Home care service planning. The case of Landelijke Thuiszorg

• Published in: European journal of operational research 2015-05, Vol.243 (1), p.292-301
• Main Authors: Maya Duque, P.A., Castro, M., Sörensen, K., Goos, P.
• Format: Article
• Language: English
• Subjects: Bi-objective optimisation; Computation; Computer programs; Decision support systems; Home care; Home health care; Local search; Mathematical analysis; Mathematical models; Mathematical programming; Operations research; Optimization; Pilots; Routing; Staff planning; Strategy; Studies
• ISSN: 0377-2217; 1872-6860
• DOI: 10.1016/j.ejor.2014.11.008

•We studied a home care planning problem faced by Landelijke Thuiszorg, a non-profit organisation that provides home care services in several provinces of Belgium.•A set partitioning formulation of the bi-objective problem is proposed.•We propose a two-phase approach that allows to analyse the trade-off between the prioritised objectives.•Our solution strategy is flexible in the sense that it can be easily adapted to tackle additional and complicating constraints.•The solution strategy is simple and easy to understand which facilitates the communication and the acceptance of our algorithm within the organisation.•Our two-phase algorithm will be the core optimisation component of a decision support system. Landelijke Thuiszorg is a “social profit” organisation that provides home care services in several Belgian regions. In this paper, the core optimisation component of a decision support system to support the planning of the organisations’ home care service is described. Underlying this decision support system is an optimisation problem that aims to maximise the service level and to minimise the distance travelled by the caregivers of the organisation. This problem is formulated as a bi-objective mathematical program, based on a set partitioning problem formulation. A flexible two-stage solution strategy is designed to efficiently tackle the problem. Computational tests, as well as extensive pilot runs performed by the organisation’s personnel, show that this approach achieves excellent performance, both in terms of the service level and total travelled distance. Moreover, computational times are small, allowing for the weekly planning to be largely automated. The organisation is currently in the process of implementing our solution approach in collaboration with an external software company.