Maintenance planning is one of the main challenges of the management of production systems. Thanks to recent technological innovations and the development of sensors, it is possible to construct degradation models of equipment condition, thus providing an assessment of the equipment health modeled by a health index indicator. In this work, we present an original tactical maintenance planning problem involving a Health Index indicator on the components of the equipment. Within this framework, maintenance can be seen as a form of delayed production. This observation allows us to derive an original lot-sizing based approach for our tactical maintenance planning problem. Aside from the classical cost minimization objective, the problem also involves a resource consumption objective to be minimized. The problem is proven to be strongly NP-hard in the general case, and a bi-objective Mixed Integer Linear Program is proposed. Then, two matheuristics are defined. A degradation matheuristic allowing several degradations of the objective functions and a Pareto-based matheuristic selecting solution of good quality from a restricted pareto front.