Advancing maritime performance analytics
at the intersection of data science and naval architecture
What we do
Vessel Performance Evaluation Based on Noon Reports
Navalytica uses a combination of naval architecture principles with statistical theory and robust machine learning algorithms to optimize the performance of the maritime vessels. We are able to utilize Noon Reports data for ship performance optimization initiatives, and provide the contractual stakeholders of the shipping industry with accurate objective vessel performance evaluation metrics.
Our Method
A key innovation of Navalytica’s patented method is that it is built on a novel proprietary closed-form mathematical expression that accurately predicts the shaft power demand and fuel consumption of a ship.
Our solution is based on a mathematical model that accurately predicts the power demand of a self-propelled full-scale vessel sailing under any operational speed, draft, trim, or weather conditions, even with limited vessel operational data availability, or without any previous information of the particulars of the vessel.
By working with Navalytica you are able to
- Save fuel and comply with various maritime regulations by detecting inaccurate data entry or possible fraud in noon reports data.
- Initiate performance benchmarking of vessels and crews to promote best practices.
- Define optimal hull cleaning and propeller polishing schedule.
- Quantify the ROI of the strategic performance initiatives (hull coating, PBCF, Wind Assisted Propeller Systems such as Flettner rotors, wing sails, suction sails, kites, etc).
- Determine the optimum trim for any given set of sailing conditions.
A cut above the competition
Current state of the art vessel performance evaluation methods quantifying fuel consumption produce a 5-10% error due to the data bias and resulting overfitting. The method developed by Navalytica minimizes the error inherent in the current system of Noon Reporting to 1-2% by utilizing the closed-form proprietary equation before applying the machine learning methods.
Do you think that these numbers are unsubstantiated?
Who is manipulating and interpreting your data?
"…the way IA techniques work these days, are driven by the data, and you are as good as the data that you are fed; and detecting bias in the data itself is actually one of the more important research and technical challenges…"
Ruchir Puri, Chief Scientist, IBM Research; IBM Fellow, 2016-19 “IBM Watson” CTO
2018 Isaac Asimov Memorial Debate: Artificial Intelligence
Who we are
Javier Zamora
Javier is the Founder and the Chief Data Scientist behind Navalyitica´s revolutionary method. Having sailed as a marine engineer as a graduate of both the Universitat Politècnica de Catalunya and the Universidad Politécnica de Madrid, ever a curious soul, Javier pursued the Master´s Degree in Naval Architecture at the prestigious Stevens Institute of Technology, where he received the full research scholarship. Javier´s rich work experience in maritime engineering consulting, cruise, shipping and maritime software start-up companies has sparked his interest in creating a reliable method of vessel performance evaluation. To help him solve the all-consuming problem of quantifying the vessel performance, Javier received the M.A. in Statistics degree from the Columbia University in the City of New York. As a result of that (and many years of sleepless nights), his method and the proprietary closed-form mathematical equation were born.
Javier´s publications and patents currently include
J. Zamora. Obtaining and Utilizing Power Demand data of Self-Propelled Vehicles. U.S. Patent Application No. 17/225,019. U.S. Patent and Trademark Office. 2021.
J. Zamora. A Propulsive Power Demand Predictive Model for Self-Propelled Vessels. Ocean Engineering. 2021 (under peer-review)
J. Zamora. Accurate Vessel Performance Quantification using Noon Reports Data Collection Systems. 6th Hull Performance & Insight Conference (HullPic), Pontignano, 2021