@MASTERSTHESIS\{IMM2006-04710,
    author       = "K. Thomsen",
    title        = "Optimization on Home Care",
    year         = "2006",
    school       = "Informatics and Mathematical Modelling, Technical University of Denmark, {DTU}",
    address      = "Richard Petersens Plads, Building 321, {DK-}2800 Kgs. Lyngby",
    type         = "",
    note         = "Supervised Jesper Larsen, {IMM,} {DTU,} and co-supervisor Rene Munk J{\o}rgensen, {CTT,} {DTU}.",
    url          = "http://www2.compute.dtu.dk/pubdb/pubs/4710-full.html",
    abstract     = "The purpose of this project is to investigate how methods from operational research can be applied in the home care sector. A problem very similar to the {VRPTW} arises, when scheduling the routes for the caretakers. The problem is very complex, and hence it is simplified in this project. The conditions are limited to include the time windows of the visits, the working hours of the caretakers, visits locked to caretakers and if two caretakers share a visit. In a shared visit two caretakers have to start and finish the visit at the same time. The aim is to minimize the travelling time and maximize the number of visits, which are attended by a regular caretaker. 

An intelligent insertion heuristic is applied on the problem. The insertion heuristic uses the regret measure to evaluate where the best insertion position is. The solutions found via the insertion heuristic are used as initial solutions for a tabu search, which allow infeasible solutions.

The results show, that when maximizing the number of visits with a regular caretaker, the total travelling time is likely to increase. The initial solutions found by the insertion heuristic are improved by the tabu search up to 27 \%.

The solutions are compared with solutions found by a programme called {ABP}. The {ABP} incorporates more wishes and conditions. The results show, that the solutions found by the methods investigated in this project are better in all cases."
}