Prof. Ing. Tomáš Hruška, CSc.
RUDNITCKAIA Julia and HRUŠKA Tomáš. Time Series Analysis and Prediction Statistical Models for the Duration of the Ship Handling at an Oil Terminal. In: RELIABILITY and STATISTICS in TRANSPORTATION and COMMUNICATION. Riga: Springer International Publishing, 2017, pp. 127136. ISBN 9789984818863. ISSN 23673370.  Publication language:  english 

Original title:  Time Series Analysis and Prediction Statistical Models for the Duration of the Ship Handling at an Oil Terminal 

Title (cs):  Analýza časových řadů a Statistické modely predikce pro dobu zpracování lodí na ropném terminalu 

Pages:  127136 

Proceedings:  RELIABILITY and STATISTICS in TRANSPORTATION and COMMUNICATION 

Conference:  Reliability and Statistics in Transportation and Communication 

Series:  Lecture Notes in Networks and Systems 

Place:  Riga, LV 

Year:  2017 

ISBN:  9789984818863 

Journal:  Lecture Notes in Networks and Systems, No. 36, Cham, CH 

ISSN:  23673370 

DOI:  10.1007/9783319744544_12 

Publisher:  Springer International Publishing 

Keywords 

time series, statistical models, ARIMA, time prediction, ship handling, oil terminal 
Annotation 

This work relates to the whole series of papers aimed at creating a marine transport and logistics process map. This map is a reflection of a real process model (descriptive model) with the possibility of extension (scaling process), determination bottlenecks (traffic jam), detecting of deviations for operational response, representation of different perspectives (controlflow, resources, performance). Also, the map can be used as a basis for prediction and decision making systems. As the object of the study, the port module was chosen, namely its component part  the oil terminal. The analysed process includes the whole ship handling from the moment of its arrival to the port (activity Notice received) till the departure (operation Pilotage). Today there are a huge number of ways to model the processes and the main aim is searching of optimal and effective methods of modern intelligent analysis (from the field of Machine Learning, Data Mining, statistics, Process Mining) for building a process map. The main point of this paper is to conduct research of time series and, then, to build statistical prediction model based on obtained characteristics. At the beginning of the article, the analysed time series is presented, which shows the distribution of the ship handling duration for the last 3 years. The main components of the time series, an explanation of their values and their effect on the prediction model are given below. In this article, the famous statistical model auto regression integrated moving average (ARIMA) was chosen for the prediction. The paper presents the results of its application to the port data, the advantages and disadvantages are indicated. 
BibTeX: 

@INPROCEEDINGS{
author = {Julia Rudnitckaia and Tom{\'{a}}{\v{s}}
Hru{\v{s}}ka},
title = {Time Series Analysis and Prediction Statistical
Models for the Duration of the Ship Handling at an
Oil Terminal},
pages = {127136},
booktitle = {RELIABILITY and STATISTICS in TRANSPORTATION and
COMMUNICATION},
series = {Lecture Notes in Networks and Systems},
journal = {Lecture Notes in Networks and Systems},
number = 36,
year = 2017,
location = {Riga, LV},
publisher = {Springer International Publishing},
ISBN = {9789984818863},
ISSN = {23673370},
doi = {10.1007/9783319744544_12},
language = {english},
url = {http://www.fit.vutbr.cz/research/view_pub.php?id=11538}
} 
