A constructive heuristic for staff scheduling in the glass industry

• Type: Article in International Scientific Journal
• Date: 2014

Authors

Marta Rocha (Author) Other; Jose Fernando Oliveira (Author) FEUP; Maria Antonia Carravilla (Author) FEUP

Journal

Title: Annals of Operations Research

Vol. 217

Pages: 463-478

ISSN: 0254-5330

Publisher: Springer Nature

Indexing

ISI Web of Knowledge - 4 Citations

ISI Web of Science

Scopus - 6 Citations

INSPEC

Scientific classification

CORDIS: Technological sciences > Engineering > Industrial engineering

FOS: Engineering and technology > Industrial biotechnology

Other information

Authenticus ID: P-009-HKS

DOI: 10.1007/s10479-013-1525-y

Abstract (EN): In this paper a constructive heuristic for solving the staff scheduling problem of a glass manufacture unit is proposed. Based on simple calculations and algorithms, the developed procedure assigns working shifts and days-off to teams of employees, ensuring the satisfaction of a mandatory sequence of working shifts and the balance of the workload between employees. The computational times for the experiments with the case study company, with three eight-hour working shifts and five teams of employees, fell consistently below 5 seconds for a set of different planning periods. Results are compared with the ones achieved with an optimization model (MIP), demonstrating the good performance of the heuristic, also in terms of the quality of the achieved solutions. The heuristic rarely fails to produce a feasible solution and whenever the solution is feasible then it is also optimal. When tackling problems with a large number of teams, the heuristic maintains the good performance while the MIP model is not able to find any solution within 16 hours of running time. Although it was designed for a particular problem of the glass industry, tests show that the heuristic is flexible enough to be applied to problems with different features, from other activity sectors, encouraging further extensions of this work.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 16