Novel approaches to cyclic job-shop problems with transportation

Scheduling problems can be found in almost any field of application in the real world. These problems may not only have different characteristics but they also imply more or less complex requirements. One specific class within this domain is the cyclic job-shop problem. It occurs in various areas reaching from industrial production planning down to the systems architecture of computers. With manufacturers in particular, one can find increasing demand for effective solution methods in order to tackle these scheduling problems efficiently. This thesis will deal with the Cyclic Job-Shop Problem with Blocking and Transportation. It arises in modern manufacturing companies, where the products move automatically between the different workstations, for instance. The problem itself is not new to the research community, but hardly any work has been done in solving it. Within this thesis we will try to close this gap and present some first approaches, discussing the structure of the problem and how it can be solved. As a result, we will provide three different solution methods, including an integer programming formulation, which is solved with a commercial solver, a branch and bound algorithm and a tabu search heuristic. All algorithms are tested on a range of data sets and compared with each other. Additionally, we have worked on a polynomial solvable subproblem, which has gained more interest in the literature. As a result, a new polynomial algorithm, that outperforms the existing ones in theory as well as in empirical tests (except for some special cases) is presented. This thesis concludes with a discussion about ideas of how to improve the presented methods and some other extensions to the investigated problem.