Identify bottlenecks before they arise
How we use CHESSCON Shift Preview and AI to make shift planning in container terminals more predictive
Container terminals are under constant pressure to be more efficient. Even small planning errors in a shift can cause high costs – for example, if too few vehicles are available at a critical point. This is exactly where CHESSCON Shift Preview comes in: as an operational simulation that calculates a planned shift and makes possible bottlenecks visible before real operations begin. In this article, I show how this approach can be extended by AI to further improve productivity at the terminal based on data.
By Oliver Jelsch
Why shift planning in container terminals is so challenging
Shift planning must bring together the expected volume, the available equipment and personnel capacity as well as the specific processes at the quay and yard on a shift-time-basis. Here, the Terminal Operating System (TOS) classically offers preliminary planning, which is checked and finalized by experienced yard planners before the start of the shift.
Nevertheless, unexpected bottlenecks occur again and again in everyday life. We looked at how we can further improve shift planning with Shift Preview and AI. Because of the many target and actual data on the layers, we have an enormously large, structured amount of data that we want to use.
What Shift Preview does in terminal operations
First of all, let's talk about the tool itself: CHESSCON Shift Preview complements the existing TOS as a forward-looking check instance. For this purpose, the planning data is taken from the main system and recalculated in an operational simulation. The container terminal will be digitally mapped; the stored algorithms are closely based on the logics of the TOS. This makes it possible to simulate before the start of the shift whether bottlenecks or overloads are likely at certain points in the terminal.
The practical benefit lies in timing: According to the transcript, the simulation takes only a few minutes, even with larger terminals. Planners can check the results before the end of the working day, adjust the planning and only then hand it over to operational use. This makes Shift Preview a tool for daily operational decision preparation.
Operational simulation instead of pure retrospective
Shift Preview thus shows where problems are likely to arise in current planning – for example, at intersections, at individual container gantry cranes or in certain yard areas. This procedure can be described as an operational simulation. While strategic simulations tend to look at long-term scenarios or non-existent terminal configurations, Shift Preview docks directly onto operational reality and simulates a specific layer in advance using real data from the TOS.
Why the TOS doesn't just optimize itself
An obvious question is: If the TOS plans anyway, why doesn't it optimize the layer directly itself? Why is a tool the CHESSCON Preview necessary? This is because the system is based on parameters and basic logics that it can only question to a limited extent. Therefore, an additional, upstream simulation with a separate system creates enormous added value. It looks analytically at the same data and makes it visible where the existing planning reaches its limits.
Where AI could start in container terminals in the future
Now not only individual shifts can be simulated in advance. The results of many backlogs with their respective plan and real data can be merged and evaluated using AI methods. The goal: to make patterns visible that are easily overlooked in day-to-day business. In this way, structural weaknesses in the operating model can be made visible over several weeks and months.
Such an approach could, for example, recognize that bottlenecks occur regularly when several preconditions come together, such as a bottleneck at a certain container gantry crane at a certain time. This would be valuable for terminal operators because it could be used to derive strategic causes – such as unfavourable traffic routing or unsuitable surface logic. The AI could also use the knowledge over many shifts to make better suggestions for the next shift planning.
Our approach for enhancement: Testing AI and Shift Preview in verifiable steps
In scenarios with AI, it is important to take a realistic look at the actual benefits. We are therefore currently examining how Shift Preview and AI can create concrete added value for terminal operators. In a test project with a customer, we deliberately proceed in small, validatable steps. Currently, we are starting where AI is demonstrably strong: recognizing recurring relationships in large amounts of data and highlighting anomalies. Only when the AI has been proven to work here do we follow it up with AI tests for interpretations and recommendations for shift planning.
AI approach not as an end in itself, but with a view to productivity
We are convinced that Shift Preview and the targeted use of AI can be used to develop concrete tools for improving operations that are also economically viable. We see AI as a concrete tool to make operational and medium-term strategic planning of shift planning more robust and faster.
