Paid SpecialistCertificate

From Physics to Models

Building a predictive model of a ship that someone should be willing to act on: what physics contributes that data cannot, which variables are inputs and which merely emerge, why a time-series split is the honest test, what a neural network buys and costs, and how to attach an uncertainty that survives contact with a decision.

Lessons
8
Quiz questions
67
Exercises
1 (10 min)
Calculators
1
Time
About 96 min
Price
€189+ VAT
Certificate
Graded final exam, then a PDF certificate with a number anyone can verify
Who it's for
Analysts and engineers who build, buy or audit performance models: anyone deciding what goes into a fuel-consumption prediction, and anyone who has to say whether its output can be trusted with a routing or maintenance decision.
Assumed knowledge
The three preceding courses in this specialisation, or equivalent. This course assumes you can describe a dataset, clean it defensibly and read its dependency structure, because a model is only as good as those three steps made it.
Taught by

Christos Papandreou

CEO & Founder, Blue Autonomy

  • Naval architect and marine engineer, National Technical University of Athens
  • 15 years of experience in fleet performance and operational data
  • Chief Scientific Officer, NEMO, previously DeepSea Technologies, StormGeo, Athenian Sea Carriers, PwC

Part of the Performance Data Science specialisation.

What you will be able to do

By the end of the course. Each outcome names the lessons that deliver it, and the final exam samples every one of them.

  1. State what a physics-based model contributes that a purely data-driven one cannot, where each approach is stronger, and the hybrid most performance work actually needs. Lesson 1
  2. Distinguish commanded from emergent variables in the coupled ship-propeller-engine system, and use that distinction to rule candidate model inputs in or out. Lessons 2, 4
  3. Write the physics chain from resistance to fuel rate term by term, locating each operational lever on it and the terms that carry the uncertainty. Lesson 3
  4. Classify candidate inputs as known, forecast, emergent or latent, handle latent states through documented proxies, and engineer features that carry the physics rather than the raw channel. Lesson 4
  5. Fit a multiple linear regression as the honest baseline, diagnosed from its residuals rather than its R² and with multicollinearity recognised, and judge any model on a time-ordered train/test split, naming the leakage a random split creates. Lessons 5, 6
  6. Weigh a neural network's flexibility against its costs in extrapolation, auditability and data appetite, and judge whether a specific decision should be taken on a model's output, using the asymmetry of the payoffs and the being-right-for-the-wrong-reason test. Lessons 7, 8

A ship's fuel consumption is governed by physics that has been understood for a century, and predicted by models that are usually fitted without any of it. Both approaches work, in different regions, and the interesting engineering is in knowing which region you are standing in.

This course builds a predictive model and then attacks it. What is inside: what a physics-based model contributes that no amount of data supplies, above all the ability to extrapolate and to say why something changed; the ship, propeller and engine as one dynamic system, and the resulting distinction between variables you set and variables that merely emerge, which decides what may legitimately be a model input; the energy chain from tank to thrust, written out so that every term is a place where uncertainty enters; linear regression as the baseline any complicated model must beat; the train-and-test discipline, and why a random split of time-series data flatters every model that receives it; neural networks, what their flexibility buys and what it costs in auditability and extrapolation; and finally the question the whole specialisation has been building toward, which is not whether a model is accurate but whether a specific decision should be taken on its output.

The opening lesson is free to read as a sample. By the end you should be able to specify a performance model, defend every variable in it, test it in a way that will not flatter it, and tell a technical director in plain language what its number is worth.

What the course covers

  1. Two Ways to Predict: Physics and Data 9 min
  2. The Ship as a Dynamic System: What You Set and What Emerges 9 min
  3. The Physics Model of Consumption, End to End 10 min
  4. Choosing and Building the Inputs 9 min
  5. Linear Regression as the Honest Baseline 26 min + exercise
  6. Train, Test, and the Generalisation Trap 14 min
  7. Neural Networks: What They Buy, and What They Cost 9 min
  8. Trusting a Model: Uncertainty and the Decision It Feeds 10 min

€189 + VAT

€234.36 including 24% VAT, which is what an individual in the EU pays. Outside the EU no VAT is added. A company in another EU country pays the net price once its VAT number is confirmed in VIES, the EU's register; a company in Greece pays the 24%. Checkout shows the total before you pay.

Paid once, for one person, for as long as the Academy runs: every lesson, the Excel workbook, the final examination and the certificate. Part of the Performance Data Science specialisation, which costs less than its courses bought one at a time.

Buying online needs JavaScript switched on in your browser. Otherwise write to info@blueautonomy.gr and we arrange access.

All prices · Refunds: 7 days, unless you have started

Common questions

What exactly is included?

8 lessons (about 96 minutes in total), 67 quiz questions with explained answers, and 1 working Excel calculator: Speed-power regression.

How much does it cost?

€189 plus VAT, paid once: €234.36 with 24% VAT, which is what an individual in the EU pays; outside the EU no VAT is added, and a company in another EU country with a VAT number valid in VIES pays the net price. That buys every lesson, the Excel workbook, the final examination and the certificate, for one person, for as long as the Academy runs. The course is also part of the Performance Data Science specialisation, which costs less than its courses bought one at a time. Everything else is on the pricing page.

Is there work to do, or is it only reading?

1 lesson carries a required exercise, about 10 minutes of hands-on work in total. You download that lesson's Excel workbook, follow the numbered steps against the numbers the lesson states, then change one input and watch the answer move. The exercises are where the method stops being something you have read and becomes something you can run; the stated course time includes them.

Do I get a certificate?

Yes. The course ends with a graded final exam, marked on our server rather than in your browser: you see your score and which questions you got wrong, but not the answers, because the questions are drawn from a pool. Pass it and you get a PDF certificate in your name, with a certificate number on it.

That number is the point. Anyone you show the certificate to can check it themselves at academy.blueautonomy.gr/verify, with no account and no email to us: the page states the name, the course and the date we issued it. A certificate nobody can verify is a decorated PDF, and we would rather issue one that stands up.

Can I preview the course?

Yes: the first lesson, Two Ways to Predict: Physics and Data, is free to read in full. It is a real lesson, not a trailer: same depth, same quiz.

Do I need prior knowledge?

The three preceding courses in this specialisation, or equivalent. This course assumes you can describe a dataset, clean it defensibly and read its dependency structure, because a model is only as good as those three steps made it.