The design of a container terminal is a process that goes through a number of phases. One of the first is the concept design phase, which is needed to make a first assessment of the project's technical and financial feasibility. This phase is exploratory of nature, because of the early stage in the design cycle and consists of several tasks that often dependent on certain design choices. One of the most influential choices in the early stages of the design cycle is the stack equipment choice. It has a significant influence on the two primary deliverables of a concept design: the land use and the cost estimate.
The problem definition holds that in practice, there is often only limited time available for the concept design phase. The short duration of the process affects the design effort by engineers in two ways. First, not all possible stack equipment options can be considered in detail, which generally is solved by an expert judgement based design freeze early in the process.
And secondly, not all designs that are considered are visualised, as often only the preferred design option is visualised at the end of the process. A quick visualisation would help to better understand the design itself, and the interaction with its environment.
An important consequence of the above work method is that not all potential solutions are being assessed throughout the design process. This limits the adaptability of the design process and can potentially lead to suitable solutions being overlooked. To improve the design process, more stack equipment types should be considered and evaluated throughout the process, and all considered design options should be visualised. This is as yet not possible in the limited time available time in the concept phase with the currently available tools. This problem leads to the following research question:
What are the consequences of accelerating the generation and visualisation of concept level terminal designs, by modelling the automatable tasks, on the concept design phase?
A typical concept design process, as defined in this report, consists of six stages and has an indicative duration of four to eight weeks, depending on the exact outline of the project and the local conditions. The tasks that correspond to the generation of the various terminal concept design options (i.e., concept level calculations, layout generation, cost estimation and visualisation) are considered to be automatable. An examination of currently existing design tools demonstrated that there is currently no available tool that meets the requirements and therefore, a specific design tool must be developed. Parametric engineering is the chosen method for automating the concept level design process as it allows for different solutions to be explored and provides for flexibility during the process. Based on both expert interviews and studied literature, it has been decided to build the design tool using a combination of two packages in which Python is used for concept level terminal calculations and Grasshopper is used for layout generation. The developed tool can calculate the required terminal elements (e.g., storage capacity, quay length, equipment numbers), arrange these elements into a layout, make a cost estimate and instantly produce a 3D visualisation of the corresponding terminal concept design.
A case study is used to validate the output of the tool and to demonstrate the tool's ability to evaluate all types of stack equipment in parallel throughout the process. During this research, the design tool was used exhaustively for a wide range of terminal designs, of which the required durations were logged. In the end, this averaged to an estimated reduction in the required time for the concept phase of around half of the original four to eight weeks. It should be noted that this number is indicative and that many factors influence this number, such as the complexity of the project, the amount of 'tailoring' required and the experience the terminal planner has with using the tool. Nonetheless, based on these findings, it can be concluded that the tool allows for evaluating more options in less time.
To further explore the tool's capabilities, the design tool is used to investigate the effect of four prominent local cost parameters (i.e., cost of land, labour, fuel and electrical power) on each of the considered stack equipment options. The results show that the cost of labour has the most significant influence on the terminal's cost estimate, but that the most significant discrepancies between the stack equipment options are observed when increasing the cost of land. Although the cost of labour and land are considered to have a significant influence on the concept phase, the results demonstrate that for a medium-sized container terminal located in Western Europe, the most economical stack equipment option is the RTG, regardless of the influence of the examined local cost parameters.
Based on the results from the case study and the exploratory research, the tool's impact on the concept design can be described as follows: firstly, the tool is able to consider all potential stack equipment types in parallel throughout the process. This new working method, therefore, enables a design freeze much later in the process than before, providing improved flexibility as changes can be anticipated throughout the concept phase. Furthermore, the time saved as a result of the design tool can now be used for more extensive expert judgement throughout the concept phase. The more time available for expert judgement, the better the terminal planner can assess the solution space of suitable design options and improve the 'fit' with local conditions. Finally, instant visualisation of the considered design options creates the ability to obtain a better understanding of the design itself and the interaction with its environment.
To conclude: the developed automated design tool is able to accelerate the generation and visualisation of concept level container terminal designs and thereby evaluate more design options in less time. This newly established working method is able to improve the concept design phase in three ways: i) the design process is a lot more flexible and allows for all stack equipment options to be considered during the process, ii) the time-savings enable more time for extensive expert judgement throughout the concept phase, and iii) the instant visualisation of all potential options provides the terminal planner with the ability to better assess the terminal design itself and the relation to its surroundings.