Scalable In-Database Machine Learning for the Prediction of Port-to-Port Routes

The correct prediction of subsequential port-to-port routes plays an integral part in maritime logistics and is therefore essential for many further tasks like accurate predictions of the estimated time of arrival. In this paper we present a scalable AI-based approach to predict upcoming port destinations from vessels based on historical AIS data. The presented method is mainly intended as a fill in for cases where the AIS destination entry of a vessel is not interpretable. We describe how one can build a stable and efficient in-database AI solution built on Markov models that are suited for massively parallel prediction tasks with high accuracy. The presented research is part of the PRESEA project (“Real-time based maritime traffic forecast”).

Re-routing using Contraction Hierarchies in Software-Defined Networks

According to the Open Networking Foundation (ONF), one of the reasons to reexamine traditional network architectures is the increment of mobile devices and its data transmission. The global IP traffic forecast by CISCO estimates an overall traffic increase to 396 exabytes per month in 2022, more than three times the traffic on 2017 (122 exabytes per month). In this work, we research the similarities between vehicular networks and computer networks. These similarities will allow us to implement the Contraction Hierarchies algorithm (CH) in computer networks. CH is an interdisciplinary algorithm from vehicular networks which can provide us with the elements and logic to optimize specific routing problems in computer networks. In order to implement CH, we use Software Defined Networks (SDN). SDN is a computer networks paradigm that separates the Data and Control planes. The Data plane is left to the network devices to distribute the packages, and the control plane is centralized into a Controller. By having a controller with a broad view of the network, we implement CH in order to optimize route selection. Once the route is determined, we study the possibility of using the advantages of CH to redistribute traffic in case the network elements suffer from unforeseen circumstances.

Passengers' Traffic Forecast of the Nigeria Airports using the Holt-Winters Additive Model

This research work was carried out in response to the need as a result of increase in Nigeria population and the demand for air transport facilities, this research was carried out using the data of the Federal Airport Authority of Nigeria, which has a total number of 23 airports out of which four are international airports, seven are domestic airports and twelve are other domestic airports with the total number of 75,879,653 passengers between Jan. 2003 and Dec. 2011. The passengers' traffic of FAAN's data of 2003 to 2011 was collected and forecasted using the NCSS computer package to generate the Holt-Winters additive model with coefficient of determination, R2 of 90.99% and the models were used to forecast for the years 2012 to 2019 using the models.