Available in English for the first time, this classic and influential book by the late Kohei Ohtsu presents real examples of ships in motion under irregular ocean waves, how to understand the characteristics of fluctuations of stochastic phenomena through spectral analysis methods and statistical modeling. It also explains how to realize prediction and optimal control based on time series models.
In recent years, the need to improve safety and reduce environmental impact in ship operations has been increasing, and the statistical methods presented in this book will be increasingly needed in the future. In addition, the recent development of innovative AI technology and highspeed communications will make it possible to adapt this method not only to ship monitoring and control, but also to any field that involves irregular fluctuations, and it is expected to contribute to solving issues that have been difficult to solve in the past.
Part 1 describes classical spectral method for the analysis of stochastic phenomena. In Part 2, this book explains methods to construct time series models using the information criterion, to capture the characteristics of ship and engine motions using the model, to design a model-based monitoring system that informs navigators operating the ship and managers ashore. Furthermore, it explains statistical control method to design an autopilot system and the governor of a marine engine, while showing actual examples. Part 3 presents the basic knowledge necessary for understanding these topics of the book, namely, the basic theory of ship motion, probability and statistics, Kalman filter and statistical optimal control theory.
Preface
1. Prologue
Part 1: Concept of Spectrum and its Estimation Methods
2. Covariance Function and Spectrum
3. Fourier Transform and its Properties
4. Statistical Estimation of Spectrum
5. Analysis of Linear Systems
6. Cross-Spectrum Analysis of Ship and Engine Motions in Waves
Part 2: Applications of Time Series Statistical Models
7. Continuous and Discrete Time Series
8. Autoregressive Modeling of Ship and Engine Motions
9. Monitoring of Slowly Changing Ship and Engine Motions
10. Analysis and Prediction of Ship and Engine Motions Using State-Space Models
11. Estimation of Trend and Seasonal Adjustment Models and Time-Varying Spectrum
12. Classification of Ship and Engine Motion Time Series
13. On Shore Simulation of Ship’s Motions at Sea
14. Multivariate Autoregressive Model and Statistical Analysis of Ship and Engine Motions
15. Optimal Control of Ships by Statistical Modeling
Part 3: For a Better Understanding
16. Introduction to Maneuvering, Ship Motions, and Propulsion Theory
17. Basics of Probability
18. Kalman Filter
19. Statistical Optimal Control Theory
Bibliography
Index
Biography
The late Kohei Ohtsu was a Professor Emeritus of Tokyo University of Marine Science and Technology. He has served as Vice President of Tokyo University of Mercantile Marine, Captain of the Training Ship Shioji Maru of the University, and President of Ohtsu Maritime Institute, Co. Ltd. His main research interests include statistical analysis of ship motion and optimal steering, ship maneuverability and practical application of ship-to-shore communication and ship operation data monitoring. He received awards from The Society of Naval Architects of Japan, Japan Institute of Navigation, The Society of Instrument and Control Engineers and the Japan Society of Naval Architects and Ocean Engineers, etc. He passed away in 2016 at the age of 72.
Genshiro Kitagawa is a professor emeritus of The Institute of Statistical Mathematics, and of Graduate University for Advanced Study, former President of the Research Organization of Information and Systems, and the former Director-General of the Institute of Statistical Mathematics. His primary research interests are time series modeling, nonlinear filtering and statistical modeling. He is the author of several English books on time series analysis and information criteria. He was awarded the Japan Statistical Society Prize and Ishikawa Prize, etc. and is a Fellow of the American Statistical Association.
