An Adaptive Problen]-Sohing S@ lkm to I. arge-Scale Scheduling Problems

1994

Abstract Although the general class of most scheduling problems is NP-hard in worst-case complexity, in practice, domain-specific techniques frequently solve problems in much better than exponential time. Unfortunately, constructing special-purpose systems is a knowledge���intensive and time-consuming process that requires a deep understanding of the domain and problem-solving architecture.

NASA has a need for planning and scheduling systems that are expressive enough to encode real-world problems, flexible enough to keep up with shifting problem requirements, and powerful enough to provide high quality solutions within reasonable time bounds. These competing demands are particularly difficult to meet for the many NASA planning and scheduling problems that contain interacting combinatorial optimization sub-problems. General purpose planners are expressive and flexible, but perform poorly on these complex problems. Combinatorial optimization solvers have excellent perfonnance, but are only applicable to sub-problems. This paper describes a simple method for integrating certain kinds of combinatorial sub-problem solvers within a general purpose plannerhcheduler framework that demonstrably increases solution speed and quality.

Computer Aided Chemical Engineering, 2009

2007

Algorithms

Algorithms, 2023

This i the Editorial for a Special Issue 'Scheduling: Algorithms and Applications'.

TR 10-02, FMA, OvGU Magdeburg, 2010, 107 pages

Journal of the Operational Research Society, 2009

2024

"Sicilia Antiqua" XXI (2024) - Studi in memoria di Gioacchino Francesco La Torre

Post-Ottoman Coexistence: Sharing Space in the Shadow of Conflict, 2016