Don't just automate it, orchestrate it!

Autonomy companies are racing to put increasingly complex, expensive AI directly onto autonomous vehicles, which is not the best way to go about automating a container terminal.

The problem is that this kind of autonomy-centric approach invites terminals to wait for some magical just-around-the-corner autonomous solution that will allow them to automate their operations with minimal effort and investment, when there already exists a superior alternative: field-proven automation systems such as Kalmar One.

Coordinating the whole show

As a general rule, it is impossible to optimise the operation of an automated terminal unless you optimise the operation at the level of the entire terminal. This might sound obvious, but it is something that is often forgotten in the AI hype driven rush to cram ever more advanced autonomous features into container handling machines.

While autonomous machines can do some very impressive things, they are fundamentally built to perform a few tasks: to drive safely in a specific environment, execute designated movements, and stop when necessary. However, it is unreasonable to expect autonomous machines of any current or future level of sophistication to be able to make routing or scheduling decisions that are efficient at the level of the entire terminal.

This is because individual autonomous vehicles lack a real-time, holistic "world state" view of the operation. They are not aware of the positions or future destinations of other automated and manual equipment, dynamic traffic rules, container handling task queues, or the exact status and location of the quay cranes.

A busy container terminal can be incredibly crowded, and every metre of space will be in demand for some part of the operation. If autonomous machines are expected to navigate this environment solely based on their own sensors, collision avoidance systems and onboard routing algorithms, it is guaranteed to lead to hesitant machine movements, traffic conflicts and costly bottlenecks.

A harmonious ensemble

A good analogy for the importance of a centralised automation system is the conductor of a symphony orchestra. The automation system as conductor holds the complete, real-time understanding of what needs to be accomplished, while the individual machines in the fleet focus on playing their own parts that all mesh together in perfect synchronization.

This kind of orchestration is critical for automated terminals. Shifting the "brain" of the operation away from individual machines and into a centralized automation system such as Kalmar One is the only way to achieve true operational certainty, eliminate vendor lock-in, enable investments today and make terminal automation cost-effective in the long term.

Yes, there is a place for advanced onboard intelligence in driverless vehicles and container handling machines. AI, machine learning and advanced sensoring can bring great improvements to the driving performance, container handling and safety functions of the machines, and this is where we should focus these development efforts. However, someone or something still needs to coordinate the operation.

By removing decision-making from the isolated machines and centralising it in a solution that can optimise the performance of the entire terminal, we can perfectly orchestrate machine interactions, designate and update optimal routes in real time, avoid congestion, and ensure the whole fleet operates in seamless harmony. And that's really what we all want, isn't it?

Tomi Tuulkari

Tomi Tuulkari works as Senior Manager, Product Management in Kalmar. Tomi’s expertise comes from delivering multiple Automatic Stacking Crane projects, extensive customer collaboration as well as from Product Management, i.e. ensuring that the developed yard cranes fulfill the needs of the industry and that they can be maintained easily during the product's lifetime. Today Tomi focuses on the product management tasks dedicated to Kalmar One. He spends his time communicating with Kalmar's customers and partners, steering product development, as well as designing and scouting new technologies that can help customers operate automated sites even more smoothly while increasing productivity.