Abstract
We implement machine learning techniques to predict the destination for Latin American crude oil exports. Utilizing a unique dataset of micro-level crude oil shipment data, derived from the Automatic Identification System (AIS) for ship tracking, we investigate the micro- and macro-level determinants of the destination choice. We use decision trees, Random Forests, and boosted trees techniques in training a model to predict the export destinations which can help to identify seller/buyer groups with similar oil trade requirements. The results show that while macro data, such as regional oil price differences and crack spreads, impacts the crude oil flow, micro-level information about the oil shipment are key attributes in the destination prediction. Our research has practical implications, particularly with regards to the prediction of oil transportation demand, spatial price arbitrage, and short-term forecasting of regional crack spread.
-
Author Keywords
- big data,
- crude oil,
- machine learning
Introduction
Latin America as a region may not always catch the world’s attention on the global energy scene, yet some of the countries in the region possess some of the largest oil and gas reserves in the world. Venezuela and Mexico have been large oil producers for decades, and oil discoveries ten years ago in the Tupi fields off the coast of Brazil has made the country one of the major oil and gas producers in the world. According to the World Factbook [1], seven countries in Latin America are net oil and gas exporters as of 2017, with a total exporting volume of 8.3 million barrels per day (mbbl/day). Though the picture of the region’s economic outlook is mixed – recessions in Brazil and Argentina are said to come to an end; while the situation in Venezuela continues to be disturbing – the region’s dependence on oil will continue to be an essential feature of its integration into the world economy.
The eventual destination of oil exports is the result of a complex and dynamic system including, for instance, trade agreements (long-term bilateral agreement and short-term commercial contracts), political relationships (sanctions or restrictions), supply and demand, and price fluctuations. Discrete choice models, see, for instance, Malchow and Kanafani [2], Rich et al. [3], Steven and Corsi [4], and Piendl et al. [5], provide a theoretical foundation for this research in terms of the choices are statistically related to some attributes. However, we opt for utilizing machine learning techniques that have the advantage of dealing with high dimensionality, mixture date type, and nonstandard data structure [6]. As there has been no academic research on the topic, the contribution of this paper is to provide an in-depth investigation of the attributes that determine the destination of seaborne oil exports using machine learning algorithms. Based on actual micro-level crude oil shipment data for the period January 2013 through mid-March 2016, we investigate how the destinations are determined based on cargo information (such as seller’s identity, loading port, and cargo grade, etc.) and economic data (e.g. oil prices and crack spread). We train the machine learning algorithm based on historical data and test the out-of-sample prediction performance of the model. Our results are potentially important as a building block in commercial applications that deal with oil and freight market analysis.
Conclusion
This paper is the first academic research to apply cutting-edge machine learning models in predicting destinations of seaborne oil trade. We base the training of the models on a rich micro-level dataset of shipments with detailed information on crude quality, oil buyer and seller identity, cargo size, and other attributes. Our application to Latin American crude oil exports at the country level results in test accuracies ranging between 70 and 90% – a strong performance. Predicting oil export destinations allows for better forecasting of regional and local market balance, improved knowledge of inventory levels, and monitoring of the supply chain.
We acknowledge that the relative rigidity of global crude oil trade, with a predominance of long-term offtake agreements and national oil buyers and refinery operations should increase predictability relative to other applications of choice models in transportation. However, our work still points to an important application of micro-level data and machine learning models in an effort to improve the oil and tanker freight market analysis. Further research in this area should look at improving prediction accuracy by diving further into the details, such as vessel type, vessel ownership, and trade characteristics.
Acknowledgment
This research is partially funded by the Norwegian Research Council, under the project “Smart Digital Contracts and Commercial Management” (project number: 280684).
About KSRA
The Kavian Scientific Research Association (KSRA) is a non-profit research organization to provide research / educational services in December 2013. The members of the community had formed a virtual group on the Viber social network. The core of the Kavian Scientific Association was formed with these members as founders. These individuals, led by Professor Siavosh Kaviani, decided to launch a scientific / research association with an emphasis on education.
KSRA research association, as a non-profit research firm, is committed to providing research services in the field of knowledge. The main beneficiaries of this association are public or private knowledge-based companies, students, researchers, researchers, professors, universities, and industrial and semi-industrial centers around the world.
Our main services Based on Education for all Spectrum people in the world. We want to make an integration between researches and educations. We believe education is the main right of Human beings. So our services should be concentrated on inclusive education.
The KSRA team partners with local under-served communities around the world to improve the access to and quality of knowledge based on education, amplify and augment learning programs where they exist, and create new opportunities for e-learning where traditional education systems are lacking or non-existent.
FULL Paper PDF file:
Latin American Oil Export Destination Choice: A Machine Learning Approachjia2019
Bibliography
author
Year
2020
Title
Latin American Oil Export Destination Choice: A Machine Learning Approach
Publish in
2019 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM), Macao, Macao, 2019, pp. 345-348
Doi
10.1109/IEEM44572.2019.8978548.
PDF reference and original file: Click here
Somayeh Nosrati was born in 1982 in Tehran. She holds a Master's degree in artificial intelligence from Khatam University of Tehran.
Professor Siavosh Kaviani was born in 1961 in Tehran. He had a professorship. He holds a Ph.D. in Software Engineering from the QL University of Software Development Methodology and an honorary Ph.D. from the University of Chelsea.
Nasim Gazerani was born in 1983 in Arak. She holds a Master's degree in Software Engineering from UM University of Malaysia.
