In recent years, developments of the behind the meter energy managements systems (BTM-EMSs) has been considered as an effective approach to manage the energy usage of the industrial/commercial units. As the one of the most important missions, BTM-EMSs are responsible to reduce the customers' monthly demand peaks which is rewarded by significant decrease in the monthly demand charge. However, the unpredicted behavior of individual electricity loads casts a shadow over the profitability of installing BTM storage units. In this paper, a data driven demand charge management solution (DCMS) is proposed to find and realize the minimum achievable monthly demand peak, also called demand charge threshold (DCT), by appropriate battery storage charging and discharging. The proposed approach uses the last few months load profile to calculate time series of DCTs and then searches for the similar DCT time series of other observed loads stored in the data set. Finally, the obtained similar time series are employed to forecast the DCT of the coming month. The efficiency of the proposed approach is validated through the simulation studies on the real value load data.