Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
A simulation study of a synchromodal logistic network
Vrije Universiteit Brussel (VUB), Belgium
Synchromodal transport employs multiple transport modes in a flexible and dynamic way in order to induce a modal shift towards more environmentally friendly transport modes. This leads to more efficient and sustainable logistic systems, but it also imposes a number of challenges that must be addressed before implementing the concept in practice. As part of the Dispatch project , in this study an agent-based simulation model that represents a large-scale logistic network in a GIS environment is developed, including transport by truck, train and inland waterways. The aforementioned model can be used to test different scenarios in terms of disruption occurrences, as well as different levels of cooperation and transparency between the different actors of the logistic system. Moreover, the model includes optimization algorithms related to routing, mode choice and capacity utilization problems. The model estimates several metrics related to the logistic operations, including orders lead times, costs for different actors, capacity utilization indicators and environmental impacts. Therefore, the model provides a testbed for synchromodal opportunities within a risk-free environment. This allows the evaluation and design of new strategies and initiatives oriented to improve the operations of the logistic networks. In the next stages of the project, the model will be modified so that it can be fed with real time data, in order to develop a digital twin to further enhance the synchromodal concept.
Machine-Learning-based ETA predictions of logistics networks
Weinke, Manuel; Poschmann, Peter; Straube, Frank
Technische Universität Berlin, Germany
Logistics networks are faced with increasing demands for higher flexibility and sustainability with a simultaneous raising of cost pressure. Against this background, it is important for the companies to ensure an optimal utilization of their assets, to reduce the vulnerability of their processes towards disruptions and to foster the attractiveness of eco-friendly transportation modes. Innovative data technologies such as machine learning (ML) as part of AI offer great potential for overcoming these challenges through the ability to predict process and arrival times (ETA) as well as upcoming disruptions very precisely. The presentation shows the procedure and the results of investigating the feasibility of the ML technology for this purpose in different logistics network. Starting with a developed and publicly available self-learning system for intermodal land transport, which includes data for 4 years out of 15 IT systems of several logistics services provides, the presenters will provide an insight into their current research activities on the implementation of a ML-based ETA prediction for inland waterway logistics chains. This comprises an overview about processes to be considered, relevant features and corresponding data availability as well as the current status of the model development with its predictions results. The presentation will conclude with derived potentials and barriers of the used ML-based approaches which might be considered in corresponding projects.
Yard Management: Identification and Evaluation of Critical Sub-processes with AHP
1FOM University of Applied Sciences Essen, Germany; 2Fraunhofer IML Dortmund, Germany; 3Georg-August-University of Göttingen, Germany
Yard management is crucial for logistics and transport operations due to the high influence towards smooth and efficient intralogistics on dedicated depot sites of logistics service providers. Yet, this field has interesting new insights to offer, especially regarding prioritizing and decision-making concepts with the support of Multi-Criteria Decision Analysis (MCDA) tools. Within this publication five critical yard sub-processes are identified and prioritized in the following order: management of the shunting system, registration at the gateway, allocation of trucks to gates/parking spaces, removal of a transport unit from the gate and exit control.