In this paper, the semi-empirical battery degradation prediction model proposed considers electrochemical degradation characteristics and represents degradation effects under various conditions, including different states of charge (SoC) areas. This model is specifically designed to address degradation during cycling and short-term rest periods in lithium-ion batteries using liquid electrolytes. Cycle aging incorporates the impact of solid electrolyte interphase (SEI) growth, a known dominant factor, and the model for short-term resting periods captures potential aging impacts on subsequent cycles due to internal material concentration gradients, moving away from the traditionally used calendar life approach. The derivation of the model presented in this paper is based on 14 data sets under different SoC conditions and 8 data sets under various Crates, explaining the degradation effects at 10 % SoC intervals and three different Crate points. Moreover, the model's performance was validated through capacity prediction for two data sets experimented with dynamic operational schedules of actual energy storage systems (ESS), including various conditions. The results showed a root mean square error (RMSE) of 0.564 and a mean absolute percentage error (MAPE) of 0.346. The superiority of this model is demonstrated by comparing its performance with four other types of degradation models derived through the same process in the validation data. • The model was designed to consider electrochemical aging characteristics for precise battery degradation prediction. • New insights on degradation during rest states move beyond the traditional calendar life approach. • The flexibility of the proposed model was evaluated using data from 22 battery lifespan experiments. • The performance was assessed using experimental data on two real-world ESS schedules, compared to other models.