@phdthesis{Krudewig2015, author = {Krudewig, Christian}, title = {Transport and capacity planning for reusable containers under fixed-step transport costs}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:824-opus4-2889}, school = {Katholische Universit{\"a}t Eichst{\"a}tt-Ingolstadt}, pages = {viii, 138 Seiten : Diagramme}, year = {2015}, abstract = {Many companies use reusable containers to store and transport products. Under good conditions this can help to reduce costs and improve the ecological footprint of logistic operations. On the other hand, the acquisition, maintenance and distribution of the containers need to be organised. The aim of this thesis is to provide mathematical models for planning transports and capacity adaptations in a container system. First, the main characteristics of this type of system and the planning tasks are analysed. A key characteristic of container transports is identified: containers can often be moved in larger quantities on the same vehicle, so that transport costs should not be regarded as linear but increasing stepwise with the number of vehicles employed. Hence, the costs represent a fixed-step function of the quantity. It is shown that extensive literature is available concerning related topics. Mathematical models already exist, which allow for some of the facets of transport and capacity planning that are important for reusable containers. However, these approaches never cover all relevant aspects. Some are missing completely; among them the challenge to model fixed-step transportation costs. To get a flexible, yet mathematically tractable modelling, a toolbox with two basic models and multiple add-ons was developed. These can be combined to reflect a certain application. The first basic model considers linear transport costs. It can be reduced to the well investigated minimum cost flow problem which is known to be solvable in a very efficient manner. To consider fixed-step transport costs, another basic model was proposed. This is a more complex mixed integer program. It can be solved by a tailored branch-and-cut approach, which combines the positive characteristics of a heuristic and an exact solution algorithm: it provides a reasonable and feasible solution quickly and reaches the optimum solution in the end. The model add-ons include the aspects of lost orders, delayed deliveries and take-along transports as well as two capacity measures, namely purchase and disposal of containers as well as leasing. All of these extensions can be integrated into both basic models, still preserving their mathematical characteristics.}, subject = {Ganzzahlige lineare Optimierung}, language = {en} }