Decision-making for the refurbishment of multi-family residential buildings is a complex and computationally difficult task. Therefore, the authors have developed a model that supports modernization planning in a long-term and comprehensive manner, i.e. from assessing the building to indicating the optimal scope of modernization. The comprehensive scope of the model includes the acquisition and provision of relevant knowledge to the model. The original methods proposed for its acquisition are derived from common expert knowledge based on linguistic terms. The methods adopted are not mandatory and may be replaced by others that provide more reliable knowledge. The fundamental aim of the proposed approach, however, is to select the optimal modernization option and allocate it over the planned modernization time horizon. An innovative optimisation approach based on decision matrices is used, allowing the selection of possible scenarios of repair options at each stage. These matrices are a set of constraints written in binary variables allowing the optimisation calculus to maintain a fixed sequence of repairs. In addition, the solutions used in the optimisation modules make it possible to take into account assumptions regarding the assumed assessment of the building's condition and financial constraints. The developed model provides a practical and versatile tool that can be used by managers at the maintenance stage of residential buildings.