Neighbourhood functions are key components of local-search algorithms such as Simulated Annealing or Threshold Accepting. These functions take a solution and return a slightly-modified copy of it, i.e. a neighbour. The package provides a single function, neighbourfun(), that produce a neighbourhood function, based on various parameters such as minimum or maximum values. Currently supported are numeric and logical solutions. The algorithms were originally created for portfolio-optimization models, but can also be used for other models.


Installing the package

The latest released version is available from In an R session, just type:

                 repos = c('', getOption('repos')))