This paper presents a novel methodology for the on-board estimation of the actual battery capacity of lithium iron phosphate batteries. The approach is based on the detection of the actual degradation mechanisms by collecting plateau information. The tracked degradation modes are employed to change the characteristics of the fresh electrode voltage curves (mutual position and dimension), to reconstruct the full voltage curve and therefore to obtain the total capacity. The work presents a model which describes the relation between the single degradation modes and the electrode voltage curves characteristics. The model is then implemented in a novel battery management system structure for aging tracking and on-board capacity estimation. The working principle of the new algorithm is validated with data obtained from lithium iron phosphate cells aged in different operating conditions. The results show that both during charge and discharge the algorithm is able to correctly track the actual battery capacity with an error of approx. 1%. The use of the obtained results for the recalibration of a hysteresis model present in the battery management system is eventually presented, demonstrating the benefit of the tracked aging information for additional scopes. (C) 2016 Elsevier B.V. All rights reserved.