@article{KastiusSchlosser2022, author = {Kastius, Alexander and Schlosser, Rainer}, title = {Dynamic pricing under competition using reinforcement learning}, series = {Journal of revenue and pricing management}, volume = {21}, journal = {Journal of revenue and pricing management}, number = {1}, publisher = {Springer Nature Switzerland AG}, address = {Cham}, issn = {1476-6930}, doi = {10.1057/s41272-021-00285-3}, pages = {50 -- 63}, year = {2022}, abstract = {Dynamic pricing is considered a possibility to gain an advantage over competitors in modern online markets. The past advancements in Reinforcement Learning (RL) provided more capable algorithms that can be used to solve pricing problems. In this paper, we study the performance of Deep Q-Networks (DQN) and Soft Actor Critic (SAC) in different market models. We consider tractable duopoly settings, where optimal solutions derived by dynamic programming techniques can be used for verification, as well as oligopoly settings, which are usually intractable due to the curse of dimensionality. We find that both algorithms provide reasonable results, while SAC performs better than DQN. Moreover, we show that under certain conditions, RL algorithms can be forced into collusion by their competitors without direct communication.}, language = {en} } @article{Schlosser2022, author = {Schlosser, Rainer}, title = {Heuristic mean-variance optimization in Markov decision processes using state-dependent risk aversion}, series = {IMA journal of management mathematics / Institute of Mathematics and Its Applications}, volume = {33}, journal = {IMA journal of management mathematics / Institute of Mathematics and Its Applications}, number = {2}, publisher = {Oxford Univ. Press}, address = {Oxford}, issn = {1471-678X}, doi = {10.1093/imaman/dpab009}, pages = {181 -- 199}, year = {2022}, abstract = {In dynamic decision problems, it is challenging to find the right balance between maximizing expected rewards and minimizing risks. In this paper, we consider NP-hard mean-variance (MV) optimization problems in Markov decision processes with a finite time horizon. We present a heuristic approach to solve MV problems, which is based on state-dependent risk aversion and efficient dynamic programming techniques. Our approach can also be applied to mean-semivariance (MSV) problems, which particularly focus on the downside risk. We demonstrate the applicability and the effectiveness of our heuristic for dynamic pricing applications. Using reproducible examples, we show that our approach outperforms existing state-of-the-art benchmark models for MV and MSV problems while also providing competitive runtimes. Further, compared to models based on constant risk levels, we find that state-dependent risk aversion allows to more effectively intervene in case sales processes deviate from their planned paths. Our concepts are domain independent, easy to implement and of low computational complexity.}, language = {en} } @article{HagedornHuegleSchlosser2022, author = {Hagedorn, Christopher and Huegle, Johannes and Schlosser, Rainer}, title = {Understanding unforeseen production downtimes in manufacturing processes using log data-driven causal reasoning}, series = {Journal of intelligent manufacturing}, volume = {33}, journal = {Journal of intelligent manufacturing}, number = {7}, publisher = {Springer}, address = {Dordrecht}, issn = {0956-5515}, doi = {10.1007/s10845-022-01952-x}, pages = {2027 -- 2043}, year = {2022}, abstract = {In discrete manufacturing, the knowledge about causal relationships makes it possible to avoid unforeseen production downtimes by identifying their root causes. Learning causal structures from real-world settings remains challenging due to high-dimensional data, a mix of discrete and continuous variables, and requirements for preprocessing log data under the causal perspective. In our work, we address these challenges proposing a process for causal reasoning based on raw machine log data from production monitoring. Within this process, we define a set of transformation rules to extract independent and identically distributed observations. Further, we incorporate a variable selection step to handle high-dimensionality and a discretization step to include continuous variables. We enrich a commonly used causal structure learning algorithm with domain-related orientation rules, which provides a basis for causal reasoning. We demonstrate the process on a real-world dataset from a globally operating precision mechanical engineering company. The dataset contains over 40 million log data entries from production monitoring of a single machine. In this context, we determine the causal structures embedded in operational processes. Further, we examine causal effects to support machine operators in avoiding unforeseen production stops, i.e., by detaining machine operators from drawing false conclusions on impacting factors of unforeseen production stops based on experience.}, language = {en} } @article{RichlySchlosserBoissier2022, author = {Richly, Keven and Schlosser, Rainer and Boissier, Martin}, title = {Budget-conscious fine-grained configuration optimization for spatio-temporal applications}, series = {Proceedings of the VLDB Endowment}, volume = {15}, journal = {Proceedings of the VLDB Endowment}, number = {13}, publisher = {Association for Computing Machinery (ACM)}, address = {[New York]}, issn = {2150-8097}, doi = {10.14778/3565838.3565858}, pages = {4079 -- 4092}, year = {2022}, abstract = {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.}, language = {en} } @article{SchlosserChenavazDimitrov2021, author = {Schlosser, Rainer and Chenavaz, R{\´e}gis Y. and Dimitrov, Stanko}, title = {Circular economy}, series = {International journal of production economics}, volume = {236}, journal = {International journal of production economics}, publisher = {Elsevier}, address = {Amsterdam}, issn = {0925-5273}, doi = {10.1016/j.ijpe.2021.108117}, pages = {13}, year = {2021}, abstract = {In a circular economy, the use of recycled resources in production is a key performance indicator for management. Yet, academic studies are still unable to inform managers on appropriate recycling and pricing policies. We develop an optimal control model integrating a firm's recycling rate, which can use both virgin and recycled resources in the production process. Our model accounts for recycling influence both at the supply- and demandsides. The positive effect of a firm's use of recycled resources diminishes over time but may increase through investments. Using general formulations for demand and cost, we analytically examine joint dynamic pricing and recycling investment policies in order to determine their optimal interplay over time. We provide numerical experiments to assess the existence of a steady-state and to calculate sensitivity analyses with respect to various model parameters. The analysis shows how to dynamically adapt jointly optimized controls to reach sustainability in the production process. Our results pave the way to sounder sustainable practices for firms operating within a circular economy.}, language = {en} } @article{RichlyBrauerSchlosser2020, author = {Richly, Keven and Brauer, Janos and Schlosser, Rainer}, title = {Predicting location probabilities of drivers to improved dispatch decisions of transportation network companies based on trajectory data}, series = {Proceedings of the 9th International Conference on Operations Research and Enterprise Systems - ICORES}, journal = {Proceedings of the 9th International Conference on Operations Research and Enterprise Systems - ICORES}, publisher = {Springer}, address = {Berlin}, pages = {12}, year = {2020}, abstract = {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.}, language = {en} } @article{KossmannHalfpapJankriftetal.2020, author = {Kossmann, Jan and Halfpap, Stefan and Jankrift, Marcel and Schlosser, Rainer}, title = {Magic mirror in my hand, which is the best in the land?}, series = {Proceedings of the VLDB Endowment}, volume = {13}, journal = {Proceedings of the VLDB Endowment}, number = {11}, publisher = {Association for Computing Machinery}, address = {New York}, issn = {2150-8097}, doi = {10.14778/3407790.3407832}, pages = {2382 -- 2395}, year = {2020}, abstract = {Indexes are essential for the efficient processing of database workloads. Proposed solutions for the relevant and challenging index selection problem range from metadata-based simple heuristics, over sophisticated multi-step algorithms, to approaches that yield optimal results. The main challenges are (i) to accurately determine the effect of an index on the workload cost while considering the interaction of indexes and (ii) a large number of possible combinations resulting from workloads containing many queries and massive schemata with possibly thousands of attributes.
In this work, we describe and analyze eight index selection algorithms that are based on different concepts and compare them along different dimensions, such as solution quality, runtime, multi-column support, solution granularity, and complexity. In particular, we analyze the solutions of the algorithms for the challenging analytical Join Order, TPC-H, and TPC-DS benchmarks. Afterward, we assess strengths and weaknesses, infer insights for index selection in general and each approach individually, before we give recommendations on when to use which approach.}, language = {en} } @article{Schlosser2020, author = {Schlosser, Rainer}, title = {Risk-sensitive control of Markov decision processes}, series = {Computers \& operations research : and their applications to problems of world concern}, volume = {123}, journal = {Computers \& operations research : and their applications to problems of world concern}, publisher = {Elsevier}, address = {Oxford}, issn = {0305-0548}, doi = {10.1016/j.cor.2020.104997}, pages = {14}, year = {2020}, abstract = {In many revenue management applications risk-averse decision-making is crucial. In dynamic settings, however, it is challenging to find the right balance between maximizing expected rewards and minimizing various kinds of risk. In existing approaches utility functions, chance constraints, or (conditional) value at risk considerations are used to influence the distribution of rewards in a preferred way. Nevertheless, common techniques are not flexible enough and typically numerically complex. In our model, we exploit the fact that a distribution is characterized by its mean and higher moments. We present a multi-valued dynamic programming heuristic to compute risk-sensitive feedback policies that are able to directly control the moments of future rewards. Our approach is based on recursive formulations of higher moments and does not require an extension of the state space. Finally, we propose a self-tuning algorithm, which allows to identify feedback policies that approximate predetermined (risk-sensitive) target distributions. We illustrate the effectiveness and the flexibility of our approach for different dynamic pricing scenarios. (C) 2020 Elsevier Ltd. All rights reserved.}, language = {en} } @article{KossmannSchlosser2020, author = {Kossmann, Jan and Schlosser, Rainer}, title = {Self-driving database systems}, series = {Distributed and parallel databases}, volume = {38}, journal = {Distributed and parallel databases}, number = {4}, publisher = {Springer}, address = {Dordrecht}, issn = {0926-8782}, doi = {10.1007/s10619-020-07288-w}, pages = {795 -- 817}, year = {2020}, abstract = {Challenges for self-driving database systems, which tune their physical design and configuration autonomously, are manifold: Such systems have to anticipate future workloads, find robust configurations efficiently, and incorporate knowledge gained by previous actions into later decisions. We present a component-based framework for self-driving database systems that enables database integration and development of self-managing functionality with low overhead by relying on separation of concerns. By keeping the components of the framework reusable and exchangeable, experiments are simplified, which promotes further research in that area. Moreover, to optimize multiple mutually dependent features, e.g., index selection and compression configurations, we propose a linear programming (LP) based algorithm to derive an efficient tuning order automatically. Afterwards, we demonstrate the applicability and scalability of our approach with reproducible examples.}, language = {en} } @article{Schlosser2020, author = {Schlosser, Rainer}, title = {Scalable relaxation techniques to solve stochastic dynamic multi-product pricing problems with substitution effects}, series = {Journal of revenue and pricing management}, volume = {20}, journal = {Journal of revenue and pricing management}, number = {1}, publisher = {Palgrave Macmillan}, address = {Basingstoke}, issn = {1476-6930}, doi = {10.1057/s41272-020-00249-z}, pages = {54 -- 65}, year = {2020}, abstract = {In many businesses, firms are selling different types of products, which share mutual substitution effects in demand. To compute effective pricing strategies is challenging as the sales probabilities of each of a firm's products can also be affected by the prices of potential substitutes. In this paper, we analyze stochastic dynamic multi-product pricing models for the sale of perishable goods. To circumvent the limitations of time-consuming optimal solutions for highly complex models, we propose different relaxation techniques, which allow to reduce the size of critical model components, such as the state space, the action space, or the set of potential sales events. Our heuristics are able to decrease the size of those components by forming corresponding clusters and using subsets of representative elements. Using numerical examples, we verify that our heuristics make it possible to dramatically reduce the computation time while still obtaining close-to-optimal expected profits. Further, we show that our heuristics are (i) flexible, (ii) scalable, and (iii) can be arbitrarily combined in a mutually supportive way.}, language = {en} }