The importance of cycling as a sustainable transport mode has been widely recognised, and recently, its effectiveness in mitigating the spread of infectious diseases has also been under the spotlight. Fostering its use requires developing and deploying decision tools to help authorities assess the performance of their cycle infrastructure for maintenance and improvements. This paper presents a multicriteria methodology based on engineering best practices and uses the Elimination and Choice Translating Reality (Electre) Tri method to assign segments of the cycling network to predefined performance classes, aiming at maintenance planning. The approach is demonstrated with a case study, which also proves the scalability of the data-collection procedure of the method. The case study results show that lack of safety and inadequate intersections are the main problems. These stem mostly from non-existent segregation between motorised traffic and cyclists, both along the segments and at intersections. This is typical of cities that, over the years, have prioritised motorised transportation.