PILOT : VALENCIA SEA PORT AS INTELLIGENT LOGISTICS HUB

A port terminal could be depicted as a hub process where the clients are inputs/outputs from different transportation models: vessel, train, and truck. The main performance goal for a port terminal is to provide the shortest possible time to leave or pick up containers at the lowest possible cost. This time-cost ratio is an overall indicator to measure performance that is transversal for the domain. The most widely used Key Performance Indicator (KPI) is the Truck/Train/Vessel Turnaround Time (T/VTT): the time a specific means of transport spends in the terminal to fulfil an order. At this point turnaround time is known in isolation, i.e. in a specific terminal, and calculated after the order is processed. The integration of Big Data of several stakeholders involved into the containers management process will provide better insights and KPIs about the overall efficiency at both port and terminal levels. The defined KPIs provide benchmarking capabilities (e.g. related to costs and performance) that may indicate different levels of competiveness.

Furthermore, harvesting operational data from sensors placed inside machines makes it possible to monitor failures of specific machine parts. One of the most important aspects in the improvement of operational efficiency is the anticipation of possible failures and, thereby to avoid unscheduled production downtime. These random stops reduce production throughput, increase unscheduled downtime, and lead to a reduction in production time optimization, resulting in the need for more resources to reach the required productivity levels.

By processing Big Data efficiently in the prediction algorithms, unscheduled production downtime can be prevented by providing the necessary information indicating whether a machine element must be repaired or replaced. This improves the planning and execution of maintenance tasks and enhances the continuity of the operations. All scheduled planning within a production process improves the productive efficiency and reduces the necessary resources to keep the production.

This developed in the domain of intelligent ports as part of the Transforming Transport project have the opportunity to assess the potential of merging, processing, and accessing data by different stakeholders as they need it. The pilot will be deployed in Valencia Port, and specifically in the Noatum Terminal: one of the leading terminals in the Mediterranean Sea according to the number TEUs managed. The pilot will provide a set of predictive models to support both a predictive maintenance strategy and the development of a terminal productivity cockpit. These models will provide insights regarding trends related with container traffic and potential maintenance alerts.

The objectives of the pilot are:

• Improve the crane scheduling by calculating the optimum sequence of crane movements, an thus, reducing the delay to pick-up/deliver a container
• Prevent unexpected failures of crane spreaders thanks to the application of predictive maintenance techniques.
• Collect and calculate and predict relevant indicators to the port and terminal stakeholders, both (i) providing real-time observation of the status of port operations and (ii) providing the necessary insights to improve the yard planning

• The implementation of an optimization algorithm that provides the user with the best sequence of crane movements, taking into account the real monitoring of a crane.
• The development of a set of predictive maintenance models to cranes’ spreaders, starting with the study and deployment of a set of sensor devices to gather information regarding the spreaders of the STS cranes. The current preventive maintenance strategy will be enriched with a prognosis model that will alert of possible failures as they are detected.
• An advanced dashboard for better decision-making, including a predictive decision support system that considers all the historical data available at the Port and Terminal levels. Registered and planned accesses, environmental data, and information coming from the TOS and SEAMS Platform will provides the basis to create useful indicators and model their next-future trends. This dashboard will provide relevant indicators in order to visualize trends and enact relevant information for the yard planning, such as behavior patterns of container arrivals.

The main KPIs of the pilot are:

• The reduction of the Terminal Truck Turnaround (TTT). The TTT is defined as the time a truck spends in the terminal yard to fulfil one or several transactions, such as load/discharge a container. Currently a TTT threshold in minutes is established, so the goal is that the 95% of the trucks accomplish such threshold.
• The reduction of the Port Truck Turnaround (PTT). This KPI is similar to the previous one but it is made up of the time to arrive to the terminal from the port gates and the time to exit from the port. This KPI provides an indirect view of the waiting time in the Port Access gates. Similarly, our goal is that 95% of the trucks meet a waiting time threshold.
• Reduce the maintenance cost defined as the sum of labour costs related with a specific failure in a crane and the cost of the crane downtime. Each maintenance report states the number of hours (labour) to fix a specific failure of a machine, whereas the downtime will be reported as the cost/hour of a working crane