We consider an academic timetabling and rostering problem involving periodic retraining of large numbers of employees at an Australian electricity distributor. This problem is different from traditional high-school and university timetabling problems studied in the literature in several aspects. We propose a three-stage heuristic consisting of timetable generation, timetable improvement, and tr...

Duty rostering problems occur in different application contexts and come in different flavors. They give rise to very large scale integer programs which typically have lots of solutions and extremely fractional LP relaxations. In such a situation, heuristics can be a viable algorithmic choice. We propose an improvement method of the Lin-Kernighan type for the solution of duty rostering problems...

A novel nurse rostering model is developed to represent real world problem instances more accurately. The proposed model is generic in a way that different problem instances can be successfully represented. Novel local search neighbourhoods are developed to take advantage of the problem properties represented by the model. These neighbourhoods are utilised within a variable neighbourhood search...

The goal of hyper-heuristics is to design and choose heuristics to solve complex problems. The primary motivation behind the hyper-heuristics is to generalize the solving ability of the heuristics. In this paper, the authors propose a Hyper-heuristic using GRASP with Path-Relinking (HyGrasPr). HyGrasPr generates heuristic sequences to produce solutions within an iterative procedure. The procedu...

The production of effective workforce rosters is a common management problem. Rostering problems are highly constrained and require extensive experience to solve manually. The decisions made by expert rosterers are often subjective and are difficult to represent systematically. This paper presents a formal description of a new technique for capturing rostering experience using case-based reason...

Assigning staff to specific duties according to their contract, qualifications, skills, etc. within a working environment characterised by multi-disciplinarity and statutory regulations is problematic. This paper discusses an approach to nurse rostering, using a strategy of distributing the computational effort required in the scheduling process. The technique involves a hybrid approach that de...

Matrix models are ubiquitous for constraint problems. Many such problems have a matrix of variablesM, with the same constraint defined by a finitestate automaton A on each row of M and a global cardinality constraint gcc on each column of M. We give two methods for deriving, by double counting, necessary conditions on the cardinality variables of the gcc constraints from the automaton A. The fi...

Many rostering methods first create shifts from some given staffing levels, and after that create rosters from the set of created shifts. Although such a method has some nice properties, it also has some bad ones. In this paper we outline a method that creates rosters directly from staffing levels. We use a Branch-and-Price (B&P) method to solve this rostering problem and compare it to an ILP f...

We investigate the applicability of an existing framework for algorithm runtime prediction to the field of metaheuristics, in particular applied to the real world problem of nurse rostering. Apart from predicting the runtime, we look at other performance criteria as well. These so called empirical hardness models are based on readily computable features of the problem instances. These problem f...