摘要:AbstractTo maintain a production or logistic system in a state as it could perform its output is the core maintenance goal. Maintenance represents a cost for industrial realities but it is also a strategic function necessary to be competitive in the market. The Age Replacement Policy is the most diffused maintenance strategy and it is characterized by periodic replacement of components to avoid failures. Data as MTTF are used to estimate replacement intervals and the relative costs of maintenance.Weibull distribution is the most diffused model to fit spare parts lifetime, and so to define the optimal periodic interval of replacement; in case of censored data, Weibull parameters could be wrong, and be influenced by a certain % of error. The aim of this study is to investigate the relationship between the % error in parameter estimation and the additional costs related to this error. Economical decision graph and maps have been considered to support the study. Finally, considerations about consequences on costs in case of overestimation or underestimation are carried out.