Table 5 Overview of AI solutions to tackle port challenges
Title of AI solution | Description | Source |
|---|---|---|
Smart waterway | Developing autonomous barges in urban waterways with low-cost sensor systems Detecting and localizing obstacles in the waterway Finding the optimal path by a navigation agent localizing seamless handovers, even blind spots (e.g., under bridges) Using computer vision and sensor fusion technology | IDlab Antwerp (2021) |
Demand prediction | Predicting demand for better matching with supply Using reinforcement learning as a technology | |
Resource allocation | Optimizing gate allocation in a warehouse to reduce the distance Using reinforcement learning as a technology | |
Optimized shelf placement | Optimizing shelf placement in the warehouse Using reinforcement learning as a technology | |
Resource-efficient AI | Designing computational power- and resource-efficient autonomous robotic platform for an industrial warehouse Improving human worker job satisfaction and safety Using ML as a technology | |
Data-driven control | Controlling chemical plants, processes, systems, robotics, ships Employing reinforcement learning as a technology | |
Lock optimization | Navigating and scheduling solutions for inland waterway transport Predicting ETA and accurate vessel positioning Using ML as a technology | |
Large-scale simulation | Simulation-based testing of large-scale Internet of Things applications Focusing on efficiency and accuracy Simulating operations of various port-related activities Using digital twin as a technology | |
Control flexibility in industrial processes | Controlling algorithm to characterize Reducing CO2 emissions to bed for the overall industrial and energy ecosystem Employing deep learning and the Markov decision process as a technology | IDlab Gent (2021) |
Applied building photovoltaics | Improving the energy yield prediction of solar panels Developing techniques for predictive maintenance of solar panels Reducing the overall carbon footprint Employing neural networks as a technology | |
Electric vehicle charging | Reducing the port's carbon emissions Shifting from fossil fuels to electricity Reducing energy consumption by smart charging Employing reinforcement learning and the Markov decision process as a technology | |
Truck guidance system | Optimizing engine trading off local versus global cost functions Optimizing logistic flows when global information is available Using ML as a technology | |
Predictive planning | Optimizing statistical forecasting Forecasting traffic conditions for trucks to optimize the planning Employing ML as a technology | Carlan et al. (2019) |
Booking of slots | Optimizing matching of free slots in container terminals Making faster handling time Employing ML as a technology | Carlan et al. (2019) |
Optimized maintenance scheduling | Predicting accessibility for offshore assets taking context, weather, vessel, routes Detecting anomaly, semantic stream reasoning, rule mining Using neural networks as a technology | IDlab Gent (2023) |
Boat landing | Predicting accessibility for offshore assets taking context, weather, vessel, routes Employing neural networks as a technology | IDlab Gent (2023) |
Detection of fouling using AI | Modeling the performance of a ship Preventing extreme fouling on the hull and propeller of the ship Reducing fuel consumption Using ML as a technology | Gillis et al. (2017) |