Jake Walker & Iliya Valchev
The IEEE OES Strathclyde SBC (Student Branch Chapter) was delighted to invite Michele Sancricca – Head of WW Technology for Transportation and Logistics at Amazon Web Services (AWS) – to present his take on the next data revolution awaiting the Maritime Industry.
At the Strathclyde SBC, we have taken advantage of the new status quo for event planning and moved our presentation series Talks at NAOME (which is now in its 3rd year) fully online. The Naval Architecture, Ocean & Marine Engineering Department at the University of Strathclyde has previously hosted speakers from more than 20 organizations and academic institutions, including BMT on the Application of the Digital Twin and ABS International on Cybersecurity and Big Data in the Maritime Industry.
This time around, Michele – a retired Lieutenant Commander with 12 years in the Italian Navy – is our first guest from the USA. He was keen to present his opinions on technology innovation from the perspective of the world’s most comprehensive and broadly adopted cloud platform.
His presentation “How Cloud Technology can transform the Maritime Industry, reduce costs and cut emissions” outlined one of the biggest problems currently facing the industry and how AWS is leveraging state-of-the-art cloud computing to reduce operational costs in shipping.
Michele quickly emphasized the benefits already experienced by maritime companies who already embraced the service-based economy in which AWS thrives. By serving a multitude of clients, operating at both a national and multi-national level, AWS allows more companies to benefit from an ‘Economy of Scale.’ This principle enables AWS to offer faster technology, more secure networks, and better analytics than even the largest shipping companies could organize internally. Put simply, Maritime companies’ time and resources are best spent on transportation of goods, not data processing, storage, and analytics.
“Maritime companies are sitting on a goldmine of data.”
Data storage was outlined as one of the critical reasons why the industry is falling behind with data. The explosion in low-cost sensors now means that Maritime companies are sitting on a goldmine of data. However, many internal databases rely on fully structured Data Silos, which are disconnected and often overlap in purpose. The traditional silos are an expensive method to store data and are usually not designed to process the volume of data generated from shipping (often in the petabytes).
In contrast, AWS is pioneering Data Lakes – which are large, unstructured, and cheap storage platforms for all the data streams generated in the supply chain. This platform paves the way for more advanced data warehouses to sit on top of the data lakes and smaller databases that facilitate the analytics. Michele explains that by collating the data, Artificial Intelligence (AI), and analytics together on the cloud, shipping operators can gain a holistic view of their business without high costs.
Figure 1: Traditional Data Silos vs New Data Lakes
This platform’s advantages have already become apparent to the early adopters, as what was once a business-to-business environment is now closer to a business-to-consumer world. Business clients desire the same experience as, for example, your day-to-day buying experience using e-commerce. They want to know exactly where their shipment is, when they can expect it, and their delivery driver’s first name.
Channeling this information in a timely and secure manner is dependent on real-time analytics powered by AI. The predictive models that fuel the analytics platforms require a subset of AI called Machine Learning (ML). Specifically, Deep Learning (which is a type of ML) uses deep neural networks trained on large volumes of data to untangle the complex interactions hidden within the data and extract the knowledge.
Businesses have become interested in the broad applicability of Machine Learning and are now asking questions about what they can do with ML.
“Can I predict the time of arrival of a vessel so I can plan my landside logistics more efficiently?”
“Can I predict the time of arrival of a ship at the doorstep of my customer?”
“Can I predict the number of containers that I need to process in a port so I can plan my human resources efficiently?”
The answer to all these questions is yes. However, it depends on the available data. There are ML implementations in many fields (including logistics and transportation), but the key is good data – otherwise, there is no point. Above all, having data that is as close to real-time as possible enables a very good alignment between what the models are predicting and what is actually happening on the business side.
ML went from an aspirational technology to mainstream very fast, especially in the maritime industry. This transformation was made possible by adopting cloud-based services, which increased access to computing power and the data for ML projects. ML is now impacting every industry because it is easy to use, whereas companies previously struggled to acquire enough IT power or storage to run big projects. Today 63% of global enterprises are investing to catch up with ML competitors, and total spending in AI is estimated at around $50 billion for 2021.
Michele closed his presentation with a “call to action” for the audience:
“Become a solution builder! Learn, be curious and start becoming familiar with cloud computing because this is probably the best thing to do to really strengthen your résumé!”
This call offered the 50 enthusiastic participants who attended the talk a great way to get involved with cloud technology and was an impactful way to end the 2020 Talks at NAOME series.
The Strathclyde SBC would like to extend their thanks to Michele for his presentation, and we look forward to continuing the series in 2021!
For further information about this topic, or the IEEE OES Strathclyde SBC, please contact Jake Walker or Iliya Valchev {jake.walker}{iliya.valchev}@strath.ac.uk.