News / The paradoxes of port productivity

Photo: DP World

This year, the port of Tianjin revealed that automation had provided a boost to the efficiency of its rubber-tyred gantry (RTG) cranes. From the 15 moves per hour controlled by human workers, the terminal achieved 20 using automated cranes connected to the Sany Marine’s H-Move2.0 system.

A 33% productivity increase is certainly impressive. One human operator can control six RTGs simultaneously, with what is termed an ‘intervention rate’ – the number of times a human is needed to become involved in a crane move – limited to 2.78%. And the system can compensate for the shaking and twisting movements that occur when an RTG picks up a container.

Automated stacking cranes (ASCs) have been extant for some time, but the handling rates of Tianjin’s automated RTGs speaks to some major progress in the field. In a 2010 whitepaper by ABB, the Swiss engineering giant that is a major supplier to port automation projects, suggested the maximum number of ASC moves at just 10 per hour.

Almost every step in the port handling chain can be automated today, monitored by a small number of workers based in an onsite office. The ship-to-shore crane unloads a container to a vehicle on the quayside, such as a straddle carrier or automated guided vehicle (AGV), which then takes it to the yard, where it is positioned in a ‘block’, handled by vast ASCs or RTGs which will finally load it onto a truck departing into the cargo hinterland.

By redesigning existing terminals it is hoped that automation will address one of the most significant needs facing 21st century ports – densification. Many ports have little prospect of acquiring more land for additional yard space, meaning capacity increases can only be achieved by stacking containers higher and closer together. This means administering blocks or stacks with a greater number of machines; and either hiring additional port staff or using increasingly automated fleets of vehicles and cranes.

At LBCT, 36 blocks, stacked six containers high and ten wide, are administered by two ASCs each. In times past, each of these would have required a driver, and a corresponding spotter on the quayside to oversee each crane movement; now, just 14 workers oversee the entire fleet. Those blocks are attended by nearly 100 AGVs.

The very largest systems, cantilever automated rail-mounted gantries (c-ARMGs), can stack containers up to eight high and 14 wide. These help to achieve maximum density at Haifa’s Carmel Terminal.

An even more outlandish effort is under way at DP World’s flagship terminal, Jebel Ali. There, a test version of the newly designed and automated Boxbay high-bay system can accommodate up to 1,300 teu in a footprint of 230 metres by 26 metres. The system has completed over 330,000 container moves thus far, and the terminal operator is now rolling out the technology at its facility in the South Korean hub of Busan.

There is more to port automation, then, than shedding jobs. Rather than reducing labour costs, automation’s chief benefits are centred around greater density, higher stack height and a higher number of container moves overall.

A recent study by US west coast terminal employer group the PMA [Pacific Maritime Association] points to a 44% increase in teu throughput at LA and Long Beach, despite some 572 jobs being eliminated in 2020-21, “…thanks to autonomous vehicles and cranes that stack containers higher, closer together and more efficiently for transferring to trains and trucks”.

However, other findings suggest there are problems that need to be overcome if automation is to deliver on its benefits. For example, unable to operate independently, connection issues can spell disaster for automated machines, incurring substantial downtime.

“I’m seeing all types of automated equipment breaking down, STS cranes, ASCs, AGVs,” Rebecca Schlarb, an automation coordinator at LBCT and ILWU Local 63 member, told Jacobin recently. “For the ASC in the yard, the average minimum downtime is 20 minutes to an hour… if one ASC goes down in a block, it is not only the broken crane that’s taken out of operation, but the second crane within that block as well.

“Additionally, for the safety of the mechanics, both ASCs in the adjacent yard block (a total of four ASCs) are also taken out of operation to provide safe passage to the broken crane.”

Straddles and AGVs are similarly vulnerable. “An AGV follows a travel path, generated by the program software, from one part of the yard to another,” Ms Schlarb explained. “When the wireless connection is lost, the AGV immediately stops driving, and a large safety block is generated by the software, stopping all other AGVs in close proximity.”

The need for highly trained engineering staff to fix issues when they arise competes with the cost of manual equipment operators, suggested a 2017 study conducted between McKinsey and Shanghai International Port Group, which found underwhelming results from the automation of a greenfield terminal.

In the study, savings on operating expenses were around 26%, compared with a target of 25%-55%; but productivity also fell, by around 11%. Container moves in a fully automated terminal were in the low 20s, versus high-30s for conventional container terminals.

It was not that human port workers were better than the machines, McKinsey determined; rather, that the highly trained staff required to administer an automated terminal were in short supply and took a long time to train.

This helps to decode the reason for ports’ recent fascination with 5G. Instead of using wi-fi, whose connections are famous for dropping out at inopportune times, port operators interested in automation are seeking to install private 5G networks, which entail more robust connectivity, less power use, less risk of signal stoppages and higher bandwidth for exchanges of data between equipment and the offices from where it is monitored and controlled.

Following a 2019 trial of 5G on an STS crane at the port of Qingdao, study partners Ericsson and China Unicom claimed that if 5G-driven automation was extended to the entire terminal, it would reduce labour costs by 70%, even compared with a conventional port automation project.