TY - GEN A1 - Richly, Keven A1 - Brauer, Janos A1 - Schlosser, Rainer T1 - Predicting location probabilities of drivers to improved dispatch decisions of transportation network companies based on trajectory data T2 - Postprints der Universität Potsdam : Reihe der Digital Engineering Fakultät N2 - The demand for peer-to-peer ridesharing services increased over the last years rapidly. To cost-efficiently dispatch orders and communicate accurate pick-up times is challenging as the current location of each available driver is not exactly known since observed locations can be outdated for several seconds. The developed trajectory visualization tool enables transportation network companies to analyze dispatch processes and determine the causes of unexpected delays. As dispatching algorithms are based on the accuracy of arrival time predictions, we account for factors like noise, sample rate, technical and economic limitations as well as the duration of the entire process as they have an impact on the accuracy of spatio-temporal data. To improve dispatching strategies, we propose a prediction approach that provides a probability distribution for a driver’s future locations based on patterns observed in past trajectories. We demonstrate the capabilities of our prediction results to ( i) avoid critical delays, (ii) to estimate waiting times with higher confidence, and (iii) to enable risk considerations in dispatching strategies. T3 - Zweitveröffentlichungen der Universität Potsdam : Reihe der Digital Engineering Fakultät - 9 KW - trajectory data KW - location prediction algorithm KW - Peer-to-Peer ridesharing KW - transport network companies KW - risk-aware dispatching Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-524040 IS - 9 ER - TY - JOUR A1 - Richly, Keven A1 - Brauer, Janos A1 - Schlosser, Rainer T1 - Predicting location probabilities of drivers to improved dispatch decisions of transportation network companies based on trajectory data JF - Proceedings of the 9th International Conference on Operations Research and Enterprise Systems - ICORES N2 - The demand for peer-to-peer ridesharing services increased over the last years rapidly. To cost-efficiently dispatch orders and communicate accurate pick-up times is challenging as the current location of each available driver is not exactly known since observed locations can be outdated for several seconds. The developed trajectory visualization tool enables transportation network companies to analyze dispatch processes and determine the causes of unexpected delays. As dispatching algorithms are based on the accuracy of arrival time predictions, we account for factors like noise, sample rate, technical and economic limitations as well as the duration of the entire process as they have an impact on the accuracy of spatio-temporal data. To improve dispatching strategies, we propose a prediction approach that provides a probability distribution for a driver’s future locations based on patterns observed in past trajectories. We demonstrate the capabilities of our prediction results to ( i) avoid critical delays, (ii) to estimate waiting times with higher confidence, and (iii) to enable risk considerations in dispatching strategies. KW - trajectory data KW - location prediction algorithm KW - Peer-to-Peer ridesharing KW - transport network companies KW - risk-aware dispatching Y1 - 2020 PB - Springer CY - Berlin ER - TY - THES A1 - Richly, Keven T1 - Memory-efficient data management for spatio-temporal applications BT - workload-driven fine-grained configuration optimization for storing spatio-temporal data in columnar In-memory databases N2 - The wide distribution of location-acquisition technologies means that large volumes of spatio-temporal data are continuously being accumulated. Positioning systems such as GPS enable the tracking of various moving objects' trajectories, which are usually represented by a chronologically ordered sequence of observed locations. The analysis of movement patterns based on detailed positional information creates opportunities for applications that can improve business decisions and processes in a broad spectrum of industries (e.g., transportation, traffic control, or medicine). Due to the large data volumes generated in these applications, the cost-efficient storage of spatio-temporal data is desirable, especially when in-memory database systems are used to achieve interactive performance requirements. To efficiently utilize the available DRAM capacities, modern database systems support various tuning possibilities to reduce the memory footprint (e.g., data compression) or increase performance (e.g., additional indexes structures). By considering horizontal data partitioning, we can independently apply different tuning options on a fine-grained level. However, the selection of cost and performance-balancing configurations is challenging, due to the vast number of possible setups consisting of mutually dependent individual decisions. In this thesis, we introduce multiple approaches to improve spatio-temporal data management by automatically optimizing diverse tuning options for the application-specific access patterns and data characteristics. Our contributions are as follows: (1) We introduce a novel approach to determine fine-grained table configurations for spatio-temporal workloads. Our linear programming (LP) approach jointly optimizes the (i) data compression, (ii) ordering, (iii) indexing, and (iv) tiering. We propose different models which address cost dependencies at different levels of accuracy to compute optimized tuning configurations for a given workload, memory budgets, and data characteristics. To yield maintainable and robust configurations, we further extend our LP-based approach to incorporate reconfiguration costs as well as optimizations for multiple potential workload scenarios. (2) To optimize the storage layout of timestamps in columnar databases, we present a heuristic approach for the workload-driven combined selection of a data layout and compression scheme. By considering attribute decomposition strategies, we are able to apply application-specific optimizations that reduce the memory footprint and improve performance. (3) We introduce an approach that leverages past trajectory data to improve the dispatch processes of transportation network companies. Based on location probabilities, we developed risk-averse dispatch strategies that reduce critical delays. (4) Finally, we used the use case of a transportation network company to evaluate our database optimizations on a real-world dataset. We demonstrate that workload-driven fine-grained optimizations allow us to reduce the memory footprint (up to 71% by equal performance) or increase the performance (up to 90% by equal memory size) compared to established rule-based heuristics. Individually, our contributions provide novel approaches to the current challenges in spatio-temporal data mining and database research. Combining them allows in-memory databases to store and process spatio-temporal data more cost-efficiently. N2 - Durch die starke Verbreitung von Systemen zur Positionsbestimmung werden fortlaufend große Mengen an Bewegungsdaten mit einem räumlichen und zeitlichen Bezug gesammelt. Ortungssysteme wie GPS ermöglichen, die Bewegungen verschiedener Objekte (z. B. Personen oder Fahrzeuge) nachzuverfolgen. Diese werden in der Regel durch eine chronologisch geordnete Abfolge beobachteter Aufenthaltsorte repräsentiert. Die Analyse von Bewegungsmustern auf der Grundlage detaillierter Positionsinformationen schafft in unterschiedlichsten Branchen (z. B. Transportwesen, Verkehrssteuerung oder Medizin) die Möglichkeit Geschäftsentscheidungen und -prozesse zu verbessern. Aufgrund der großen Datenmengen, die bei diesen Anwendungen auftreten, stellt die kosteneffiziente Speicherung von Bewegungsdaten eine Herausforderung dar. Dies ist insbesondere der Fall, wenn Hauptspeicherdatenbanken zur Speicherung eingesetzt werden, um die Anforderungen bezüglich interaktiver Antwortzeiten zu erfüllen. Um die verfügbaren Speicherkapazitäten effizient zu nutzen, unterstützen moderne Datenbanksysteme verschiedene Optimierungsmöglichkeiten, um den Speicherbedarf zu reduzieren (z. B. durch Datenkomprimierung) oder die Performance zu erhöhen (z. B. durch Indexstrukturen). Dabei ermöglicht eine horizontale Partitionierung der Daten, dass unabhängig voneinander verschiedene Optimierungen feingranular auf einzelnen Bereichen der Daten angewendet werden können. Die Auswahl von Konfigurationen, die sowohl die Kosten als auch Leistungsanforderungen berücksichtigen, ist jedoch aufgrund der großen Anzahl möglicher Kombinationen -- die aus voneinander abhängigen Einzelentscheidungen bestehen -- komplex. In dieser Dissertation präsentieren wir mehrere Ansätze zur Verbesserung der Datenverwaltung, indem wir die Auswahl verschiedener Datenbankoptimierungen automatisch für die anwendungsspezifischen Zugriffsmuster und Dateneigenschaften anpassen. Diesbezüglich leistet die vorliegende Dissertation die folgenden Beiträge: (1) Wir stellen einen neuen Ansatz vor, um feingranulare Tabellenkonfigurationen für räumlich-zeitliche Workloads zu bestimmen. In diesem Zusammenhang optimiert unser Linear Programming (LP) Ansatz gemeinsam (i) die Datenkompression, (ii) die Sortierung, (iii) die Indizierung und (iv) die Datenplatzierung. Hierzu schlagen wir verschiedene Modelle mit unterschiedlichen Kostenabhängigkeiten vor, um optimierte Konfigurationen für einen gegebenen Workload, ein Speicherbudget und die vorliegenden Dateneigenschaften zu berechnen. Durch die Erweiterung des LP-basierten Ansatzes zur Berücksichtigung von Modifikationskosten und verschiedener potentieller Workloads ist es möglich, die Wartbarkeit und Robustheit der bestimmten Tabellenkonfiguration zu erhöhen. (2) Um die Speicherung von Timestamps in spalten-orientierten Datenbanken zu optimieren, stellen wir einen heuristischen Ansatz für die kombinierte Auswahl eines Speicherlayouts und eines Kompressionsschemas vor. Zudem sind wir durch die Berücksichtigung von Strategien zur Aufteilung von Attributen in der Lage, anwendungsspezifische Optimierungen anzuwenden, die den Speicherbedarf reduzieren und die Performance verbessern. (3) Wir stellen einen Ansatz vor, der in der Vergangenheit beobachtete Bewegungsmuster nutzt, um die Zuweisungsprozesse von Vermittlungsdiensten zur Personenbeförderung zu verbessern. Auf der Grundlage von Standortwahrscheinlichkeiten haben wir verschiedene Strategien für die Vergabe von Fahraufträgen an Fahrer entwickelt, die kritische Verspätungen reduzieren. (4) Abschließend haben wir unsere Datenbankoptimierungen anhand eines realen Datensatzes eines Transportdienstleisters evaluiert. In diesem Zusammenhang zeigen wir, dass wir durch feingranulare workload-basierte Optimierungen den Speicherbedarf (um bis zu 71% bei vergleichbarer Performance) reduzieren oder die Performance (um bis zu 90% bei gleichem Speicherverbrauch) im Vergleich zu regelbasierten Heuristiken verbessern können. Die einzelnen Beiträge stellen neuartige Ansätze für aktuelle Herausforderungen im Bereich des Data Mining und der Datenbankforschung dar. In Kombination ermöglichen sie eine kosteneffizientere Speicherung und Verarbeitung von Bewegungsdaten in Hauptspeicherdatenbanken. KW - spatio-temporal data management KW - trajectory data KW - columnar databases KW - in-memory data management KW - database tuning KW - spaltenorientierte Datenbanken KW - Datenbankoptimierung KW - Hauptspeicher Datenmanagement KW - Datenverwaltung für Daten mit räumlich-zeitlichem Bezug KW - Trajektoriendaten Y1 - 2024 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-635473 ER - TY - GEN A1 - Richly, Keven T1 - Leveraging spatio-temporal soccer data to define a graphical query language for game recordings T2 - IEEE International Conference on Big Data (Big Data) N2 - For professional soccer clubs, performance and video analysis are an integral part of the preparation and post-processing of games. Coaches, scouts, and video analysts extract information about strengths and weaknesses of their team as well as opponents by manually analyzing video recordings of past games. Since video recordings are an unstructured data source, it is a complex and time-intensive task to find specific game situations and identify similar patterns. In this paper, we present a novel approach to detect patterns and situations (e.g., playmaking and ball passing of midfielders) based on trajectory data. The application uses the metaphor of a tactic board to offer a graphical query language. With this interactive tactic board, the user can model a game situation or mark a specific situation in the video recording for which all matching occurrences in various games are immediately displayed, and the user can directly jump to the corresponding game scene. Through the additional visualization of key performance indicators (e.g.,the physical load of the players), the user can get a better overall assessment of situations. With the capabilities to find specific game situations and complex patterns in video recordings, the interactive tactic board serves as a useful tool to improve the video analysis process of professional sports teams. KW - Spatio-temporal data analysis KW - soccer analytics KW - graphical query language Y1 - 2019 SN - 978-1-5386-5035-6 U6 - https://doi.org/10.1109/BigData.2018.8622159 SN - 2639-1589 SP - 3456 EP - 3463 PB - IEEE CY - New York ER - TY - JOUR A1 - Schlosser, Rainer A1 - Richly, Keven T1 - Dynamic pricing under competition with data-driven price anticipations and endogenous reference price effects JF - Journal of revenue and pricing management N2 - Online markets have become highly dynamic and competitive. Many sellers use automated data-driven strategies to estimate demand and to update prices frequently. Further, notification services offered by marketplaces allow to continuously track markets and to react to competitors’ price adjustments instantaneously. To derive successful automated repricing strategies is challenging as competitors’ strategies are typically not known. In this paper, we analyze automated repricing strategies with data-driven price anticipations under duopoly competition. In addition, we account for reference price effects in demand, which are affected by the price adjustments of both competitors. We show how to derive optimized self-adaptive pricing strategies that anticipate price reactions of the competitor and take the evolution of the reference price into account. We verify that the results of our adaptive learning strategy tend to optimal solutions, which can be derived for scenarios with full information. Finally, we analyze the case in which our learning strategy is played against itself. We find that our self-adaptive strategies can be used to approximate equilibria in mixed strategies. KW - Dynamic pricing competition KW - Data-driven price anticipation KW - e-Commerce KW - Dynamic programming KW - Response strategies Y1 - 2019 U6 - https://doi.org/10.1057/s41272-019-00206-5 SN - 1476-6930 SN - 1477-657X VL - 18 IS - 6 SP - 451 EP - 464 PB - Palgrave Macmillan CY - Basingstoke ER - TY - JOUR A1 - Richly, Keven A1 - Schlosser, Rainer A1 - Boissier, Martin T1 - Budget-conscious fine-grained configuration optimization for spatio-temporal applications JF - Proceedings of the VLDB Endowment N2 - Based on the performance requirements of modern spatio-temporal data mining applications, in-memory database systems are often used to store and process the data. To efficiently utilize the scarce DRAM capacities, modern database systems support various tuning possibilities to reduce the memory footprint (e.g., data compression) or increase performance (e.g., additional indexes). However, the selection of cost and performance balancing configurations is challenging due to the vast number of possible setups consisting of mutually dependent individual decisions. In this paper, we introduce a novel approach to jointly optimize the compression, sorting, indexing, and tiering configuration for spatio-temporal workloads. Further, we consider horizontal data partitioning, which enables the independent application of different tuning options on a fine-grained level. We propose different linear programming (LP) models addressing cost dependencies at different levels of accuracy to compute optimized tuning configurations for a given workload and memory budgets. To yield maintainable and robust configurations, we extend our LP-based approach to incorporate reconfiguration costs as well as a worst-case optimization for potential workload scenarios. Further, we demonstrate on a real-world dataset that our models allow to significantly reduce the memory footprint with equal performance or increase the performance with equal memory size compared to existing tuning heuristics. KW - General Earth and Planetary Sciences KW - Water Science and Technology KW - Geography, Planning and Development Y1 - 2022 U6 - https://doi.org/10.14778/3565838.3565858 SN - 2150-8097 VL - 15 IS - 13 SP - 4079 EP - 4092 PB - Association for Computing Machinery (ACM) CY - [New York] ER - TY - GEN A1 - Richly, Keven T1 - A survey on trajectory data management for hybrid transactional and analytical workloads T2 - IEEE International Conference on Big Data (Big Data) N2 - Rapid advances in location-acquisition technologies have led to large amounts of trajectory data. This data is the foundation for a broad spectrum of services driven and improved by trajectory data mining. However, for hybrid transactional and analytical workloads, the storing and processing of rapidly accumulated trajectory data is a non-trivial task. In this paper, we present a detailed survey about state-of-the-art trajectory data management systems. To determine the relevant aspects and requirements for such systems, we developed a trajectory data mining framework, which summarizes the different steps in the trajectory data mining process. Based on the derived requirements, we analyze different concepts to store, compress, index, and process spatio-temporal data. There are various trajectory management systems, which are optimized for scalability, data footprint reduction, elasticity, or query performance. To get a comprehensive overview, we describe and compare different exciting systems. Additionally, the observed similarities in the general structure of different systems are consolidated in a general blueprint of trajectory management systems. KW - Trajectory Data Management KW - Spatio-Temporal Data KW - Survey Y1 - 2019 SN - 978-1-5386-5035-6 U6 - https://doi.org/10.1109/BigData.2018.8622394 SN - 2639-1589 SP - 562 EP - 569 PB - IEEE CY - New York ER -