Intelligent Decision Support for Logistics and Supply Chain Management

Permanent URI for this collection


Recent Submissions

Now showing 1 - 5 of 5
  • Item
  • Item
    Online Collective Demand Forecasting for Bike Sharing Services
    ( 2023-01-03) Dickens, Charles ; Miller, Alexander ; Getoor, Lise
    We introduce a general time-series forecasting method that extends classical seasonal autoregressive models to incorporate exogenous and relational information in an online setting. Our approach is implemented using the probabilistic programming language Probabilistic Soft Logic (PSL). We leverage recent work that enables the scalable application of PSL to online problems and propose novel modeling patterns to leverage dependencies between multiple time series. We demonstrate the applicability and performance of our method for the task of station-level demand forecasting on three bike sharing systems. We perform an analysis of the demand time series and present evidence of relational dependencies among the stations, motivating the need for a forecasting model that leverages the rich relational structure in the bike sharing networks. Our approach significantly improves multi-step forecasting accuracy of autoregressive time-series models on all three datasets. Further, our approach is easily extendable and we expect applicable to a variety of other time-series forecasting problems.
  • Item
    Supporting Your Basic Needs - A Base Support Approach for Static Stability Assessments in Air Cargo
    ( 2023-01-03) Longhitano, Diego Enrico ; Mazur, Philipp Gabriel ; Wolf, Simon Maximilian ; Schoder, Detlef
    Static stability is one of the most important constraints in the design and efficient calculation of safe air cargo pallets. To calculate the static stability of a cargo layout, base-focused methods such as full or partial base support are often used. Compared to mechanical or simulation-based methods, they offer high performance and simplicity. However, these methods currently reach their limits when dealing with the practical complexity of air cargo, making them difficult to apply in practice. In this research, we extend and generalize these support point methods by modeling irregular and multilevel cargo shapes, which enables improved practical applications. We follow a design-oriented approach to capture air cargo requirements, design an artifact, and evaluate its performance. Our results show a generalized approach that covers a greater practical complexity while maintaining its efficiency.
  • Item
    Towards Just-In-Time Arrival for Container Ships by the Integration of Prediction Models
    ( 2023-01-03) Yu, Jingjing ; Voß, Stefan
    Within the context of green shipping, the concept of Just-In-Time (JIT) arrival has attracted much attention. Research achieves the JIT arrival for container ships by combining the berth allocation and quay crane assignment problem (BACAP) and the vessel speed optimization (VSO), both subject to the data exchange. Many prediction models of the research to date generally aim to reduce the uncertainty of the communicated estimated time of arrivals. There is a lack of research that simultaneously assesses the application effect of prediction models on both plans of the BACAP and the VSO. Therefore, this paper proposes a two-stage model that integrates the prediction of the vessel arrival time with the optimization of the BACAP-VSO. The application in our specific case study shows that the random forest performs best in the first stage. The results are forwarded to the second stage and lead to a reduction of the service delay, fuel consumption cost, and vessel emissions.
  • Item
    Digital Twins for Internal Transport Systems: Use Cases, Functions, and System Architecture
    ( 2023-01-03) Marinkovic, Minja ; Galka, Stefan ; Meißner, Sebastian
    Internal transport systems are an essential part of intralogistics in production and distribution facilities. These are characterized by a variety of technologies as well as a multitude of interactions with other processes, such as warehouse, picking, and production processes. Therefore, resource planning and control of these systems is complex, especially for discontinuous conveyors. In this task, users can be supported by Digital Twins for decision-making, as they are suitable for investigating both future system states and possible actions. However, relevant use cases that are generally applicable across sectors as well as a generic system architecture for Digital Twins for resource planning and process control of in-plant transport systems have not yet been sufficiently investigated. In this paper, use cases are presented, relevant functions defined, and, finally, a generic functional and a logical reference architecture described. This is conducted with the design science in information systems research method together with a Systems Engineering approach. The use cases are determined at industrial partners of the research project TwInTraSys, which explores Digital Twins for the planning and control of internal transport systems. They are generalized and, thus, also applicable to other production and distribution facilities in different sectors. Further, the reference architecture can provide a basis for the successful implementation of the Digital Twin.