Zoetermeer is facing two challenges currently. The first is a housing program which is defined only at a superficial level. Secondly, the municipality has expressed the ambition to also take into account the migration of households, but has no data on which dwellings facilitate migrations the best. This research thus set out to first define the migrations chains which are found in Zoetermeer and then subsequently use this information in designing a suiting housing program.

The outcome of the Markov chain theory, which was used to determine residential migration chain length, showed that a number of dwellings facilitated migration the most. These were the more expensive dwellings in general and middle segment dwellings for households residing for social rental apartments specifically. Second, to construct the housing program, detailing on the living environments and dwelling models was defined and combined with ambition of the city.

Around this information an LP decision-making model was built which could use these constraints and find the optimal solution, if one existed. Ten different housing programs were designed, each with a different aim, and the results were assessed. Out of the ten programs, one seemed the most appropriate overall, as it achieved reasonable values in a number of important variables, including number of migrations, spread and value of newly added dwellings and area requirement. A surprising find was however that one of the programs, which in advance was seen as an extreme, proved to fit the ambitions of Zoetermeer in terms of increasing the city’s monetary value and attracting a new type of resident with new dwelling types.

Lastly, a program was designed with the aim of combining all benefits of the other programs into one. This program showed the results of the most overall appropriate program as well as adding a lot of value and the new types of dwellings. To conclude, this research found that the link between the two challenges could be made relatively easily and that the results of the model were very well received after even the first iteration. It should however be mentioned that the input of the model turned out to be very influential and that the municipality had difficulty in defining this input throughout this research as a shift in focus had recently occurred, which resulted in a change of thinking from quantitatively to qualitatively defining goals.

Finally, the two models designed proven to work for Zoetermeer, but were also designed to be used by any other party desiring to either determine migrations patterns, design a housing program or doing a combination of the two.