Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/224718 
Erscheinungsjahr: 
2020
Quellenangabe: 
[Title:] Proceedings of the ENTRENOVA - ENTerprise REsearch InNOVAtion Conference, Virtual Conference, 10-12 September 2020 [Publisher:] IRENET - Society for Advancing Innovation and Research in Economy [Place:] Zagreb [Year:] 2020 [Pages:] 522-530
Verlag: 
IRENET - Society for Advancing Innovation and Research in Economy, Zagreb
Zusammenfassung: 
Intelligent Systems, thanks to their effectiveness and robustness, find many applications in various industries. One of such applications is optimization of distribution network of small-scale LNG market, which was highly dynamic throughout last years. LNG (Liquified Natural Gas) is a fuel produced from natural gas, but its volume is approx. 600 times smaller than in the gas (natural) state, which makes it more economically effective to transport and store. Distribution network consists of several pickup points (varying in LNG specification) and a number of destination points (varying in tanks capacities). From economic point of view, optimization of LNG truck tanks paths is an important factor in whole market development. The optimization process involves selecting a pickup point and a sequence of destination points with amount of LNG unloaded in each of them. Solution proposed in this paper is based on graph theory and advanced machine learning methods, such as reinforcement learning, recurrent neural networks and online learning. Optimization of distribution network translates directly into a number of economic benefits: reduction of LNG transport cost, shortening the delivery time, reduction of distribution costs and increase in the effectiveness of tank truck usage.
Schlagwörter: 
Liquified Natural Gas
distribution network
artificial intelligence
reinforcement learning
economic optimization
JEL: 
O31
Creative-Commons-Lizenz: 
cc-by-nc Logo
Dokumentart: 
Conference Paper

Datei(en):
Datei
Größe





Publikationen in EconStor sind urheberrechtlich geschützt.