Beschreibung
The NP-hard Resource-Constrained Project Scheduling Problem (RCPSP) is a well-known optimization problem. One new extension is called RCPSP with alternative activity chains (RCPSP-AC) presented by Tao and Dong (2017), which they describe as a generalized form of the Multi-Mode RCPSP. It consists of two sub problems, the resource constrained project scheduling and the activity selection problem, which represents the real-world possibility of alternative production ways where one way has to be selected. In this work, the RCPSP-AC is extended by multiple components, motivated by an application in the steel industry. A new activity type called "completion activity" is introduced. Thus, the production completion of multiple lots within the total production time horizon is ensured, which is necessary to satisfy all required precedence relations. Moreover, real-world inherent time restrictions are proposed. A new minimum necessary duration per activity replaces the fixed processing time per activity. Hence, it is possible to decide on the variable processing time per activity and therefore, also on the starting time of every activity. Moreover, new objective functions for this extension are modeled. The throughput time and the tardiness of single production lots or the whole project are minimized. The developed MIP model is implemented in CPLEX and solved to optimality for small instances. For larger, real-world instances, an evolutionary solution approach is proposed.| Zeitraum | 11 Juli 2018 |
|---|---|
| Ereignistitel | 29th European Conference On Operational Research |
| Veranstaltungstyp | Konferenz |
| Ort | Valencia, SpanienAuf Karte anzeigen |