Optimization of Vehicle Sharing Systems : Unconstraining, System Analysis & Dynamic Pricing

Hardt, Cornelius

User: Guest

Cornelius Hardt

If you experience problems opening the document, please try this link.

Original title:: Optimization of Vehicle Sharing Systems
Original subtitle:: Unconstraining, System Analysis & Dynamic Pricing
Translated title:: Optimierung von Vehicle-Sharing-Systemen
Translated subtitle:: Unconstraining, Systemanalyse, Dynamisches Pricing
Author:: Hardt, Cornelius
Year:: 2023
Document type:: Dissertation
Faculty/School:: TUM School of Engineering and Design
Advisor:: Bogenberger, Klaus (Prof. Dr.)
Referee:: Bogenberger, Klaus (Prof. Dr.); Pickl, Stefan (Prof. Dr.); Minner, Stefan (Prof. Dr.)
Language:: en
Subject group:: BAU Bauingenieurwesen, Vermessungswesen
Keywords:: vehicle sharing system, free floating; car sharing; pricing; dynamic; unconstraining; clustering; Analysis; MDP; ADP, MPC
Translated keywords:: vehicle sharing systeme, free-floating, carsharing, pricing, dynamisch; unconstraining; clustering; Analyse; MDP; ADP, MPC
TUM classification:: BAU 850; BAU 875
Abstract:: The thesis at hand explores the steps necessary to optimize a Vehicle Sharing System in free-floating mode. A novel unconstraining algorithm is presented to estimate real demand from rental data. A clustering is presented that reduces dimensionality in system analysis. Finally, a dynamic pricing is introduced, utilizing a Markov- Decision-Process Modelling and a full dynamic pricing policy. With the approach presented, increases of up to 10% in profit can be generated.
Translated abstract:: Die vorliegende Arbeit untersucht die Schritte zur Optimierung eines Free-floating Sharing-Systems. Ein neuer Unconstraining-Algorithmus wird präsentiert der die tatsächliche Nachfrage aus Mietdaten abschätzen, sowie ein Clustering, welches die Dimensionen in der Systemanalyse reduziert. Schlussendlich wird ein Dynamisches Pricing vorgestellt basierend auf einem Markov-Decision-Prozess sowie einer Pricing Policy. Der vorgestellte Ansatz ermöglicht Gewinnsteigerungen um bis zu 10%.
WWW:: https://mediatum.ub.tum.de/?id=1685694
Date of submission:: 27.09.2022
Oral examination:: 15.05.2023
File size:: 193480120 bytes
Pages:: 252
Urn (citeable URL):: https://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:91-diss-20230515-1685694-1-0
Last change:: 14.06.2023
BibTeX