Intelligent Decision Support for Logistics and Supply Chain Management

Permanent URI for this collection

https://hdl.handle.net/10125/72145

Browse

Recent Submissions

Now showing 1 - 5 of 5
  • Item
    Managing Uncertainty in Pharmaceutical Supply Chains: A Structured Review
    ( 2021-01-05) Blossey, Gregor ; Hahn, Gerd J. ; Koberstein, Achim
    In past decades, pharmaceutical supply chains (PSC) have become increasingly fragmented and complex making them more susceptible to supply chain (SC) risks. This has manifested in a growing number of drug shortages around the world which presents a great challenge to many national healthcare systems. PSC models provide crucial decision support and can facilitate efforts to avoid or manage such stockout situations more effectively. In this paper, we review the scientific literature on quantitative PSC models considering uncertainty in the context of drug shortages. We conduct a systematic search to obtain an overview of the current state of research in this field. The identified papers are analyzed with regards to pivotal modeling choices and their characterization of uncertainty. Our results show that many models make assumptions or abstractions which do not accurately reflect the current business environment of the pharmaceutical sector. Thus, we deduce future research avenues which might lead the way to more responsive and resilient PSCs.
  • Item
    Demand Forecasting Intermittent and Lumpy Time Series: Comparing Statistical, Machine Learning and Deep Learning Methods
    ( 2021-01-05) Kiefer, Daniel ; Grimm, Florian ; Bauer, Markus ; Van, Dinther, Clemens
    Forecasting intermittent and lumpy demand is challenging. Demand occurs only sporadically and, when it does, it can vary considerably. Forecast errors are costly, resulting in obsolescent stock or unmet demand. Methods from statistics, machine learning and deep learning have been used to predict such demand patterns. Traditional accuracy metrics are often employed to evaluate the forecasts, however these come with major drawbacks such as not taking horizontal and vertical shifts over the forecasting horizon into account, or indeed stock-keeping or opportunity costs. This results in a disadvantageous selection of methods in the context of intermittent and lumpy demand forecasts. In our study, we compare methods from statistics, machine learning and deep learning by applying a novel metric called Stock-keeping-oriented Prediction Error Costs (SPEC), which overcomes the drawbacks associated with traditional metrics. Taking the SPEC metric into account, the Croston algorithm achieves the best result, just ahead of a Long Short-Term Memory Neural Network.
  • Item
    Computable Approaches to Rational Choice and Decision-Making
    ( 2021-01-05) Seo, Hyunjin ; Thorson, Stuart
    The capacity of agents to act rationally, that is to make choices that positively reflect their interests, is a core assumption underlying democratic governance systems, microeconomics, decision science, market driven economies, and many agent based modeling efforts. In this paper we investigate axiomatic theories of rational choice from the perspective of computability. Using algorithmic complexity, we show highly general conditions under which no effective procedure can exist enabling these theories to identify sequences of choices as random. While axiomatic theories of rational choice yield powerful descriptions of choice behavior, this power comes at the expense of axioms which can be brittle with regard to computability limits.
  • Item
    An Intelligent Decision-Support System for Air Cargo Palletizing
    ( 2021-01-05) Lee, No-San ; Mazur, Philipp Gabriel ; Bittner, Moritz ; Schoder, Detlef
    Palletizing in the air cargo sector faces a large number of constraints (e.g., aviation safety regulations) and represents a highly complex problem. In air cargo operations, there is hardly any digital support to optimize the palletizing process. As a result, desired objectives (e.g., optimal utilization of the possible loading weight, maximum use of the available loading space, or both) are often only met by chance. The goal of this research is to report on the design and performance of an intelligent decision support system that we built for the air cargo context. This system supports the manual palletizing process by considering far more constraints as well as more complex item shapes and unit load devices than any other system we know. We explain the problem context, including the essential requirements; model the solution design; and develop the intelligent decision support system as an artifact, which we then evaluate.
  • Item