional dockside handling equipment often fails to optimise the use of expensive and scarce land on maritime sites, and can lead to waste of capital in inventory, and long waiting times for trucks and ships.

But automating a container terminal is more challenging than automating, say, a warehouse or factory. Surprisingly, it is not because of the scale of site, or the fact that everything is done outdoors in varying weather. These obstacles can be worked round.

AGVs are now commonplace in industrial applications; uses that have highly regularised, predictable vehicle movements ('material flow' in logistics parlance). A car assembly plant is a good example.

The biggest challenge to introducing AGVs in a container terminal is that each box movement is far less predictable. A good ship-toshore cycle time for a container, perhaps 45 seconds, can easily be followed by one or several minutes (because of disturbances such as spreader downtime). Arrival times of vessels and landside trucks are also notoriously unreliable, so this is the sort of random logistics to be faced in a container terminal. However, sophisticated process control systems can often handle these 'stochastics' in a much better way than manually controlled systems.

And as Dr Wieschemann adds, the AGV hardware is not the most difficult part to do. The navigation software and management systems (AGV-MS or AGV fleet control) that have to be integrated with the terminal operating system (TOS) so as to guide the AGVs to their correct positions, scheduled and right on time, in order to maximize the fleet performance, is challenging though.

The economic potential of AGVs should be considered according to the following criteria:

Average and peak performance for the waterside transportation (per vessel and for the whole berth) Risk of breakdown Total operational costs (cost of ownership) The operational performance depends on the AGV's characteristics (speed, acceleration/deceleration rate, curve radius, handling provisions) and also on the complementary equipment handling systems (navigation, AGV-MS, TOS). Though with experience from existing systems and simulation studies, it is not too difficult to determine an AGV system's performance that will meet expectations.

The risk of breakdowns can also be estimated with conventional failure mode and effect analysis (FMEA) and breakdown data from existing systems. These two aspects would not normally hinder the application of the AGVs when selected with the proper systems layout and proven design.

So, that leaves total cost. This is composed of the following elements:

Equipment (AGVs) fixed costs, eg: depreciation, capital cost, yearly inspection, insurance variable costs, eg: fuel, spare parts, tyres, maintenance, lubrication Facility interchange areas provisions for navigation traffic control provisions AGV Management System cost of capital and depreciation of hardware for fleet control system, interface, tele-diagnostics and radio-data communication cost of capital and depreciation for control system software (control, data logging, data communication and user interfaces).

Plus the annual cost of maintenance and modifications should be included Labour Although the AGVs themselves are unmanned, some minor labour costs are incurred for logistics control and overall monitoring.