Intelligent Search Heuristics for Cost Based Scheduling
This work addresses nurse scheduling, a domain-specific optimization challenge, with incremental improvements in solution time.
The paper tackles the nurse scheduling optimization problem by proposing an intelligent search heuristic that reduces the time needed to find feasible solutions, demonstrating superior performance compared to existing algorithms from Li et al. (2003) and Ozkarahan (1989).
Nurse scheduling is a difficult optimization problem with multiple constraints. There is extensive research in the literature solving the problem using meta-heuristics approaches. In this paper, we will investigate an intelligent search heuristics that handles cost based scheduling problem. The heuristics demonstrated superior performances compared to the original algorithms used to solve the problems described in Li et. Al. (2003) and Ozkarahan (1989) in terms of time needed to establish a feasible solution. Both problems can be formulated as a cost problem. The search heuristic consists of several phrases of search and input based on the cost of each assignment and how the assignment will interact with the cost of the resources.