Session: 06-07-01 Metocean, Measurement and Data Interpretation

Submission Number: 156597

A Systematic Data Preparation Approach for Analyzing Hotel Electrical Power Consumption on Passenger Ship 

In recent years, an increasing number of ships have been equipped with sensors and monitoring devices to track power consumption across various onboard hotel systems, resulting in a significant increase in the volume and availability of operational data. However, due to equipment faults, transmission errors, sensor malfunctions, and environmental interference, such as vibrations and harsh weather conditions, the operational data may contain erroneous data points that are critical to assess and address prior to conducting data analysis. If these issues are not addressed, they can undermine the accuracy of the analysis and limit the reliability of the insights derived from the data.

In this paper, a systematic approach to data preparation for analyzing the electrical power consumption of hotel operations onboard a passenger ship is presented. This approach addresses the unique challenges posed by a complex dataset containing power consumption information of various onboard hotel systems, weather data, and operational parameters. We employ comprehensive missing data imputation techniques such as K-nearest neighbor (KNN) imputer, backward/forward fill, and linear interpolation, tailored to the specific nature of missing data within the dataset. The data preparation strategy also includes outlier detection, data smoothing, feature engineering, and normality tests to guide appropriate correlation analysis, with the goal of identifying relationships between power consumption and other parameters within the dataset for effective feature selection.

The final result is a dataset free from distortions and unwanted anomalies yet preserving the integrity and characteristics of the original data and ensuring high data quality without compromise.

Keywords: Passenger ship, Data preprocessing, Data preparation, power consumption, Ship hotel systems

Presenting Author: Adanna Okonkwo Aalto University

Presenting Author Biography: Adanna Okonkwo is a dedicated researcher passionate about enhancing energy efficiency in the maritime industry and contributing to a low-carbon future. Her commitment to this field stems from a deep concern about climate change and its potential impact on our planet. Adanna believes we are approaching a critical tipping point where the effects of climate change could become irreversible, putting the well-being of current and future generations at serious risk. This reality drives her to explore innovative and sustainable solutions that can help make shipping more environmentally friendly.

Adanna began her career as a technical support specialist, where she gained valuable experience in data analysis and systems management. This role sparked her interest in leveraging technology and data to solve problems within the maritime sector. Her current research focuses on ship energy management, applying data analytics, machine learning, and optimization techniques to enhance the energy efficiency of ship operations and reduce the industry's carbon footprint.

She holds a Bachelor’s degree in Electronics and Computer Engineering, with a major in Telecommunications. She also earned a master’s degree in Maritime Affairs, specializing in Maritime Energy Management, from the World Maritime University in Sweden, where she graduated with distinction and received the award for best student in her specialization. Currently, she is pursuing her Ph.D. in Marine and Arctic Technology at Aalto University, where she continues to explore innovative ways to make shipping more sustainable.