IoT for ships – An untapped data resource

Kalmar 210412

Linnaeus University Centre for Data Intensive Sciences and Applications

IoT for ships – An untapped data resource

Updated version from feedback.

Background

The project is the first step to study the access and exploitation of new data resources in the domain of IoT in the shipping industry. In order to reduce the energy consumption on board, measurements are needed to make the right

decisions. Today, there is a need for being able to collect data from older ships not equipped with sensor technology. The vast majority of all ships today are of an older standard, and there are both technical challenges and costs involved in leveraging on both the existing data on board as well as installing new sensors.

A ship consists of many mechanical components that together create a

very complex energy system [1]. The goal is to design, develop, and install IoT solutions that can tap into the vast amount of available physical sensors that already exist, or can be installed on board. The project will be implemented as a cross collaboration between the Department of Computer Science and Media Technology, Kalmar Maritime Academy, and the external IoT company Sensative AB.

During the project we will:

● Installation of IoT solutions on the Kalmar Maritime Academy´s ship Calmare Nyckel

● Install IoT solutions on the Kalmar Maritime Academy´s ship Calmare Nyckel

● Provide full access to data to researchers and students via API

● Perform a benefit-cost analysis of exploiting the developed IoT solutions

● Apply for additional relevant funding (Lighthouse, Vinnova, Transportstyrelsen)

● Promote the IoT technology and platform within other faculties to utilize

IoT and real-time data in upcoming research application

Problem definition and value – An untapped data resource

Internet of Things (IoT) is rapidly being implemented and scaled up in the real estate, industry, smart cities, smart farming etc., where IoT is an important tool to a data-driven business and society [2], [3]. IoT and real-time data is an important data source for future Artificial Intelligence/Machine Learning, Big-data and scientific research to improve productivity, efficiency in a competitive landscape – however the shipping & maritime industry has not yet adapted IoT technology in the same pace [4]. One particular challenge is the metal structure limiting the RF range of wireless communications. There are also several challenges in making ships more energy efficient, and methods with Big Data analytics and machine learning requires data [5]–[7]. A ship has a couple of challenges compared to common open spaces, buildings, and cities:

● Legacy technology has not yet fully adapted to the need for open interfaces/API.

● The metal structure of the ship makes wireless communication more challenging

● The connectivity to cloud goes through an unreliable and low-speed satellite communication

Due to the technical limitations, today research within the maritime industry is limited to the data sources and systems available today. IoT provides a new and effective tool to get access to more data with high quality and can also be provided in real-time opening opportunities for far more intelligent decision-making systems.

Yet, there is no simple solution, thus there is a need for a local testbed to support collaboration between different sectors that can find and test IoT solutions for the maritime industry specifically. By bringing together the need owner (maritime industry) with external competence and resources within the university (from computer, data science to maritime faculty with researchers and students) and industrial expertise within IoT/IT. In the project we will use and IoT platform that can integrate to existing legacy technology (OT) & IoT technology, providing access through standardised API. The same platform and technology is already in use by Scandinavian municipalities, real-estate owners, agricultural – thus the same API and technology can be used for research purposes in fields outside the maritime faculty.

Research problem

Today, there is research in the maritime sector for developing digital twins, but there

are no ready to use solutions for the existing fleet of ships [8], [9]. Also, the means

of creating digital twins on ships based upon existing operational data combined

with physical equations, rather than only physics based simulations is an emerging

area [10], which requires gathering new data.

different stakeholders. As there are many different systems on board a ship (this applies also to land based industries) research is needed to combine different data sources and create a common format.

Objectives

Establish a IoT testbed for ships, in Kalmar for the ship Calmare Nyckel, managed by the Kalmar Maritime Academy:

● Access to hardware for rapid prototyping.

● Access to API and real-time data and database access, for research and analytics in the DISA group

● Apply for additional funding and resources together with a partner network to extend the IoT testbed in Kalmar (Kalmar Energi has shown interest to further develop the testbed)

● Utilize the IoT platform and API as part of the summer IoT-course where students will be able to test and evaluate their hardware prototype or applications (apps)

Expected results

● IoT data from the maritime industry will be used in new research

applications, aimed specifically for the maritime industry stakeholders and funding institutions such as Lighthouse and Transportmyndigheten.

● Interest from the maritime academy to further develop the IoT testbed for research purposes

● A draft proposal for a full fledged follow-up project Consortium

● Fredrik Ahlgren, Faculty of Technology, LNU – Project leader for the IoT testbed

● Björn Pundars, Kalmar Maritime Academy, LNU – Need owner for the Maritime faculty & industry

● Håkan Lundström, Sensative AB (External) – IoT expertise and SW provider for the IoT installation.

Activities and time plan

● March 17th: Launch of IoT testbed at “IoT Meet Up”.

● March 31st: Selection and sourcing of IoT hardware.

● May 31st: Deadline for installing IoT hardware on the ship.

● December 31st: Deadline for consortium to apply for additional funding.

Budget

● Hardware: 50 kKr (IoT gateways, sensors, for rapid prototyping). External, Sensative AB.

● Software deployment, configuration, installation and training (IoT platform): 5 kKr, External, Sensative AB

● Work time, Internal LNU, researchers and technicians/students o Fredrik Ahlgren, 25 kKr. Project management, initiating

applications, sharing results within DISA-group.

o Technicians/students. 20 kKr.

● Work time, External Sensative. 55 kKr o Writing application for funding, Total budget: 155 kKr

- from DISA: 100 kKr - Sensative AB: 55 kKr

The funding from DISA is allocated for the Faculty of Technology, Sensitive will cover their own work time in the project. The hardware & software (IoT platform) will be accessible for all faculties at the university and students interested in testing the IoT hardware/service for testing purposes after approval by the project leader (Fredrik Ahlgren).

Fredrik Ahlgren Senior Lecturer

[1] B. Pena, L. Huang, and F. Ahlgren, “A Review on Applications of Machine Learning in Shipping Sustainability,” Sep. 2020, Accessed: Mar. 03, 2021.

[Online].

[2] B. S. Chaudhari and M. Zennaro, “Introduction to low power wide area networks,” in LPWAN Technologies for IoT and M2M Applications, Elsevier, 2020, pp. 1–13.

[3] D. Weyns, G. S. Ramachandran, and R. K. Singh, “Self-managing Internet of Things,” in SOFSEM 2018: Theory and Practice of Computer Science, vol.

10706, A. M. Tjoa, L. Bellatreche, S. Biffl, J. van Leeuwen, and J.

Wiedermann, Eds. Cham: Springer International Publishing, 2018, pp. 67–84.

[4] F. Ahlgren and M. Thern, “Auto Machine Learning for predicting Ship Fuel Consumption,” presented at the ECOS, GUIMARÃES, Jun. 2018.

[5] F. Ahlgren, M. E. Mondejar, and M. Thern, “Predicting Dynamic Fuel Oil Consumption on Ships with Automated Machine Learning,” Energy Procedia, vol. 158, pp. 6126–6131, Feb. 2019, doi: 10.1016/j.egypro.2019.01.499.

[6] C. Gkerekos, I. Lazakis, and G. Theotokatos, “Machine learning models for

predicting ship main engine Fuel Oil Consumption: A comparative study,”

130, pp. 351–370, Jan. 2017, doi: 10.1016/j.oceaneng.2016.11.058.

[8] S. S. Johansen and A. R. Nejad, “On Digital Twin Condition Monitoring Approach for Drivetrains in Marine Applications,” Volume 10: Ocean Renewable Energy, Jun. 2019, doi: 10.1115/OMAE2019-95152.

[9] F. Perabo, D. Park, and M. K. Zadeh, “Digital Twin Modelling of Ship Power and Propulsion Systems: Application of the Open Simulation Platform (OSP),”

p. 6.

[10]S. O. Erikstad, “Merging Physics, Big Data Analytics and Simulation for the

Next-Generation Digital Twins,” p. 12.