In the recent years, the high cost for purchasing the software, inability to access the source code and some other reasons, encouraged many experts to develop open source software. The main goal of this study was to develop a urban land use allocation on a open source software. This study with review the theoretical foundations, a variety of strengths and weaknesses open source software, compared four  GIS desktop open source software  GRASS, UDig, QGIS and MapWindow  in order to develop a decision support system for land allocation and  Finally the MapWindow  software was chosen and a system  that called OSLAS was developed for the purpose. OSLAS has four potential, land demand, land allocation and evaluate modules and the basic ability of a Geographic Information System. The developed system is implemented for the allocation of urban land use in district one of Tehran and predicted land use map for the year 2021 is created.