SubscribeThis article is a follow-up of the review of stacking systems, and therefore we refer to this article for the introduction, the methodology of assessment and the review of existing systems.
The remainder of this article will encompass the following sequential topics: the qualities of the various technologies available with a focus on transportation systems. Subsequently we summarize the main results, performance and cost-wise. Finalize we wrap up with some concluding remarks.
The considered alternatives: transportation systems
Although the market for automated transportation equipment is still small - in comparison to the market for manned equipment - there are various systems on the market. Not all proven technology yet - a major requirement from many operators - but available for a conceptual battle on similar terms.
We have selected the following opponents:
- The good old AGV (Gottwald), currently running in Rotterdam and Hamburg. The diesel-electric version has been used.
- The automated shuttle / straddle carrier (Kalmar), currently running in Brisbane.
- The cassette AGV (TTS, diesel-hydraulic), not yet operational, but several trials in ports have been conducted. This AGV carries cassettes around, and is able to drop and pick-up cassettes by itself.
- The lift AGV (Gottwald, also diesel-electric), based on the existing AGV but with the possibility to pick-up and drop a container by itself at the interchange of the ASC. This concept allows mixing with existing fleets of AGVs, as the machines are almost identical, apart from a lifting platform on top of the AGV.
- The manual shuttle carrier (Kalmar, Noell); a 1 over 1 straddle carrier, which will soon be running in several places (Antwerp, Norfolk), and currently in operation in for instance Southampton.
We have left out the tractor-trailer, although feasible, since it has quite similar characteristics as an AGV - coupled interchange at ASC and QC.
Figure 4: From left to right: the AGV, the lift-AGV, the cassette AGV and, and the shuttle carrier.
The 5 alternatives have been compared under similar operational conditions: an operation with 10 quay cranes (single trolley, single hoist) and 25 stacking modules, equipped with twin ASCs (we also compared the operations with the other two types of stacking cranes, but we focus on this operation in this article). Each handling system has been tuned to utilize its capabilities to the full extent, i.e. decoupling and usage of the buffers at the stacking crane, efficient placing of empty cassettes, efficient job dispatching of transportation vehicles and stacking cranes.
The cassette AGV and the lift AGV do not decouple at the QC. Although being an opportunity, we experienced a greater need for empty cassette transportation, only decreasing performance when decoupling at the QC. The lift AGV requires platforms where the containers can be placed, and these would have to move with the QC which was considered too complex in practice. Of course, all systems apart from the traditional AGV decouple at the ASC transfer point.
The results
Comparison transportation systems
In Figure 6 the main result of the comparison of transportation systems can be seen: the quay crane productivity depending on the number of vehicles deployed. In all cases, there was global pooling of vehicles, allowing minimizing empty travel. Obvious for the automated equipment types, less obvious for the manned machines. The maximum achievable productivity is approximately 45 container moves per hour, starting with an average cycle time of the QCs of 80 seconds, and 85 seconds when operating in backreach (only in case of the manned shuttle carriers).
Figure 6: Net QC productivity for each transportation system, depending on the number of vehicles
What can we conclude from Figure 6? First of all, that all systems can deliver a similar productivity by deploying sufficient vehicles, and in case of adequate control software. So the perception that automated systems cannot deliver the same productivity as manually driven machines, is in our opinion not true; it requires more vehicles though.
Second conclusion is that the effect of decoupling is considerable: when just comparing the traditional AGV with the lift AGV, a performance increase of up to 30% can be observed, or a reduction of vehicles (at comparative performance ) of up to 50% (@40 bx/h). The cassette AGV also shows an improvement compared to the AGV, but less, as a result of the empty cassette moves (the average driving distance per container is almost 40% higher). Also the process at the QC takes quite long due to the necessity to lower the cassette onto the ground to avoid impact loads on the vehicle.1
Finally: what is needed to achieve 40 bx/h? In this particular terminal setting, 27 manned shuttle carriers (pooled), 30 automated shuttle carriers, 33 lift AGVs, 50 cassette AGVs and 65 AGVs. In the next section we will discuss what this means for CAPEX and OPEX.
Cost analysis
The cost analysis departs from a few assumptions:
The labour cost per operating man hour are assumed to be 40 euro. The cost of fuel 80 eurocents per liter. The prices (in euro) of the vehicles (incl. platforms for the lift AGV and the cassettes for the cassette AGV):
- AGV: 380 K
- Lift AGV: 500 K
- Cassette AGV: 565 K
- Automated shuttle carrier: 960 K
- Manned shuttle carrier: 480 K
The fuel consumption for the various vehicles (which determine to a large extent the operating cost per hour) is also a result from the simulation, based on the engine characteristics. Also considered are the additional investments in software required for the automated systems (up to 2.3 million euro additionally).
CAPEX and OPEX are shown in Figure 7.
Figure 7: CAPEX to purchase the transportation equipment to support 10 QCs at a peak berth productivity of 400 bx/h. Yearly OPEX of operating this fleet of equipment, including labour cost, running costs, maintenance and capital costs at an interest rate of 7%.
What can be concluded? First of all: all automated systems are obviously less expensive than the manned alternative, even considering the higher capital expenses. This means that the additional investment for an automated system (ranging from 7 million euro for the lift AGV to 19 million euro for the automated shuttle carrier) will be earned back in 2 (lift AGV) to 5 years (automated shuttle carrier). Secondly: the least interesting automated system, despite its capability to fully decouple is the automated shuttle carrier, mainly due to its high price, and relatively high maintenance costs.
Conclusions and reflection
Hop on the bus Gus, as Paul Simon used to sing? Well, there is still a lot of ground to cover for fully automated terminals, we believe. However, with the lessons learned from past implementations (lack of decoupling, little software robustness, equipment specifications), the case for automated terminals is strong. Economically, there is no question that full automation pays off in a short period of time. Performance wise, automated systems require more equipment, but they can achieve similar performance levels as manned equipment. Prerequisite is intelligent, flexible and robust software, which has to be proved in practice yet.
Finally, what about the social impact of automation? Yes, in direct labour, jobs will be lost, or more moves per labour hour will be moved. On the other hand, it requires more intelligence in the construction and maintenance of equipment and software, also creating more interesting jobs. Furthermore, can we imagine the automotive industry without automation? So, in order to stay competitive, automation in yard handling and horizontal transportation needs to be considered, and balanced against other interests.
1 Another study showed that in case of more transfer points at the QC (4 instead of the 2 used in this study) the latter effect could be minimized. However, more than 2 transfer points are not feasible in case of larger crane clusters, as used in this study (up to 7 cranes in one cluster). The only systems with 3 transfer points per QC are the shuttle carriers, manned and automated.